Econstudentlog

Autism Spectrum Disorder (I)

Assburger(smbc)

I recently realized that I had actually never read a textbook like this on this topic. I did get some reading materials back when I got diagnosed so it’s not like I’ve never read anything about the stuff (and there was a lot of verbal information back then as well), but as mentioned I haven’t read a text on the topic. It was actually due to the old reading materials in question that I ended up deciding to read this book; I was looking for some other stuff the other day and I ended up perusing some of these materials (which I hadn’t seen in years), and I figured I should probably go read a book on the topic. Now I am.

The book is sort of okay. There are various complaints one might make, the most important one of which in the context of me reading the book is perhaps that children with autism-spectrum disorders grow up and become adults, and adults prefer to read chapters about adult stuff, not stuff about e.g. how to teach the preschooler with the diagnosis social skills. I’ve read roughly half the book at this point, and there’s not in my opinion been enough stuff about the adult setting at this point. Another complaint is that I as usual am somewhat mistrustful when guys like these talk about the conclusions to be drawn from some types of empirical evidence; the coverage has in my opinion been of a decidedly mixed quality in terms of the stuff dealing with behavioural interventions, in the sense that they on the one hand at one point reasonably frankly acknowledge that the evidence is sparse and of poor quality, and on the other hand later on seem to become very excited about a longitudinal study and start drawing big conclusions from that single study – which would be sort of fine, I like longitudinal studies, if not for the fact that the study was based on 6 (!) individuals. Similar things happen elsewhere in that part of the coverage – potential power issues are never mentioned in the book, at least they have not been so far – you find yourself reading about a ‘seminal’ study on 19 individuals, and then you move on to their comments about how there have been several other studies supporting those findings, including a study looking closer at 9 of the individuals involved in the original study. Sometimes it’s hard to know what to think, especially in the situations where the only people evaluating the interventions are the people who came up with them in the first place – this doesn’t seem like a particularly smart way to conduct business, though in some parts of psychology it seems to be more or less standard practice.

The stuff on behavioural interventions has in my opinion been some of the weakest stuff in the book so far, which is why I have not talked about this stuff in my coverage below. Some of the proposed interventions are incredibly expensive, and there’s probably a good reason why such things are usually not covered by public health care systems, however the authors do not really seem to consider economic aspects to be all that important, except to the extent that economic factors unfortunately restrict access to all these nice things we could do for these children; they’re aware that parents may not be able to afford the treatment options which are recommended at this point by people who would benefit from these treatment options being more widely used, but they don’t seem to be aware of the existence of things like cost-effectiveness analyses. It’s one thing to argue that there may be developmental gains to be achieved by early childhood interventions (I’ve previously done work in educational economics and I can tell you that it is a common finding in this literature that you can improve outcomes by throwing lots of money and attention after young children – a finding which should perhaps not be super surprising..), it’s quite another thing to argue that the specific interventions comtemplated are cost-effective. To be fair, cost-effectiveness is incredibly hard to evaluate when you’re contemplating evaluating interventions which may have effects lasting basically the rest of the life of the individual and the intervention is supposed to take place during the first years of a child’s life, but in my opinion you sort of need to at least pretend to try to address this aspect somehow; if you don’t, you’re quite likely to end up in a situation where it seems as if you’re acting as if there’s no (societal) budget constraint, and the authors of this book seem to me to move very close to this position at various points in the coverage.

I knew very little (nothing?) about autism-spectrum disorders before I got the diagnosis – I got diagnosed very late, in my adulthood. It’s sort of funny how you can miss important stuff like this without even knowing, and in a way it relates to a point which came up in my recent post on ethics, specifically the point that ‘bad’ people tend to think they are ‘good’ people, or at least no worse than average. How much do you really know about how good other people are at, say, interpreting nonverbal social signals? Would withdrawal from social interaction make the comparison easier or harder? If you don’t really engage in the normal patterns of non-verbal information exchanges, e.g. eye contact exchanges, during social situations, how are you to know that important information is contained in such exchanges? Individuals seem to make assumptions about these things to a large extent based on what they know themselves (about themselves?), and if you have limitations in these areas it may be difficult to figure out that this is the case; another apt analogy might be children who need glasses early on in their lives – we screen for vision impairment in young children in part because young children don’t know, and may never on their own ever realize, that the world is not supposed to be blurry, and that you’re actually supposed to be able to see all the letters written down on the blackboard.

I thought I should make one thing clear before moving on to the main text, a point particularly relevant considering the comic which I decided to start out with; which is that incompetence should not be equated with/interpreted as malicious intent. It seems to me that many people conceive of people with autism-spectrum disorders as inconsiderate jerks who don’t have a clue – I’ve seen quite smart people state relatively similar things in the past. I dislike the ‘jerk’-model because I try to be thoughtful and considerate when interacting with others, and when these people think that way I feel that they’re devaluing the work I put into this stuff. One important problem which is sort of hard to figure out how to deal with is that I’m well aware that the more thoughtful and considerate I am (…or is it: ‘try to be?’) during social encounters, the more taxing the social interactions may become, and taxing social interactions lead to social isolation and withdrawal. Coming up with a good equilibrium level of effort is not an easy task, and I think one needs to address aspects like these before making strong judgments about things like the jerkishness of specific behaviours. In a way people with social anxiety have similar concerns which other people also cannot observe (in this case it would be excessive amounts of thinking during social situations about whether they are doing stuff right now that may mean that they’ll get rejected by others, which then leads to oversensitivity to clues of rejection, leading to social avoidance because of perceived rejection). Of course people with autism-spectrum disorders may be anxious as well, as also mentioned in the coverage below. The level of self-awareness varies a lot in people with autism-spectrum disorders, but people with relatively high levels of self-awareness may certainly face some constraints and tradeoffs which are not immediately obvious to the outsider and which may actually be assumed by neurotypicals to be absent, given the diagnosis.

The textbook answered one question I’d been thinking about a few times without ever worrying enough about it to actually seek out an answer, which is the question of what the recent diagnostic changes might mean, given that I have a diagnosis which by now has been ‘retired’. It turns out that I was diagnosed with what in the textbook are considered to be the ‘gold-standard tools’, which means that this remark related to the recent diagnostic changes that have taken place seems to answer the question: “The DSM 5 noted that “Individuals with a well-established DSM-IV diagnosis of” “Asperger’s disorder” “should be given the diagnosis of autism spectrum disorder””. I’m not going to ‘ask’ for a ‘new’ diagnosis (/a ‘translation’ of my diagnosis) (and quite aside from what other people like to call this stuff, I like the word ‘eccentric’ a lot better than the word ‘autistic’…), but it’s nice to know which recommendations are being made in this area. Some of the quotes below also relate a bit to these aspects.

I’ve added some quotes from the book below.

“Autism is a developmental neurobiological disorder characterized by severe and pervasive impairments in reciprocal social interaction skills and communication skills (verbal and nonverbal), and by restricted, repetitive, and stereotyped behavior, interests, and activities. […] Autism and autistic stem from the Greek word autos, meaning “self.” The term autism originally referred to a basic disturbance, an extreme withdrawal of oneself from social life, or aloneness. […] The critical point in the scientific history of autism was in 1943, when Leo Kanner published Autistic Disturbances of Affective Conduct, a groundbreaking paper that described the symptoms of 11 children presenting similar behaviors that had not been previously recognized. […] Based on Kanner’s terminology, autism was considered for years a psychosis, and child psychiatrists were using “childhood schizophrenia” and “child psychosis” in autism as “interchangeable diagnoses.” […] A parallel line of inquiry to that of Kanner and Eisenberg is represented by the work of Hans Asperger.”

“In Autism and Pervasive Developmental Disorders, Fred Volkmar and Catherine Lord (2004) distinguished important points of differentiation and similarities between Kanner’s and Asperger’s descriptions. […] In concluding their comparison of Kanner’s and Asperger’s descriptions, Volkmar and Lord pondered whether, despite the relevant differences, it was “scientifically and clinically helpful to classify individuals with these traits into separate categories of autism or Asperger’s disorder, or whether it would be better to treat them as parts of a greater continuum.” The utility of the “greater continuum” has led to the category of autism spectrum disorder to be proposed for DSM-5. […] As a result of [various findings] and the lack of reliability in the community in making distinctions among the ASDs [Autism-Spectrum Disorders] [for example: “Variations in clinical severity among ASD cases are not valid indices of differences in pathophysiology or etiology”], the Fifth Edition of the Diagnostic and Statistical Manual (DSM-5) proposes to collapse all of these clinical syndromes into a single diagnosis of “autism spectrum disorder.” Although this revision is appropriate for community diagnosis, and thus the allocation of clinical and support services, research studies will continue to rely on research diagnostic instruments like the Autism Diagnostic Interview (ADI) and the Autism Diagnostic Observation Schedule (ADOS) [these were both part of my work-up, US] to make categorical distinctions between “autism and not autism” and “autism and autism spectrum disorder” (which includes Asperger’s disorder and PDDNOS [Pervasive Developmental Disorder Not Otherwise Specified]). These distinctions have played a vital role in advancing our understanding of the behavioral and neural profile of ASD over the past two decades”

“Recent studies and reports from the Centers for Disease Control […] have shown an increase in the prevalence of children diagnosed with an ASD to one in 110 […] The reported increase is thought to be attributable to several factors. First, there have been changes in diagnostic practices […] Second, there is greater public awareness of ASD and more case-finding […] Finally, there has been a tendency to diagnose many children with intellectual disability as PDD. […] no evidence currently exists to support any association between ASD and a specific environmental exposure. […] Numerous studies have failed to demonstrate a causal relationship between immunizations, particularly thimerosal-containing vaccines, and ASD […] The CDC (2009) reports the median age for a diagnosis of ASD to be between 4.5 and 5.5 years. […] the ASD diagnosis is four times more common for boys than in girls.”

“The essential features of Asperger’s disorder are severe and sustained impairment in social interaction (criterion A); and the development of restricted, repetitive patterns of behavior, interests, and activities (criterion B); which must cause clinically significant impairment in functioning (criterion C). There are no clinically significant delays in language (criterion D) or cognitive development (criterion E).”

“ASD (excluding Asperger’s disorder) has early language and communication impairment. […] almost two thirds of individuals with ASD also have ID [intellectual disability] […] 15%–20 % of cases of ASD are now linked to genetic or
chromosomal abnormalities […] Fragile X Syndrome (FXS) [is] the most common identifiable cause of ASD and the most common inheritable cause of ID. […] Thirty percent of individuals with FXS demonstrate characteristics of ASD.” [In some other conditions penetrance is even higher – examples that could be mentioned are 15q duplication and Timothy Syndrome, but prevalence is lower in these cases and especially in the latter case some might argue that the autism is the least of that child’s problems..]

Challenging behaviors [in individuals with ASD] may reflect pain that is not communicated verbally […] Challenging behaviors may [also] reflect the child’s difficulty with communication, changes, new places, new situations, new experiences, new sounds, new smells, and new people” [I wonder if you can spot a pattern here in terms of what these children (/people) don’t like? I think an important distinction here is to be made between curiosity and the desire to try out new things. I’m often, hesitant, about trying out new things, yet I’m also quite curious about a lot of things. Be careful which categories you apply here and how they may impact your thinking… In a related vein:] “Insistence on sameness and difficulty with change are common symptoms of an ASD. These behaviors should not typically be considered a behavior done to exert control over others.”

“Psychiatric comorbidity is now acknowledged as quite common in ASD [and] psychiatric comorbidity increases the level of impairment […] There is a handful of questionnaires [aiming at spotting psychiatric comorbidities] that have been developed specifically for use in developmentally disordered or ASD populations. […] none of the measures has the level of research support possessed by questionnaires used in other branches of psychiatry. The vast majority of these instruments have just one study behind their development, or have been studied only by the developer of the instrument. […] one of the main challenges in diagnosing psychiatric disorders in individuals with ASD is the possibility of different presenting symptoms and difficulty in differentiating impairment related to the underlying ASD from impairment due to a separate condition. […] While we do not want to miss true comorbid diagnoses, over-diagnosing comorbidity can be equally harmful. […] Mood disorders, such as depression and bipolar disorder, in ASD have recently begun to receive a great deal of attention […] there are many potential psychosocial stressors that could be possible triggers. For example, higher-functioning individuals who are aware of their deficits and badly desire friends, but lack success in this area, are at particular risk. […] Although there is little research on emotion regulation in ASD, there is clear evidence that emotion regulation is highly variable and often problematic in this population, regardless of psychiatric comorbidity […] Therefore, particularly for mood disorders, it is imperative to consider baseline functioning and not over-diagnose mood disorders when the concern may be more temperamental in nature.”

“Anxiety is considered by some to be the most common comorbid psychiatric concern in ASD […]. The DSM-IV-TR notes that individuals with ASD might have unusual fear reactions, and it is also not uncommon for there to be a general tendency toward anxiety for many individuals with ASD. […] There are many aspects of having an ASD that may lead to this increased risk for anxiety, to the degree that some consider anxiety and the social impairment in ASD to have a bidirectional relationship […] An increase in self-awareness is considered a risk factor for higher anxiety; therefore, anxiety is typically thought of as more common among individuals with ASD who have higher intellectual abilities, and older children, adolescents, and adults.”

“autism may be conceptualized as a disorder of complex information processing resulting from disordered development of the connectivity of cortical systems (e.g., failure of cortical systems specialization) […] approximately 15%–20% of infants with an older sibling diagnosed with autism will ultimately be diagnosable with ASD by three to four years of age. […] [Findings from longitudinal sibling studies] do not support the view that autism is primarily a social-communicative disorder and instead suggest that autism disrupts multiple aspects of development rather simultaneously. […] When both elementary and higher-order abilities in many domains are assessed, it becomes evident that deficits exist in several domains not considered to be integral parts of the autism syndrome, including aspects of the sensory-perceptual, motor, and memory domains. Furthermore, there are enhanced skills and impaired abilities within the same domains as deficits (e.g., memory, language, abstraction). […] Causal explanations for ASD must account for the comprehensive pattern of both deficits and intact aspects of the disorder both within and across multiple domains. […] There is no single primary deficit or triad of deficits, brain regions, or neural systems causing autism. […] Rather, autism broadly affects many abilities at the same time and systematically from its earliest presentation and throughout life. […] This pattern [can] be characterized overall as reflecting a disorder of complex or integrative information processing, which results from altered development of cerebral cortical connectivity in ASD. […] Just as the infant sibling studies have clearly demonstrated, studies of children and adults with autism have also demonstrated a broad but selective profile of deficits and intact or enhanced abilities that all reflect a relationship to information-processing demands. […] it is likely that genes affecting signaling pathways that regulate neuronal organization are strongly implicated in the etiology of autism.”

“ASD is now conceptualized as a developmental neurobiological disorder affecting elaboration of the forebrain circuitry that underlies the abilities most unique to human beings. […] Wiring the brain requires that neurons proliferate, acquire the correct identities, migrate to the appropriate locations, extend axons, and make guidance decisions with a high degree of spatial and temporal fidelity. Converging evidence indicates that more than one of these processes may be altered in various combinations to produce the heterogeneous phenotypes observed in ASD. […] Studies examining head circumference (HC) and brain volume (BV) in individuals with ASD have demonstrated altered brain growth trajectories across the lifespan. […]
• Up to 70 % of infants with ASD exhibit abnormally accelerated brain growth in the first year of life. Approximately 20% to 25% of infants in this subset actually meet formal criteria for macrocephaly (i.e., HC of 2.0 standard deviations above the mean) in the first year.
• BV is significantly larger by two to four years of life, and some children meet criteria for megalencephaly (i.e., BV 2.5 S.D. above mean).
• The first two years of life are usually a period of rapid brain growth in infants as neurons undergo significant postnatal growth in cell size and elaboration (actually overproduction) of axons, synapses, and dendrites. It is possible that this process is exaggerated somehow in at least a subset of ASD.
• Whatever the neurobiological basis, abnormal growth rates in ASD tend to decline significantly after the initial acceleration, causing an apparent “normalization” of BV by adolescence or early adulthood. […]
At the time of maximal brain growth in very early childhood, cerebral gray matter (GM) and white matter (WM) are both increased […] The frontal cortical GM and WM show the most enlargement, followed by the temporal lobe GM and WM and the parietal GM.”

“Thus far, [fMRI and fcMRI] studies have identified underconnectivity with the frontal cortex as a specific characteristic of the altered connectivity in autism, and this characteristic is present across the same wide range of domains of complex information processing that are affected in the disorder, including social, language, executive, and motor processes. […] measures of functional connectivity between specific areas have been shown to reliably predict the degree of impairment in specific domains among those diagnosed with autism. For instance, individuals with poorer social functioning measured by the ADI-R show lower functional connectivity between frontal and parietal cortices. These findings gave rise to the underconnectivity theory in autism, which now has sufficient support that it is accepted as a central feature of the pathophysiology of autism […] Results from these studies are consistent with the notion that autism is a disorder of distributed neural systems (e.g., the connections between structures rather than the structures themselves). […] Diffusion-weighted imaging measures the direction and speed of microscopic water movement in the brain, allowing inferences about the microstructure of the tissue that constrains such movement. These studies have consistently found reduced structural integrity of white matter in adults with ASD, indicating reduced anatomical connectivity […] like measures of functional connectivity, measures of anatomical connectivity derived from diffusion imaging have been shown to reliably predict symptom severity among individuals with autism.”

“In thinking about the genetic basis of autism, it is important to contrast syndromic (or complex) and non-syndromic (or idiopathic/essential) ASD. […] Syndromic ASD includes identifiable autism syndromes with known genetic causes, such as tuberous sclerosis complex, Fragile- X syndrome, Rett syndrome, and Smith-Magenis syndrome.
• Syndromic ASD is associated with a relatively higher propensity for dysmorphic features (including anatomical brain abnormalities), intellectual disability (ID), seizures, and female sex (sex ratios are almost equal).
• Syndromic ASD is also associated with a higher frequency of chromosomal abnormalities in general, many of which have been identified […]. However, it is not yet clear for many of these syndromes which features are typical of autism and which are unique.
Non-syndromic ASD is also called idiopathic autism and consists of cases with and without identifiable micro-deletions or duplications to the DNA. […] individuals with idiopathic ASD are more likely to be male, with sex ratios approximately 1:4 (F:M) but approaching 1:7 in milder cases.”

“Overall, approximately 10 % of children being evaluated for ASD are found to have an identified medical condition with a known genetic lesion such as Fragile X or tuberous sclerosis. An additional 10 % or more have an identifiable chromosomal structural abnormality or copy number variation associated with ASD. […] Recent genome-wide scans using microarray technology have demonstrated a substantial role for small chromosomal deletions or duplications (i.e., copy number variation or CNV) in the etiology of ASD. […] There is [however still] considerable debate concerning the genetic architecture underlying […] the majority of idiopathic autism. Arguments can be made for either the effects of single, but rare Mendelian causes (for which documented CNVs are presumably the tip of the iceberg) or the interaction of numerous common, but low-risk alleles. Genetic linkage and association studies have been traditionally employed to address the latter model, but have failed to consistently identify susceptibility loci.” [An important point I should perhaps make before finishing this post is that if incidence/prevalence of a condition is increasing fast in a population, which seems to be the case here, such an increase is in general considered to be unlikely to only be the result of genetic changes at the population level – that type of pattern is usually indicative of environmental factors playing an important role. It may well be that the ‘average cause’ is different from the ‘marginal cause’, and that it may be a good idea to be careful in terms of which tools to use to explain base rates and growth rates. It might be argued that increased assortative mating among nerds in Silicon Valley has increased incidence locally (I’m sure this might be argued as I’m quite sure I’ve seen this exact argument before…) and I’m not saying this may not be the case, but if close to one percent of the American population get diagnosed, what goes on in Silicon Valley probably isn’t super relevant one way or the other – only roughly 1% of the population live in that area altogether. Even if you were to argue that a similar process is going on everywhere else in the country, it sort of strains belief that ‘something else’ is not going on as well].

July 25, 2014 Posted by | books, genetics, health, medicine, personal, Psychology | Leave a comment

Sexually Transmitted Diseases (4th edition) (III)

I read the first nine chapters of this very long book a while back, and I decided to have another go at it. I have now read chapters 10-18, the first seven of which deal with ‘Profiles of Vulnerable Populations’ (including chapters about: Gender and Sexually Transmitted Diseases (10), Adolescents and STDs Including HIV Infection (11), Female Sex Workers and Their Clients in the Epidemiology and Control of Sexually Transmitted Diseases (12), Homosexual and Bisexual Behavior in Men in Relation to STDs and HIV Infection (13), Lesbian Sexual Behavior in Relation to STDs and HIV Infection (14) (some surprising stuff in that chapter, but I won’t cover that here), HIV and Other Sexually Transmitted Infections in Injection Drug Users and Crack Cocaine Smokers (15), and STDs, HIV/AIDS, and Migrant Populations (16)), and the last two of which deal with ‘Host Immunity and Molecular Pathogenesis and STD’ (Chapters about: ‘Genitourinary Immune Defense’ (17) and ‘Normal Genital Flora’ (19 as well as ‘Pathogenesis of Sexually Transmitted Viral and Bacterial Infections’ (19) – I have only read the first two chapters in that section so far, and so I won’t cover the last chapter here. I also won’t cover the content of the first of these chapters, but for different reasons). The book has 108 chapters and more than 2000 pages, so although I’ve started reading the book again I’m sure I won’t finish the book this time either. My interest in the things covered in this book is purely academical in the first place.

You can read my first two posts about the book here and here.

Some observations and comments below…

“A major problem when assessing the risk of men and women of contracting an STI [sexually transmitted infection], is the differential reporting of sexual behavior between men and women. It is believed that women tend to underreport sexual activity, whereas men tend to over-report. This has been highlighted by studies assessing changes in reported age at first sexual intercourse between successive birth cohorts15 and by studies that compared the numbers of sex partners reported by men and by women.10,13,16, 17, 18 […] There is widespread agreement that women are more frequently and severely affected by STIs than men. […] In the studies in the general population that have assessed the prevalence of gonorrhea, chlamydial infection, and active syphilis, the prevalence was generally higher in women than in men […], with differences in prevalence being more marked in the younger age groups. […] HIV infection is also strikingly more prevalent in women than in men in most populations where the predominant mode of transmission is heterosexual intercourse and where the HIV epidemic is mature […] It is generally accepted that the male-to-female transmission of STI pathogens is more efficient than female-to-male transmission. […] The high vulnerability to STIs of young women compared to young men is [however] the result of an interplay between psychological, sociocultural, and biological factors.33

“Complications of curable STIs, i.e., STIs caused by bacteria or protozoa, can be avoided if infected persons promptly seek care and are managed appropriately. However, a prerequisite to seeking care is that infected persons are aware that they are infected and that they seek treatment. A high proportion of men and of women infected with N. gonorrhoeae, C. trachomatis, or T. vaginalis, however, never experience symptoms. Women are asymptomatic more often than men. It has been estimated that 55% of episodes of gonorrhea in men and 86% of episodes in women remain asymptomatic; 89% of men with chlamydial infection remain asymptomatic and 94% of women.66 For chlamydial infection, it has been well documented that serious complications, including infertility due to tubal occlusion, can occur in the absence of a history of symptoms of pelvic inflammatory disease.65

“Most population-based STD rates underestimate risk for sexually active adolescents because the rate is inappropriately expressed as cases of disease divided by the number of individuals in this age group. Yet only those who have had intercourse are truly at risk for STDs. For rates to reflect risk among those who are sexually experienced, appropriate denominators should include only the number of individuals in the demographic group who have had sexual intercourse. […] In general, when rates are corrected for those who are sexually active, the youngest adolescents have the highest STD rates of any age group.5

“Although risk of HPV acquisition increases with number of partners,67,74,75 prevalence of infection is substantial even with limited sexual exposure. Numerous clinic-based studies,76,77 supported by population-based data, indicate that HPV prevalence typically exceeds 10% among young women with only one or two partners.71

“while 100 years ago young men in the United States spent approximately 7 years between [sexual] maturation and marriage, more recently the interval was 13 years, and increasing; for young women, the interval between menarche and marriage has increased from 8 years to 14. […] In 1970, only 5% of women in United States had had premarital intercourse by age 15, whereas in 1988, 26% had engaged in intercourse by this age. However, in 1988, 37% of never married 15-17-year-olds had engaged in intercourse but in 2002, only 30% had. Comparable data from males demonstrated even greater declines — 50% of never married 15-17-year-olds reported having had intercourse in 1988, compared with only 31% in 200299

“Infection with herpes simplex type 2 (HSV-2) is extremely common among FSWs [female sex workers], and because HSV-2 infection increases the likelihood of both HIV acquisition in HIV-uninfected individuals, and HIV transmission in HIV-infected individuals, HSV-2 infection plays a key role in HIV transmission dynamics.100 Studies of FSWs in Kenya,67 South Africa,101 Tanzania,36 and Mexico72 have found HSV-2 prevalences ranging from 70% to over 80%. In a prospective study of HIV seronegative FSWs in Nairobi, Kenya, 72.7% were HSV-2 seropositive at baseline.67 Over the course of over two years of observation […] HSV-2 seropositive FSWs were over six times more likely to acquire HIV infection than women who were HSV-2 seronegative.”

“Surveys in the UK133 and New Zealand134 found that approximately 7% of men reported ever paying for sex. A more recent telephone survey in Australia found that almost 16% of men reported having ever paid for sex, with 1.9% reporting that they had paid for sex in the past 12 months.135 Two national surveys in Britain found that the proportion of men who reported paying women for sex in the previous 5 years increased from 2.0% in 1990 to 4.2% in 2000.14 A recent review article summarizing the findings of various surveys in different global regions found that the median proportion of men who reported “exchanging gifts or money for sex” in the past 12 months was approximately 9-10%, whereas the proportion of men reporting who engaged in “paid sex” or sex with a sex worker was 2-3%.136

“There are currently around 175-200 million people documented as living outside their countries of birth.3 This number includes both voluntary migrants, people who have chosen to leave their country of origin, and forced migrants, including refugees, trafficked people, and internally displaced people.4 […] Each year about 700 million people travel internationally with an estimated 50 million originating in developed countries traveling to developing ones.98 […] Throughout history, infectious diseases of humans have followed population movements. The great drivers of population mobility including migration, economic changes, social change, war, and travel have been associated with disease acquisition and spread at individual and population levels. There have been particularly strong associations of these key modes of population mobility and mixing for sexually transmitted diseases (STDs), including HIV/AIDS. […] Epidemiologists elucidated early in the HIV/AIDS epidemic that there was substantial geographic variability in incidence, as well as different risk factors for disease spread. As researchers better understood the characteristics of HIV transmission, its long incubation time, relatively low infectivity, and chronic disease course, it became clear that mobility of infected persons was a key determinant for further spread to new populations.6 […] mobile populations are more likely to exhibit high-risk behaviors”

“Studies conducted over the past decade have relied on molecular techniques to identify previously noncultivable organisms in the vagina of women with “normal” and “abnormal” flora. […] These studies have confirmed that the microflora of some women is predominated by species belonging to the genus Lactobacillus, while women having BV [bacterial vaginosis] have a broad range of aerobic and anaerobic microorganisms. It has become increasingly clear that even with these more advanced tools to characterize the microbial ecology of the vagina the full range of microorganisms present has yet to be fully described. […] the frequency and concentration of many facultative organisms depends upon whether the woman has BV or Lactobacillus-predominant microflora.36 However, even if “normal” vaginal microflora is restricted to those women having a Lactobacillus-dominant flora as defined by Gram stain, 46% of women are colonized by G. vaginalis, 78% are colonized by Ureaplasma urealyticum, and 31% are colonized by Candida albicans.36 […] Nearly all women are vaginally colonized by obligately anaerobic gram-negative rods and cocci,36 and several species of anaerobic bacteria, which are not yet named, are also present. While some species of anaerobes are present at higher frequencies or concentrations among women with BV, it is clear that the microbial flora is complex and cannot be defined simply by the presence or absence of lactobacilli, Gardnerella, mycoplasmas, and anaerobes. This observation has been confirmed with molecular characterization of the microflora.26, 27, 28, 29, 30, 31, 32, 33, 34, 35

Vaginal pH, which is in some sense an indicator of vaginal health, varies over the lifespan (I did not know this..): In premenarchal girls vaginal pH is around 7, whereas it drops to 4.0-4.5 in healthy women of reproductive age. It increases again in post-menopausal women, but postmenopausal women receiving hormone replacement therapy have lower average vaginal pH and higher numbers of lactobacilli in their vaginal floras than do postmenopausal women not receiving hormone replacement therapy, one of several findings indicating that vaginal pH is under hormonal control (estrogen is important). Lactobacilli play an important role because those things produce lactic acid which lowers pH, and women with a reduced number of lactobacilli in their vaginal floras have higher vaginal pH. Stuff like sexual intercourse, menses, and breastfeeding all affect vaginal pH and -microflora, as does antibiotic usage, and such things may play a role in disease susceptibility. Aside from lowering pH some species of Lactobacilli also play other helpful roles which are likely to be important in terms of disease susceptibility, such as producing hydrogen peroxide in their microenvironments, which is the kind of stuff a lot of (other) bacteria really don’t like to be around: “Several clinical studies conducted in populations of pregnant and nonpregnant women in the United States and Japan have shown that the prevalence of BV is low (4%) among women colonized with H2O2-producing strains of lactobacilli. By comparison, approximately one third of women who are vaginally colonized by Lactobacillus that do not produce H2O2 have BV.45, 46, 47“.

My interest in the things covered in this book is as mentioned purely academical, but I’m well aware that some of the stuff may not be as ‘irrelevant’ to other people reading along here as it is to me. One particularly relevant observation I came across which I thought I should include here is this:

“The lack of reliable plenotypic methods for identification of lactobacilli have led to a broad misunderstanding of the species of lactobacilli present in the vagina, and the common misperception that dairy and food derived lactobacilli are similar to those found in the vagina. […] Acidophilus in various forms have been used to treat yeast vaginitis.144 Some investigators have gone so far as to suggest that ingestion of yogurt containing acidophilus prevents recurrent Candida vaginitis.145 Nevertheless, clinical studies of women with acute recurrent vulvovaginitis have demonstrated that women who have recurrent yeast vaginitis have the same frequency and concentration of Lactobacillus as women without recurrent infections.146 […] many women who seek medical care for chronic vaginal symptoms report using Lactobacillus-containing products orally or vaginally to restore the vaginal microflora in the mistaken belief that this will prevent recurrent vaginitis.147 Well-controlled trials have failed to document any decrease in vaginal candidiasis whether orally or vaginally applied preparations of lactobacilli are used by women.148 Microbial interactions in the vagina probably are much more complex than have been appreciated in the past.”

As illustrated above, there seems to be some things ‘we’ know which ‘people’ (including some doctors..) don’t know. But there are also some really quite relevant things ‘we’ don’t know a lot about yet. One example would be whether/how hygiene products mediate the impact of menses on vaginal flora: “It is unknown whether the use of tampons, which might absorb red blood cells during menses, may minimize the impact of menses on colonization by lactobacilli. However, some observational data suggests that women who routinely use tampons for catamenial protection are more likely to maintain colonization by lactobacilli compared to women who use pads for catamenial protection”. Just to remind you, colonization by lactobacilli is desirable. On a related and more general note: “Many young women use vaginal products including lubricants, contraceptives, antifungals, and douches. Each of these products can alter the vaginal ecosystem by changing vaginal pH, altering the vaginal fluid by direct dilution, or by altering the capacity of organisms to bind to the vaginal epithelium.” There are a lot of variables at play here and my reading of the results indicate that it’s not always obvious what is actually the best advice. For example an in this context large (n=235) prospective study about the effect of N-9, a compound widely used in contraceptives, on vaginal flora “demonstrated that N-9 did have a dose-dependent impact on the prevalence of anaerobic gram-negative rods, and was associated with a twofold increase in BV (OR 2.3, 95% CI 1.1-4.7).” Using spermicides like those may on the one hand perhaps decrease the likelihood of getting pregnant and perhaps lower the risk of contracting a sexually transmitted disease during intercourse, but on the other hand usage of such preparations may also affect the vaginal flora in a way which may make users more vulnerable to sexually transmitted diseases by promoting E. coli colonization of the vaginal flora. On a more general note, “The impact of contraceptives on the vaginal ecosystem, including their impact on susceptibility to infection, has not been adequately investigated to date.” The book does cover various studies on different types of contraceptives, but most of the studies are small and probably underpowered, so I decided not to go into this stuff in more detail. An important point to take away here is however that there’s no doubt that the vaginal flora is important for disease susceptibility: “longitudinal studies [have] showed a consistent link between increased incidence of HIV, HSV-2 and HPV and altered vaginal microflora […] there is a strong interaction between the health of the vaginal ecosystem and susceptibility to viral STIs.” Unfortunately, “use of probiotic products for treatment of BV has met with limited success.”

I should note that although multiple variables and interactions are involved in ‘this part of the equation’, it is of course only part of the bigger picture. One way in which it’s only part of the bigger picture is that the vaginal flora plays other roles besides the one which relates to susceptibility to sexually transmitted disease – one example: “Studies have established that some organisms considered to be part of the normal vaginal microflora are associated with an increased risk of preterm and/or low birth weight delivery when they are present at high-density concentrations in the vaginal fluid”. (And once again the lactobacilli in particular may play a role: “high-density vaginal colonization by Lactobacillus species has been linked with a decreased risk of most adverse outcomes of pregnancy”). Another major way in which this stuff is only part of the equation is that human females have a lot of other ways to defend themselves as well besides relying on bacterial colonists. If you don’t like immunology there are some chapters in here which you’d be well-advised to skip.

 

July 5, 2014 Posted by | books, data, demographics, health, medicine | Leave a comment

Nutrition at a Glance

This is a Wiley-Blackwell publication about human nutrition. It is also perhaps the strangest W-B publication I’ve ever read, because of the combination of the following two facts: i. Each chapter is two pages long (the book has 62 chapters). ii. This is an academic text without a single source or reference. The latter of those two points is the main reason why I have not rated the book.

The chapters are denser than you’d think (they have a lot of information considering what you’d expect from two-page chapters), and many chapters ‘come in pairs’ or deal with related stuff; for example there are three main chapters dealing exclusively with proteins – one about the ‘chemistry and digestion’ of proteins, another one about the ‘functions of proteins in the body’, and a third one about the ‘needs and sources’ of proteins. Carbohydrates and fats also get multiple chapters each, and micronutrients get 7 chapters of ‘exclusive coverage’ dealing only with those things. The level of detail is reasonably high (again considering what you’d expect), but of course there’s only so much stuff you can cram into a chapter two pages long. I think in many ways it is a really neat book to have for looking up stuff in this area that you’re wondering about and/or can’t quite remember (‘what was the role of butyric acid in the colon again?’ ‘Which factors affect calcium absorption?’ ‘What are the roles of the various B-vitamins in metabolism? How might I get into trouble if I don’t get enough riboflavin, and what can I do to avoid that situation?’). Quite a few of the things she talks about I don’t really consider it too problematic that she does not source; she knows a lot more about which role folate plays in the regulation of homocysteine levels than I do, and I don’t think there’s a big risk involved in just taking her word for it that those things work the way she says they do. Much of the book covers stuff on a level where I could justify thinking along such lines; many of the chapters are a bit like book versions of short Khan Academy nutrition lectures (perhaps a bit like lectures like this and this, I guess without having watched those lectures), and I figure if I’m okay with watching lectures like those I should be okay with reading a book like this as well, which is a big part of the reason why I didn’t just throw it away the moment I realized that there was actually not a single page of references to be found anywhere.

As I didn’t rate the book because of it, the lack of sourcing of course bothered me. One thing which puzzled me is why she decided to write the book this way – I simply do not understand that decision at all. Given the current state of affairs of nutritional science (and the current state at the time the book was written in 2007), I will say that I think the way she has chosen to write this book is simply flabbergasting to me. Nothing tempts people to disregard your information like not telling them where it comes from. There is not a single sentence in this book with the words ‘[X & Y] found that…’ – she only ever writes ‘a study found that…’, and this is just infuriating. There are various recommendations of daily intake of various substances in the book, but you have no idea who came up with those recommendations or which evidence base they are based on – there’s not even a source indicated in the headings in those cases. I simply don’t understand why she’d write the book that way – the lack of sources makes much of the stuff look deeply suspect, regardless of whether or not it’s actually all of it based on ‘the best available evidence’, and occasionally it seems as if she’s gone out of her way to avoid adding a source even in situations where it would make a lot more sense to add it than to not do that. To add insult to injury, a couple of the reported estimates in the last half of the book were so out of line with other estimates I could find elsewhere that I seriously considered throwing the book away. So, yeah.

But the book has a lot of good stuff as well which presumably a lot of people would benefit from knowing about, so it’s really hard for me to know what to think about it. To take an example of what I’m talking about here, Rees et al. observed in their book – as I have pointed out – that: “Vegans who omit all animal products from their diet often have subclinical vitamin B12 deficiency.” They probably wouldn’t have if they’d read this book. Anemia is a very common condition worldwide, and iron deficiency is estimated to be the most common cause. An estimated 250 million preschool children are vitamin A deficient. We humans need a lot of different stuff to keep going, and the food we eat plays many roles most of us probably haven’t given any thought. Everybody needs to eat, so there aren’t many people who would not benefit from knowing more about how these things work; even people following ‘an ideal balanced diet’ can be at risk of developing deficiency states due to malabsorption syndromes or various disease states which may change nutritional requirements.

I’ve added some ‘sample observations’ from the book, as well as a few comments, below:

“There is a continuous turnover of protein in the body, which in healthy adults exhibits a balance between synthesis and breakdown, and amounts to 3–6 g/kg body weight per day. During growth there is an excess of synthesis over breakdown, and in wasting conditions (e.g. starvation, cancer and after surgery or trauma), breakdown exceeds synthesis. Protein synthesis is regulated principally by insulin, and catabolism by glucocorticoids. […] The body is unable to make nine of the amino acids used in protein synthesis […] Lack of any one of these will limit the synthesis of protein, even if all the other required amino acids are present in adequate amounts. […] In addition, there are a number of other amino acids that can be synthesised in the body under normal conditions, given the necessary supply of precursor molecules. In the absence of these precursors, the amino acids become ‘conditionally indispensable […] Protein synthesis is an energy-demanding process; it has been calculated that the energy requirement is 4.2 kJ (1 kcal)/g of protein synthesised. Protein synthesis occurs more rapidly after a meal than in the fasting state, due to the greater supply of amino acids. On average the energy used in protein synthesis accounts for 12% of the basal metabolic rate. […] The digestibility of proteins from animal sources is much greater than that from plant sources. Digestibility for egg is given as 97%. […] Poor digestibility, of between 60 and 80%, is found in legumes and cereals with tough cell walls, particularly when uncooked, and is a factor in diets that are low in protein. […] Worldwide, the availability of plant proteins is relatively consistent, at about 50 g/person/day. However, the availability of animal protein sources varies widely, from <5 to 50 g/head/day, highest in most Western countries. […] Inadequate protein intakes rarely occur alone, and are generally found within a wider picture of undernutrition. Insufficient intakes of energy cause protein to be used for energy, and make it unavailable for tissue maintenance or growth.”

“The most rapid period of brain growth occurs from mid-gestation to 18 months after birth. At birth the brain accounts for 10% of the body weight; an adult brain weighs about 1.4 kg, and comprises 2% of body weight. Different components of the brain grow at different rates and have ‘critical periods’ when growth is most rapid and vulnerable to adverse influence.”

‘Alcohol and folic acid’ would probably be two factors most people would know about in this context. But there’s of course a lot more to developing a human brain than these variables. Generalized undernutrition can lead to smaller brain and less extensive neural networks, long-chain polyunsaturated fatty acids are quite important for brain development, you need copper for myelin synthesis but too much of it may be toxic, iodine deficiency leads to cretinism, severe iron deficiency may lead to long-term reductions in cognitive performance whereas too much of it again may lead to toxicity, excess vitamin A intake may be teratogenic, pyridoxine deficiency may lead to seizures/neurologic symptoms, … Incidentally the arrow doesn’t just go from food intake to brain performance; the brain is also helping you figure out what to eat: “neurotransmitters acting within the brain are thought to regulate preferences for particular macronutrients. Serotonin may influence the balance between carbohydrate and protein intakes. Noradrenaline and opiates are also believed to have a role. […] Disturbances of neurotransmitter release, whether of endogenous (e.g. in disorders of brain function) or exogenous origin (e.g. by drugs), are likely therefore to affect food intake.”

Some more stuff about childhood growth from a few chapters on related matters:

“The fastest rate of growth is in the first 6 months of life, with a doubling of birth weight, and slows towards 12 months, to achieve about three times birth weight. Body weight only doubles between the ages of 1 and 5 years. Standard growth charts are useful to check that growth is progressing appropriately […] During the years of [school-age] childhood, mean growth is relatively constant, and averages 2.5 kg and 6 cm per year. During puberty, on average: • girls increase by 20 cm (height) and 20 kg (weight); • boys increase by 30 cm (height) and 30 kg (weight). These increases represent 40% of eventual adult weight. Growth is vulnerable to faltering if nutritional intakes do not keep pace with the demands. […] Body fat percentage levels increase rapidly in the first months of life, but start to fall after the first year. […] There is a further increase from about the age of 5 years (adiposity rebound), which may start earlier in larger, fatter children […] In boys, the fat content starts to fall during the pubertal growth spurt, but in girls it continues to increase, resulting in the average 10% fat content differential between the sexes seen in adults. […] Growth in infants and young children, usually recorded as weight, should progress along a centile line on standard growth charts. Reasons for centile crossing (moving from one centile line to another) need to be established. […] Infants of diabetic mothers, who are often born very large (>4.5 kg), may exhibit ‘catch-down’ growth during the first year of life. Once removed from the oversupply of nutrients in the womb, their growth rate slows. […] An infant undernourished in the womb may show ‘catch-up’ growth. This should be an increase in lean body mass, rather than fat; the latter is linked to risk of later disease.”

“It is now recognised that vitamin D is synthesised in the skin by the action of ultraviolet light on a precursor, and could strictly be termed a hormone rather than a vitamin. Further, niacin [vitamin B3] can be made in the body from the amino acid tryptophan, so a separate supply may not be needed if protein intakes are adequate. However, in both of these cases, there are situations where synthesis is insufficient, and so a dietary need remains.”

“Inadequate intakes of macronutrients will most obviously be reflected in disturbed growth in children and body weight changes in adults. […] Undernutrition in the elderly is poorly reported, but is believed to be widespread […] The consequences of undernutrition can manifest both in the short and long term, and may have intergenerational effects, through poor pregnancy outcome and low birth weight. […] Worldwide, deficiencies of iron, vitamin A and iodine affect the greatest numbers of people. […] Several other micronutrients may become deficient when diets lack specific food groups. These include: • vitamin B12, when vegan diets are consumed; • calcium, when dairy products are excluded from the diet; • riboflavin, when diets are low in green vegetables and dairy products.”

“The typical increase in weight (in the UK) during pregnancy is 11–16 kg but varies widely. Gains in the second and third trimesters should average 0.4 kg/week for normal weight women, less (0.3 kg/ week) for overweight women and more (0.5 kg/week) for women who are underweight. […] The extra energy costs of pregnancy are estimated at 310 MJ (77 000 kcal) […] The mother’s nutritional status is unlikely to affect the volume or the macronutrient content of her milk for the first few weeks of lactation. However, poorly nourished women will not be able to sustain the same level of nutrients for prolonged periods. The fat content of the milk correlates with the mother’s levels of body fat, and the pattern of fatty acids secreted in the milk partly reflects those in the mother’s dietary intake. Neither the fat-soluble vitamin content nor the mineral content of the milk fluctuates with maternal dietary intake.”

ATP is the fundamental molecule that on breakdown to ADP provides energy for contracting muscle. ATP stores are very limited and require continual replenishment; the amount stored would fuel only about two seconds of exercise. […] CHO [Carbohydrate] stores in muscle (300–800 g) and liver (80 g) are limited; fats stored mainly in subcutaneous tissue are found in very much greater amounts (minimum 5 kg in males, more in females). • Lipids are considerably more energy dense than CHOs. Metabolism of one gram of fat will deliver considerably more ATP molecules than one gram of CHO; however, more oxygen is required to metabolise fats, and fats cannot be metabolised anaerobically […] An important principle is that carbohydrates are the preferred energy source particularly for more intense and prolonged exercise […] Most athletes already consume sufficient protein in a mixed diet […] With the exception of energy intake the evidence that normal dietary supplements enhance sporting performance is poor.”

I found the two chapters about nutrition and sport interesting in a way, but mostly because they helped me figure out what happens in normal people – a book like this (which I have considered reading in the past, but has never gotten around to actually reading) is probably better at elucidating relevant mechanisms in my case.

“Consumers generally believe that foods produced organically, often by more traditional agricultural methods, have superior nutritional quality. This is not currently supported by the scientific literature, in which studies find no difference in nutrient content between organic and non-organic produce; there is also no information about impact on human health.”

June 21, 2014 Posted by | books, health, medicine | Leave a comment

Impact of Sleep and Sleep Disturbances on Obesity and Cancer (2)

Warning: Long post.*

Okay, I’ve finished the book. I gave it five stars on goodreads – it’s come to my attention that I may be judging scientific publications like this one way too harshly, when you compare them with most other books. But then again I’d probably have given it four or five stars anyway; this book is an excellent source of information about the stuff it covers, and it covers a lot of stuff. In a way it’s hard to evaluate a book like this, because on the one hand you have a pretty good idea whether it’s enjoyable to read it or not, but on the other there are small chunks of it (or huge portions of it, or entire chapters, in the case of some readers…) which you are really not at all qualified to evaluate in the first place because you’re not actually sure precisely what they’re talking about**. Oh well.

As mentioned this book has a lot of stuff, and I can’t cover it all here. I’m annoyed about this, because it’s a great book. Some of this stuff is quite technical and there were parts of a few of the chapters I will not pretend to have really understood, but most of the stuff is okay in terms of the difficulty level – the book isn’t any harder to deal with than are most of Springer’s medical textbooks – and it’s interesting. In the first post I talked a little about sleeping patterns and a bit about cancer. The book has a lot of other stuff, and it has a lot of additional stuff about those things as well. Writing posts where I go into the details of books like these takes a lot of time and it’s not always something I have a great desire to do because it’s really hard to know where to stop. Let’s say for example that I were to decide to cover this book in great detail, and that I were to start out in chapter two, dealing with ‘Effects of Sleep Deficiency on Hormones, Cytokines, and Metabolism’. In that case I might decide to start out with these observations:

“Laboratory studies of both chronic and acute partial sleep restriction indicate that insufficient sleep can lead to increased hunger and caloric intake.”

“Many studies […] report that sleep independently relates to diabetes risk, even after controlling for the confounding effects of obesity and overweight. […] Cappuccio et al. [29] analyzed ten prospective studies with a pool of over 100,000 adults to ascertain the association of type 2 diabetes with sleep duration and quality. After controlling for BMI, age, and other confounding factors, they found [that] sleeping less than 6 h per night conferred an RR of 1.28 in predicting the incidence of type 2 diabetes, and prolonged duration (>8–9 h) had a higher RR of 1.48. As for sleep quality, Cappuccio et al. found that difficulty falling and staying asleep were highly correlated with type 2 diabetes risk with RRs of 1.48 and 1.84, respectively. […] a 3-year prospective study show[ed] that of workers with prediabetic indices, such as elevated fasting glucose, night-shift workers [were] at fivefold risk for developing overt diabetes compared to day workers [79].”

And I’d move on from there. So already here we’ve established not only that sleep problems may lead to changes in appetite which may lead to weight gain; that sleep problems and type 2 diabetes may be related, and perhaps not only because of the weight gain; that different aspects of sleep may play different roles (difficulty falling asleep doesn’t seem to have the same effect as does difficulty staying asleep); and that the time course from pre-diabetes to overt diabetes may be drastically accelerated in people who work night shifts. This is a lot of information, and we’re still only scratching the surface of that chapter (there are 11 chapters in the book). If I were to go into details about the diabetes thing I might be tempted to talk about how in another chapter they describe a study where three out of eight completely healthy young men were basically (temporarily) converted into prediabetics just by messing around a bit with their circadian clock in order to cause it to get out of sync with their sleep-wake cycle (a common phenomenon in people suffering from jetlag, and actually also a common problem, it seems, in blind people, as they’re generally not capable of using light to adjust melatonin release patterns and keep the circadian clock ‘up to date’, so to speak), but I really wouldn’t need to look to other chapters to talk more about that kind of stuff as the chapter also has some coverage of studies on hormonal pathways such as those involving leptin [a ‘satiety hormone’] and ghrelin [a ‘hunger hormone’]. The role of cortisol is also discussed in the chapter (and elaborated upon in a later chapter). I might decide to go into a bit more detail about these things and explain that the leptin-ghrelin connection isn’t perfectly clear here, as some studies find that sleep deprivation reduces leptin production and stimulates ghrelin release whereas other studies do not, but perhaps I’d also feel tempted to add that although this is the case, most studies do after all seem to find the effects you’d expect in light of the results from the weight gain studies I talked about in the first post (sleep deprivation -> less leptin, more ghrelin). But maybe then I’d feel the need to also talk about how it seems that these effects may depend on gender and may change over time (/with age). And I’d add that most of the lab studies are quite small studies with limited power, so it’s all a bit uncertain what all this ‘really means’. Perhaps I’d add the observation from the last chapter, where they talk more about this stuff, that the literature on these two hormones are not equally convincing: “Conflicting results have been presented for leptin […], although increases in ghrelin, an appetite-stimulating hormone, may be more uniformly observed.” Perhaps when discussion these things I’d opt for including a few remarks about the role of other hormones and circulating peptides as well, for example the “hypothalamic factors (e.g., neuropeptide Y and agouti-related peptide), gut hormones [such as] glucagon-like peptide-1 [GLP-1], peptide YY [PYY], and cholecystokinin), and adiposity signals (e.g., leptin and adiponectin)”, all of which are briefly covered in chapter 11 and all of which “have been demonstrated to play a role in the regulation of hunger, appetite, satiety, and food intake.”

As for the increased hunger and caloric intake observation, I might decide to talk about how there’s an ‘if you’re awake, you have more time to eat’-effect that may play a role (aside perhaps from the rare somnambulist, few people eat while they’re sleeping – and I’m not sure about the somnambulists either…) – but on the other hand staying awake requires more calories than does sleeping (“Contrary to the common belief that insufficient sleep reduces energy expenditure, sleep loss increases total daily energy expenditure by approximately ~5 % (~111 kcal/day).”). Those are sort of behavioural approaches to the problem, but of course there are many others and multiple mechanisms have been explored in order to better understand what happens when people are deprived of sleep – hormonal pathways is one way to go, I’ve talked a little about them already, and of course they’re revisited later in the chapter when dealing with type 2 diabetes. As an aside, in terms of hormonal pathways there’s incidentally an entire chapter on melatonin and the various roles it may play, as well as some stuff on insulin sensitivity and related matters, but that’s not chapter 2, the one we were talking about – however if I were to cover chapter 2 in detail I’d probably feel tempted to add a few remarks about that as well. But of course chapter 2 doesn’t limit coverage to just behavioural stuff and the exploration of hormonal pathways, as it seems that sleep deprivation also has potentially important neurological effects, in that it affects how the brain responds to food – and so in the chapter they talk about a couple of fMRI studies which have suggested this and perhaps indicated how those things might work, and they talk about a related study the results of which suggest that sleep deprivation may also induce impairments in self-control.

If I we’re to talk about the weight gain stuff in the chapter, I might as well also talk a bit about how sleep patterns may affect people when they’re trying to lose weight, as they talk a little bit about that as well. Those results are interesting – for example one study on weight loss that followed individuals for two weeks found that the individuals who were assigned to the sleep-deprivation condition (5.5 hours, vs 8,5 hours in the control group) had higher respiratory rates than those who did not. The higher respiratory rate the authors of the study argued was an indicator that the sleep-deprived individuals relied more on carbohydrates and less on fat than the well-rested controls, which is important if you’re dealing with weight loss regimes; however the authors in the book do not seem convinced that this was a plausible inference… Before going any further I would probably also interpose that how sleep affects breathing – and how breathing affects sleep – is really important for many other reasons as well besides weight loss stuff, so it makes a lot of sense to look at these things; stuff like intermittent hypoxia during the sleeping state, sleep disordered breathing and sleep apnea are topics important enough to have their own chapters in the book. Perhaps I’d feel tempted to mention in this context that there’s some evidence that people with sleep apnea who get cancer have a poorer prognosis than people without such sleep problems, and that we have some idea why this is the case. I actually decided to quote a bit from that part of the book below… But anyway, back to the weight loss study, an important observation from that study I might decide to include in my coverage is that: “shorter sleep duration reduced weight loss by 55 % in sleep-restricted subjects”. This is not good news, at least not for people who don’t get enough sleep and are trying to lose weight; certainly not when combined with the observation that sleep-deprived individuals in that study disproportionately lost muscle tissue, whereas individuals in the well-rested group were far more likely to lose fat. One tentative conclusion to draw is that if you’re sleep deprived while dieting your diet may be less likely to work, and if it does work the weight loss you achieve may not be nearly as healthy as you perhaps would be tempted to think it is. Another conclusion is that researchers looking at these things may miss important metabolic effects if they limit their analyses to body mass measures without taking into account e.g. tissue composition responses as well.

Actually if I were to talk about the stuff covered in chapter 2 I wouldn’t really be finished talking the type 2 diabetes and sleep problems even though I talked a little bit about that above, and so I’d probably feel tempted to say a bit more about that stuff. Knowing that sleep disorders may lead to a higher type 2 diabetes risk doesn’t tell you much if you don’t know why. So you could perhaps talk a bit about whether this excess risk only relates to insulin sensitivity? Or maybe beta cell function is implicated as well? We probably shouldn’t limit the analysis to insulin either – cortisol is important in glucose homeostasis, and perhaps that one plays a role? – yep, they’ve looked at that stuff as well. And so on and so forth … for example what role does the sympathetic nervous system and the catecholamines play in the diabetes-sleep link? The one you’d expect, or at least what you’d expect if you knew a bit of stuff about these things. A few conclusions from the chapter:

“Overall, studies suggest a strong relationship between insufficient sleep and impaired glucose homeostasis and cortisol regulation. These proximal outcomes may explain observed associations between sleep and the diabetes epidemic.” […] “The relationship suggested between sleep loss and sympathetic nervous system dysfunction [‘increased catecholamine levels’, US] proposes another likely mediator of several of the negative metabolic effects of sleep loss and sleep disorders, including insulin resistance, decreased glucose tolerance, and reduced leptin signaling”).

I’d still leave out a bit of stuff from chapter two if I were to cover it in the amount of detail ‘outlined’ above, but I hope you sort of get the picture. There are a lot of connections to be made here all over the place, a lot of observations which you can sort of try to add together to get something resembling a full picture of what’s going on, and it gets really hard to limit your coverage to ‘the salient points’ of a specific topic without excluding many important links to other parts of the picture and overlooking a lot of crucial details. There’s way too much stuff in books like these for me to really provide a detailed coverage of all of it – most of the time I don’t even try, though I sort of did in this post, in a way. I encourage you to ask questions if there’s something specific you’d like to know about these things which might be covered in the book; if you do, I’ll try to answer. Of course it’s rather easy for me to say that you can just ask questions about stuff like this which you’d like to know more about, as part of the reason why people read books like these in the first place is so that they can get at least some idea which questions it makes sense to ask. On the other hand people who don’t know very much about science occasionally manage to ask some rather interesting questions with interesting answers on the askscience-subreddit, so…

I’ve added some additional observations from the book below, as well as some further observations and comments.

“Over the past few decades, the drastic increase in the prevalence of obesity has been reflected by substantial decreases in the amount of sleep being obtained. For example, whereas in 1960 modal sleep duration was observed to be 8–8.9 h/night, by 2004 more than 30 % of adults aged 30–64 years reported sleeping <6 h/night [3]. More recently, the results of a large, cross-sectional population-based study of adults in the United States showed that 7.8 % report sleeping <5 h/night, 28.3 % report sleeping ≤6 h/night, and 59.1 % of those surveyed report sleeping ≤7 h/night [4].”

Regardless of the extent to which you think these two variables are related (and how they’re related), this development is interesting to me. I’m pretty sure some of the authors of the book consider the (causal part of the?) link to be stronger than I do. I had no idea things had changed that much. Okay, let’s move on…

“For many years, it has been known that the timing of onset of severe adverse cardiovascular events, such as myocardial infarction, sudden cardiac death, cardiac arrest, angina, stroke, and arrhythmias, exhibits a diurnal rhythm with peak levels occurring between 6 am and noon […] It is clear that many variables and parameters within the cardiovascular system are under substantial regulation by the circadian clock, highlighting the relevance of circadian organization for cardiovascular disease. Shift work has consistently been associated with increased cardiovascular disease risk [68–71].”

“Molecular oxygen (O2) is essential for the survival of mammalian cells because of its critical role in generating ATP via oxidative phosphorylation [the link is to a featured article on the topic, US]. Hypoxia, i.e., low levels of O2, is a hallmark phenotype of tumors. As early as 1955, it was reported that tumors exhibit regions of severe hypoxia [16]. Oxygen diffuses to a distance of 100–150 μm from blood vessels. Cancer cells located more than 150 μm exhibit necrosis. The uncontrolled cell proliferation causes tumors to outgrow their blood supply, limiting O2 diffusion resulting in chronic hypoxia. In addition, structural abnormalities in tumor blood vessels result in changes in blood flow leading to cyclic hypoxia [17,18]. Measurement of blood flow fluctuations in murine [rats and mice, US – a lot of our knowledge about some of these things come from animal studies, and they’re covered in some detail in some of the chapters in the book] and human tumors by different methods have shown that the fluctuations in oxygen levels in tumors vary from several minutes to more than 1 h in duration. Hypoxia in tumors was shown to be associated with increased metastasis and poor survival in patients suffering from squamous tumors of head and neck, cervical, or breast cancers [19,20]. Tumor hypoxia is associated with resistance to radiation therapy and chemotherapy and poor outcome regardless of treatment modality. Cancer cells have adapted a variety of signaling pathways that regulate proliferation, angiogenesis, and death allowing tumors to grow under hypoxic conditions. Cancer cells shift their metabolism from aerobic to anaerobic glycolysis under hypoxia [21] and produce growth factors that induce angiogenesis [22,23]. […] It is increasingly recognized that hypoxia in cancer cells initiates a transcription program that promotes aggressive tumor phenotype. Hypoxia-inducible factor-1 (HIF-1) is a major activator of transcriptional responses to hypoxia [24]. […] It is now well recognized that HIF-1 activation is a key element in tumor growth and progression.”

“the existing epidemiologic evidence linking OSA [Obstructive Sleep Apnea] and cancer progression fits some of the key classic causality criteria [40]: the association is biologically plausible (in view of the existing pathophysiologic knowledge and in vitro evidence); the existing longitudinal evidence supports the existence of temporality in the cause-effect association; the effects are strong; there is evidence of a dose-response relationship; and it is consistent with animal experimental models and other evidence. Lacking is evidence regarding another important criterion: that treatment of OSA will result in a decrease in cancer mortality. Future studies in this area are critical.
If verified in future studies, the implications of the evidence presented here are profound. OSA might be one of the mechanisms by which obesity is a detrimental factor in cancer etiology and natural history. From a clinical standpoint, assessing the presence of OSA (particularly in overweight or obese patients) and treating it if present might have to become a routine part of the clinical management of cancer patients.”

It’s perhaps worth mentioning here that this is but one of presumably a number of areas of oncology where sleep research has shown promise in terms of potential treatment protocol optimization. It’s observed in the book that the effectiveness of- and side effect profile of chemotherapies may depend upon which time during the day (/night?) the treatment is given, which also seems like something oncologists may want to look into (unfortunately it does not however seem like they’ve made a lot of progress over the years):

“Arguably, a field in which little progress has been made in linking circadian rhythms to pathology, disease pathogenesis, and/or clinical medicine at the molecular and genetic levels is cancer. This is unfortunate given that a diurnal rhythm in efficacy and sensitivity to chemotherapeutic agents was reported in mice over 40 years ago [92]. More recently, screening studies in rodents have demonstrated clear circadian rhythmicity in the antitumor activity and side effect profile of many anticancer agents, although at present, it is not possible to predict a priori at which time of day a given drug will be maximally effective (i.e., although rhythms are clearly present, little is known of their mechanistic underpinnings) [93]. Results such as these have given rise to the concept of “chronotherapeutics,” in which the time of drug administration is taken into consideration in the treatment plan in order to maximize efficacy and minimize toxicity […] Although some progress has been made, by and large, this approach has not made significant inroads into clinical oncology”

The stuff above is probably closely related to discoveries made by other contributors, described elsewhere in the book:

“Our laboratory used actigraphy to measure circadian activity rhythms, fatigue, and sleep/wake patterns in breast cancer patients. We found that circadian rhythms were robust at baseline, but became desynchronized during chemotherapy […] desynchronization was correlated with fatigue, low daytime light exposure, and decreased quality of life [21,32].”

Here’s some more stuff on related matters:

“A diagnosis of cancer and the subsequent cancer treatments are often associated with sleep disturbances. […] Prevalence rates for sleep disturbance among oncology patients range from 30% to 55% [in another chapter it’s 30% to 75% – either way these numbers are high, US] […]  These sleep disturbances can last for years after the end of the cancer treatment. In cancer patients and survivors, sleep disturbances are associated with anxiety, depression, cognitive impairment, increased sensitivity to physical pain, impaired immune system functioning, lowered quality of life, and increased mortality. Given these associations and the high prevalence of sleep disturbance in cancer patients, it is paramount that clinicians assess sleep disturbances and treat sleep disorders in cancer patients and survivors. […] The effects of chemotherapy and anxiety on sleep quality in [cancer] patients have been well studied, and interventions to improve sleep quality and/or duration among cancer patients have shown widespread improvements in cancer mortality and outcomes, as well as mental health, and overall quality of life” [for more on quality of life aspects related to cancer, see incidentally Goerling et al.]

“We have previously demonstrated an inverse association of self-reported typical hours of sleep per night with likelihood of incident colorectal adenomas in a prospective screening colonoscopy-based study of colorectal adenomas [10]. Compared to individuals reporting at least 7 h of sleep per night, those individuals reporting fewer than 6 h of sleep per night had an estimated 50 % increase risk in colorectal adenomas […] A recent study as part of the Women’s Health Initiative (WHI) has shown similar results with regard to risk of colorectal cancer [48].”

Remember here that colorectal cancer is one of the most common types of cancer in industrialized countries – “[t]he lifetime risk of being diagnosed with cancer of the colon or rectum is about 5% for both men and women in the US” – some more neat numbers here. The more people are affected by the disease, in some sense the ‘bigger’ these ’50 % increases’ get.

“Probably, the cancer for which sleep duration has been studied most with regard to risk is breast cancer. There are also a number of epidemiological studies that have investigated the association of sleep duration and risk of breast cancer. In these studies, the association of short sleep duration and incidence of breast cancer has been mixed […] In a large, prospective cohort of over 20,000 men, Kakizaki et al. found that sleeping 6 or fewer hours was associated with an approximately 38 % increased risk of prostate cancer, compared with those reporting 7–8 h of sleep […] New evidence is also emerging on the role of sleep duration in cancer phenotype […] Breast cancer patients who reported less than 6 h of sleep per night prior to diagnosis were about twice as likely to fall into the “high-risk” recurrence category compared to women who reported at least 7 h of sleep per night before diagnosis. This suggests that short sleep may lead to a more aggressive breast cancer phenotype.”

“Pain in cancer patients is most often treated with opioids, and sedation is a common side effect of opioids. However, the relationship between opioid use and sleep has not been well studied. Limited PSG data show that opioids decrease REM sleep and slow-wave sleep [31], suggesting that rather than improving sleep by being sedated, opioids may actually contribute to the sleep disturbances in cancer patients with chronic pain. In addition, the most serious adverse effect of opioids is respiratory depression which may exacerbate the hypoxemia in those individuals with SDB [Sleep Disordered Breathing] and thus lead to more interrupted sleep […it may also promote tumor growth and/or lead to poorer treatment outcomes – see above. On the other hand not treating pain in cancer patients is also … problematic (yet probably still widespread, at least judging from the data in Clark & Treisman’s book)]. […] Although pharmacotherapy is the most prescribed therapy for cancer patients with sleep disturbances [10,35], there is a paucity of studies related to pharmacologic interventions in cancer patients. A recent review concluded that evidence is not sufficient to recommend specific pharmacologic interventions for sleep disturbances in cancer patients [6]. […] As several studies have now confirmed the beneficial effects of cognitive behavioral therapy for insomnia (CBT-I) in cancer patients (mostly breast cancer) and survivors, CBT-I needs to be considered as the first-line treatment. Hypnotics are commonly prescribed to cancer patients. Despite this common use, little to nothing is known about the safety of these drugs in cancer patients. Given the possible interaction effects of the hypnotic/sedatives with cancer treatment agents, the side effects, and potential tolerance and addiction issues, the common use of these drugs in cancer patients is concerning.”

The book is not only about sleep, and this part I found interesting:

“Emerging evidence supports the hypothesis […] that shared mechanisms exist for the co-occurrence of common [cancer] symptoms […] an increased understanding of the mechanisms that underlie the co-occurrence of multiple symptoms may prove crucial to the development of successful interventions […] The study of multiple co-occurring symptoms in cancer patients has led to the emergence of “symptom cluster” research. […] Although awareness of the co-occurrence of symptoms has existed for over two decades […], the study of symptom clusters is considerably more recent [118]. An enduring challenge in the study of symptom clusters remains the lack of consistency in the methods used to cluster symptoms [119]. Currently, the analytic methods used to cluster co-occurring symptoms include correlation, regression modeling [120,121], factor analysis [122], principal component analysis [121,123], cluster analysis [104,111], and latent variable modeling [109]. […] the decisions that dictate the use of a specific approach are beyond the scope of this chapter […] Symptom cluster research can be grouped into two categories: de novo identification of symptom clusters (i.e., clustering symptoms) and the identification of subgroups of patients based on a specific symptom cluster (i.e., clustering patients ) […] De novo identification of symptom clusters is the most common type of symptom cluster research that occurs with oncology patients.”

A lot of stuff didn’t make it into this post, but I’ll stop here. Or should I also mention that aside from what you eat, it may also matter a lot when you eat (“a study in mice showing that animals fed a high-fat diet during their inactive phase gained more weight than mice fed during their habitual active phase”)? Or should I mention that “individuals with later sleep schedules tended” … in one study … “to have higher energy intakes throughout the day than those whose midpoint of sleep was earlier?” No, probably not. I wouldn’t know where to stop…

[This is a big part of the reason why I often limit my coverage of books to mostly just quotes. Posts like these have a tendency to blow up in my face, and if they don’t I often still find myself having spent a lot of time on them.]

*Or maybe it isn’t actually all that long, perhaps it’s just slightly longer than average? Anyway now that you’ve scrolled down from the top of the post to the buttom in order to figure out what that asterisk meant (if you didn’t scroll down and are now only reading this after you’ve read the entire post above, that’s your fault, not mine…), you’ll know whether you think it’s long. The warning seemed to carry more weight this way. That a warning like this should carry some weight seems quite important to me, considering that I’m blogging a book about obesity. A book about obesity which covers dietary aspects in some detail, yet is occasionally itself a bit hard to digest. [Permission to groan: Granted.]

**An example:

“Prolyl hydroxylase (PHD) is a tetrameric enzyme containing two hydroxylase units and two protein disulphide isomerase subunits, which requires O2, ferrous iron, and 2-oxoglutarate for PHD enzyme activity. In the presence of O2, PHD covalently modifies the HIFα subunit to a hydroxylated form, which by interacting with Von Hippel-Lindau (VHL) protein, a tumor suppressor, is subjected to ubiquitylation and targeted to proteasome, where it gets degraded [25]. Hypoxia inhibits PHD activity resulting in accumulation of HIF-1α subunit, which dimerizes with HIF-1β subunit.”

Yeah, that sounds about right to me…

There isn’t much of this kind of stuff in the book; if there had been I would not have given it five stars, because in that case I would not have found it at all interesting/enjoyable to read.

May 23, 2014 Posted by | books, cancer, diabetes, health, medicine | 2 Comments

Impact of Sleep and Sleep Disturbances on Obesity and Cancer (1)

“Sleep has recently been recognized as a critical determinant of energy balance, regulating restoration and repair of many of the physiological and psychological processes involved in modulating energy intake and utilization. Emerging data indicate that sleep can now be added to caloric intake and physical activity as major determinants of energy balance with quantitative and qualitative imbalances leading to under- or overnutrition and associated comorbidities. Considerable research is now focused on disorders of sleep and circadian rhythm and their contribution to the worldwide obesity pandemic and the associated comorbidities of diabetes, cardiovascular disease, and cancer. In addition to having an impact on obesity, sleep and circadian rhythm abnormalities have been shown to have significant effects on obesity-associated comorbidities, including metabolic syndrome, premalignant lesions, and cancer. In addition to the observation that sleep disturbances are associated with increased risk for developing cancer, it has now become apparent that sleep disturbances may be associated with worse cancer prognosis and increased mortality. […] circadian misalignment, such as that experienced by “shift workers,” has been shown to be associated with an increased incidence of several malignancies, including breast, colorectal, and prostate cancer, consistent with the increasing recognition of the role of clock genes in metabolic processes […] This volume […] review[s] current state-of-the-art studies on sleep, obesity, and cancer, with chapters focusing on molecular and physiologic mechanisms by which sleep disruption contributes to normal and abnormal physiology, related clinical consequences, and future research needs for laboratory, clinical, and translational investigation.”

I’m currently reading this book. I probably shouldn’t be reading it; I realized a couple of weeks ago that if I continue at the present rate I’ll get to something like 100 books this year, and despite some of these books being rather short and/or fiction books I don’t think this is a healthy amount of reading. It’s probably worth noting in this context that despite the fact that the number of ‘books read’ is now much higher than it used to be, I incidentally am far from sure if I actually read any more stuff now than I did in the past; it may just be that these things have become easier to keep track of as I now read a lot more books and a lot less ‘unstructured online stuff’. It’s not a new problem, but it’s getting rather obvious.

But anyway I’m reading the book, and although it may not be a good way for me to spend my time I am at least learning some stuff I did not know. The book is a standard Springer publication, with 11 chapters each of which deals with a specific topic of interest (a few examples: ‘Effects of Sleep Deficiency on Hormones, Cytokines, and Metabolism’, ‘Biomedical Effects of Circadian Rhythm Disturbances’, and ‘Shift Work, Obesity, and Cancer’). I’ve added some observations from the book below as well as some comments – I’ll probably post another post about the book later on once I’ve finished reading it. The very short version is that insufficient sleep may be quite bad for you.

“Insomnia, identified by complaints of problems initiating and/or maintaining sleep, is common, especially among women. Insomnia is often associated with a state of hyperarousal and has been linked to increased risk of depression, myocardial infarction, and cardiovascular mortality [15]. Relative risks for cardiovascular disease for insomnia have been estimated to vary from 1.5 to 3.9; a dose-dependent association between frequency of insomnia symptoms and acute myocardial infarction has been demonstrated [16]. Insomnia may be particularly problematic at certain times in the lifespan, especially in the perimenopause period and in association with acute life stresses, such as loss of a loved one. The occurrence of insomnia during critical periods, such as menopause, may contribute to increased cardiometabolic risk factors at those times. Short sleep duration may occur secondary to a primary sleep disorder or secondary to behavioral/social issues. Regardless of etiology, short sleep duration has been associated with increased risk of obesity, weight gain, diabetes, cardiovascular disease, and premature mortality [17,18].”

“Sleep is characterized not only by its presence or absence (and timing) but by its quality. Sleep is composed of distinct neurophysiological stages […] associated with differences in arousal threshold, autonomic and metabolic activity, chemosensitivity, and hormone secretion [2] […] Each sleep stage is characterized by specific patterns of EEG activity, described by EEG amplitude (partly reflecting the synchronization of electrical activity across the brain) and EEG frequency. Lighter sleep (stages N1, N2) displays relatively low-amplitude and high-frequency EEG activity, while deeper sleep (slow-wave sleep, N3) is of higher amplitude and lower frequency. Stages N1, N2, and N3 comprise non-rapid eye movement (REM) sleep (NREM). In contrast, rapid eye movement (REM) sleep is a variable frequency, low-amplitude stage, in which rapid eye movements occur and muscle tone is low. […] In adults, over the course of the night, NREM and REM sleep cycles recur approximately every 90 min, although their composition differs across the night: early cycles typically have large amounts of N3, while later cycles have large amounts of REM. The absolute and percentage times in given sleep stages, as well as the pattern and timing of progression from one stage to another, provide information on overall sleep architecture and are used to quantify the degree of sleep fragmentation. Sleep characterized by frequent awakenings, arousals, and little N3 is considered to be lighter or non-restorative and contributes to daytime sleepiness and impaired daytime function. Higher levels of N3 are thought to be “restorative.””

“The circadian rhythm changes with age and one important change is a general shift to early sleep times (advanced sleep phase) with advancing age. While teenagers and college students have a tendency due to both intrinsic rhythm and external pressures to have later bedtimes, this starts to wane in young adulthood. This phase advance to an earlier sleep time has been referred to as “an end to adolescence” and happens at a younger age for women than for men [60]. […] During the transition from adolescence to adult, several changes occur to the sleep architecture. Most notably is the significant reduction in stage N3 sleep by approximately 40 % as the child progresses through the teenage years […] This means that other stages of NREM (N1 and N2) take up more of the sleep time. Functionally this translates to the child having lighter sleep during the night and therefore is easier to arouse and awaken. […] The sleep architecture of young adults is […] in a 90-min cycle with all sleep stages represented. The amount of stage N3 sleep continues to reduce at this time, at a rate of approximately 2 % per decade up to age 60 years. There is also a smaller reduction in REM sleep during early and mid-adulthood. Once through puberty and into the 20s, most adults sleep approximately 7–8 h per night. This remains relatively constant through mid-adulthood. Young adults may still sleep a bit longer, 8–9 h for a few years. The need for sleep does not change as people progress to mid-adulthood, but the ability to maintain sleep may be affected by medical conditions and environmental influences. […] although average sleep duration does not change over adulthood, there is a large degree of inter- and intraindividual variability in sleep duration. Individuals who are consistently short sleepers (e.g., <6 h per night) and long sleepers (>9 h per night) and who demonstrate high between-day variability in sleep duration are at increased risk for weight gain, diabetes, and other metabolic dysfunction and chronic disease.”

“Nine retrospective studies have indicated that shift work might be associated with a higher risk of breast cancer, including three studies in Denmark, three studies in Norway, two studies in France, and one study in the United States. […] Three of four prospective studies have provided evidence in favor of an association between shift work and breast cancer. […] evidence for a relation between shift work and prostate cancer is very limited, both by the small number of studies and by major limitations involved in those studies that have been conducted”

The increased risk of breast cancer may well be quite significant not only in the statistical sense of the word, but also in the normal, non-statistical, sense of the word; for example the estimated breast cancer odds ratio of Norwegian nurses who’d worked 30+ years of nightwork, compared to those who hadn’t done any nightwork, was 2.21 (1.10-4.45) – and that study involved more than 40.000 nurses. Another study dealing with the same cohort found that the nurses who’d worked more than five years with schedules involving more than 5 consecutive night shifts also had an elevated risk of breast cancer (odds ratio: 1.6 (1.0-2.4)). It’s noteworthy that many of the studies on this topic according to the authors suffer from identification problems which if anything are likely to bias the estimates towards zero. As you should be able to tell from the reported CIs above, the numbers are somewhat uncertain, but that doesn’t exactly make them irrelevant or useless; roughly 1 in 8 women at baseline can expect to get breast cancer during their lifetime (link), so an odds ratio of, say, 2 is actually a really big deal – and even if we don’t know precisely what the correct number is, the risk certainly seems to be high enough to warrant some attention. One mechanism proposed in the shift work chapter is that the altered sleep patterns of shift workers lead to weight gain, and that weight gain is then part of the explanation for the increased cancer risk. I’ve read about and written about the obesity-cancer link before so this is stuff I know a bit about, and that idea seems far from far-fetched to me. And actually it turns out that the link between shift work and weight gain seems significantly stronger than does the link between shift work and cancer – which is precisely what you’d expect if it’s not the altered sleep patterns per se which increase cancer risk, but rather the excess adipose tissue which so often follows in its wake:

“Numerous epidemiologic studies have examined the association between shift work and obesity in various different countries. Most of these studies have utilized existing data from employment records in particular companies, which provide convenient but typically limited information on shift work and health-related variables because this information was not originally collected for research purposes. As a result, many of these studies have methodological issues that potentially limit the interpretation of their results. Still, 22 of 23 currently published studies found some evidence that obesity is significantly more common among individuals with shift work experience compared to those without such experience [36–57]; only one study did not identify a possible link [58]. […] many analyses of shift work and obesity lack adjustment for potentially important confounding variables (e.g., other health and lifestyle factors), and therefore prospective studies with more extensive information on these variables have provided critical insight. Four such prospective studies have been conducted, all of which indicate that individuals who perform shift work tend to experience significant weight gain over time — including two studies in Japan, one study in Australia, and one study in the United States. […] in the largest and most detailed analysis to date, each 5-year increase in rotating shift work experience was associated with a gain of 0.17 kg/m2 in body mass index (95 % CI = 0.14–0.19) or 0.45 kg in weight (95 % CI = 0.38–0.53), among 107,663 women who were followed over 18 years in the US Nurses’ Health Study 2 [57]. Statistical models were adjusted extensively for age, baseline body mass index, alcohol intake, smoking, physical activity, and other health and lifestyle indicators.”

A major problem with the ‘shift work -> obesity -> cancer’ -story is however that the identified weight gain effect sizes seem really small (one pound over five years is not very much, and despite how dangerous excess adipose tissue may be, those kinds of weight differences certainly aren’t big enough to explain e.g. the breast cancer odds ratio of 1.6 mentioned above) – the authors don’t spell this out explicitly, but it’s obvious from the data. It may be slightly misleading to consider only the average effects, as some women may be more sensitive than others to these effects and outliers may be important, but not that misleading; I don’t think it’s plausible to argue that this is all about body mass. In the few studies where they have actually looked at obesity as a potential effect modifier, the results have not been convincing:

“Although it is possible that obesity predicts both shift work and cancer risk — as would be required for obesity to be a potential confounding factor of this relation — it is probably more likely that shift work predicts obesity, in addition to obesity being a risk factor for many types of cancer. This scenario is suggested by the prospective studies of shift work and obesity described above; that is, obesity is a stronger candidate for effect modification than confounding of the association between shift work and cancer, as shift work appears to influence the risk of obesity over time. Yet, only three prior studies have conducted stratified analyses based on obesity status to evaluate the possibility of effect modification. Two of these studies focused on shift work and breast cancer, but they found no evidence of effect modification by obesity [24,26]; a third study of shift work and endometrial cancer did identify obesity as an effect modifier [32]. […] Clearly, additional studies need to carefully consider the role of body mass index—a possible confounding factor, but more likely effect modifying factor—in the association between shift work and obesity.”

I should make clear that although it makes sense to assume that obesity is a potentially major variable in the sleep-cancer risk relation, there are a lot of other variables that likely play a role as well, and that the book actually talks about these things as well even though I haven’t covered them here:

“Although the exact mechanisms by which various sleep disorders may affect the initiation and progression of cancer are largely unknown, disruption of circadian rhythm, pervasive in individuals with sleep disorders, is thought to be the underlying denominator linking sleep disorders, as well as shift work and sleep deprivation, to cancer. The circadian system synchronizes the host’s daily cyclical physiology from gene expression to behavior [55]. Disruption of circadian rhythm may influence tumorigenesis through a number of mechanisms, including disturbed homeostasis and metabolism (details provided in Chap. 2), suppression of melatonin secretion (details provided in Chap. 3), intermittent hypoxia and oxidative stress (details provided in Chap. 5), reduced capacity in DNA repair, and energy imbalance.”

The obesity link relates to a few of these, but there’s a lot of other stuff going on as well. I may talk about some of those things later – I thought chapter 7 was quite interesting, so I’ve ended up talking quite a bit about that chapter in this post, and neglected to cover some of the earlier stuff covered in the book.

May 21, 2014 Posted by | books, cancer, diabetes, health | Leave a comment

Acute Muscle Injuries

“This special edition about muscular injuries provides current knowledge to orthopaedic surgeons. Acute muscular injury is the most frequent trauma encountered by sport physicians and surgeons. These injuries must be well understood: physiology, biomechanics, healing process, but also epidemiology, nutrition, and psychology may explain not only the onset of the injury but also how to manage these lesions. The consequences of these lesions are sometimes dramatic […] After a global and transversal approach, ten chapters cover different localizations of acute muscle injuries. The chapter will address the definition of the injury, clinical aspects, complementary exams, preoperative findings, assessment and therapeutic options. […] injuries are described from trauma mechanism, physical examination findings and diagnostic and treatment algorithms towards rehabilitation programs and full return to sports. The book is structured in a fashion that allows people to use it as a reference manual. Therefore, this book is directed to orthopaedic surgeons, sports medicine physicians, physiotherapists, general practitioners, sports managers, athletes and coaches.”

I found it a bit funny that the book was directed to ‘athletes and coaches.’ The passage below provides part of the reason why:

“The plantaris muscle originates from the supracondylar ridge of the lateral femoral condyle and courses toward the posteromedial side of the lower leg. It inserts just medial of the Achilles tendon on the calcaneus. The plantaris muscle is located between the more superficial gastrocnemius and the deeper soleus muscle. The general function of this muscle group is plantarflexion of the ankle [29]. The soleus muscle originates from the proximal part (±1/2) of the posterior tibia along the soleal line and the proximal part (±1/3) of the posterior fibula. The gastrocnemius muscle spans three joints: the knee, ankle, and subtalar joint. The gastrocnemius is a bipennate muscle; the lateral head originates from the posterior aspect of the lateral femoral condyle, whereas the medial head arises from the medial femoral condyle. […] Repairs of distal ruptures generally focus on restoring the insertion of the distal biceps tendon on the radial tuberosity, although tenodesis to the brachialis tendon has also been reported [11,27]. The tendon and muscle can generally be mobilized to facilitate anatomic repair in acute cases, while augmentation is occasionally required in more chronic situations [45]. Repair was traditionally performed through an extensile Henry approach to the anterior elbow [53]. A high complication rate led Boyd and Anderson to develop the classic two-incision approach to anatomic repair [6].”

If you fail to see where I’m heading, combine the above stuff with this link (Danish link). No statistics seem to exist on this stuff, but the manager of the Danish union of soccer players says in the article that he estimates that only half of the professional soccer players even finish high school. I picked a couple of random pages from chapter 7 and ran them through this neat little tool – it gave a Flesch Reading Ease number of 33.17 (according to the wiki article, this is comparable to an issue of the Harvard Law Review, which “has a general readability score in the low 30s”), and a Gunning Fog index (‘estimates the years of formal education needed to understand the text on a first reading’) of roughly 17 (a score of 12 would be what you’d be aiming for if you wanted a high school grad to easily understand your text – do remember that half of the Danish soccer guys may well be below this level). You can always argue with the algorithms, but I’d remind you that if anything in cases like these they may well tend to underestimate the difficulty involved here – it’s much easier to understand a text with words like ‘strong and specific’ than it is to understand a text with words like ‘the medial pectoral’, yet the two combinations of words score the same in these algorithms (same number of words, same number of letters in each word). And there are a lot of words of the latter kind in this book. I say good luck reading and understanding this book if you’re a Danish high school dropout pro soccer player (I haven’t even talked about the fact that the book is written in a foreign language yet). Note incidentally that roughly one-third (31%, according to the introduction) of all elite soccer injuries are muscle injuries, and that “Hamstring injury is the single most common injury in professional football” (some of the findings of this paper are included in the book) – it’s not like the stuff covered in this book isn’t potentially relevant to these people.

I’ve often seen this sort of stuff before in the introductory chapters of academic publications – academics writing stuff which they delude themselves into thinking that a lot of other people will easily be able to read and understand – and I often have some trouble figuring out how to react to this. I find it sad in a way. I should perhaps point out in connection with this that I at least occasionally think about this type of stuff when blogging; however although I do put in at least some efforts trying to only post stuff ‘other people’ might at least arguably be expected to understand, at least to some extent, and avoid superfluous technicalities, I hardly delude myself into thinking that lots of people will not encounter some problems if they try reading some of the specific posts on this blog. This is not a problem as the blog isn’t really written for the general public; rather it’s mainly written for myself and the few other people out there who find the kind of stuff I write here interesting. I assume that given my limited interaction with ‘normal people’ I occasionally make inferential mistakes regarding comprehensibility qualitatively similar to those of the authors of this book, which is another reason why I have ambiguous feelings about this kind of stuff, but at least I have some awareness of these issues. One might of course argue that the authors assume that only a few players might benefit from the book, but given that some might benefit after all they chose to include that category of potential readers as well – and although I’m far from sure, that may be what’s going on. Maybe the fact that I so often read stuff that’s technically not ‘written for people like me’ makes me more attuned to these kinds of aspects than I perhaps ought to be.

So anyway, with that major digression out of the way, let’s talk a bit more about what this book is about. Naturally it doesn’t deal with all muscle injuries – there are a lot of muscles in the human body (‘approximately 642‘, though numbers vary – this seems like yet another area of biology where there’s a splitter/lumper dynamic at play) and it’d be a very long book if it dealt with every single one of them in detail. Aside from that a lot of textbooks have also already been written about some specific muscle groups, making coverage of those in a book like this somewhat unnecessary – for example there’s quite a literature on what happens when the heart muscle gets damaged and what you can do about it when that happens, so although acute muscle injuries that affect the myocardium may well be some of the most important ones in terms of human morbidity and mortality, they don’t really talk about that kind of stuff in this book. The main focus is on sports injuries – the first two chapters deal with general principles (‘Terminology and Classification of Athletic Muscle Injuries’, ‘Basic Principles of Muscle Healing’), whereas the rest – aside from the last one dealing with ‘Muscle Research: Future Perspective on Muscle Analysis’ – deal with specific muscles or muscle groups. The topics covered are: Hamstring injuries, Acute Adductor Muscle Injury, Quadriceps Muscles, The Calf Muscle Complex, Pectoralis Major Rupture, Acute Biceps Brachii Injuries, and Rectus Abdominis Injury. So what the book does is to provide you with an overview over some common injuries; how they present, diagnosis – it actually turns out that some muscle injuries are much harder to diagnose than you’d think, or at least they’re harder to diagnose than I thought they were – and treatment, etc. It’s probably close to an ideal book to own if you’re an athlete who’s just had a muscle injury; in all likelihood the book will contain some stuff about what’s going on, how worried you should be, how to optimally deal with the injury, etc. Some might presumably also argue that it’s an ideal book to have read if you’re an athlete at risk of getting an injury, and all athletes are, to some extent. Although the book is technical I’d say that if you’re reading this blog I’m pretty sure you’ll be all right – I don’t think a lot of high-school dropout professional soccer players read this blog, although I may be wrong about that…

Here are a few more wiki links (aside from the ones included in the text above) I looked up while reading the book: Anatomical terms of motion, Fascia, Ecchymosis, Myositis ossificans, Compartment syndrome (you do not want this), Tendinitis, Iontophoresis, Metaplasia, Tenosynovitis, Patella, Antalgic gait, Osteitis pubis. This is the kind of stuff that’s covered in the book. I gave the book three stars on goodreads.

I decided to include a few passages from the book below which I thought were interesting and/or worth knowing, as well as a few comments.

“Tears of the quadriceps tendon are a rare occurrence […] Patients who have suffered a complete or partial tear of the quadriceps tendon are typically older (>40 years old) and often have conditions that can lead to degeneration in the tendon […] Other patients who are at risk for quadriceps tendon tears are those that use performance-enhancing substances such as anabolic steroids and creatine. These drugs lead to increased muscle strength, and steroids have been reported to weaken tendons, change collagen fibril structure, and decrease tendon elasticity in animal studies. The combination of amplified muscle strength and a potentially weakened tendon increases the likelihood of suffering a tendon rupture.”

I had no idea this was a potential hazard associated with the usage of anabolic steroids. Although I’ve of course never considered using such drugs it’s probably safe to assume that some of the people who actually do use these drugs are also not aware of this risk. A quadriceps tendon tear is incidentally often quite unpleasant: “Incomplete or partial tears of the quadriceps tendon can often be managed nonoperatively. The patient’s knee should be immobilized in full extension for a period of up to approximately 6 weeks depending on the size of the tear.” 6 weeks immobilized – and these are the tears that are categorized as ‘mild’ (of course I’m not saying that if the people using such drugs knew about this risk they’d change their behaviour – some of the commonly known risks involved are much worse as they can actually kill you).

How to treat muscle injuries? The stuff included below is about how to deal with problems with the quadriceps tendons, but similar principles apply to many other muscle injuries (there may be a better coverage of this aspect elsewhere in the book, but this book was one of those books where I was unable to highlight and I am not going to reread the book just in order to find the absolute best quotes to post here):

“Treatment goals for muscle strains are aimed at minimizing the bleeding and hematoma formation following injury to the muscle. There is a scarcity of literature on the specific treatment of muscle injuries, including strains. Because of this, the treatment protocols have not changed drastically in recent years. Acute treatment for strains complies with the PRICE (Protection, Rest, Ice, Compression, and Elevation) protocol for the first 24–72 h following an injury. Ice and compression should be used for approximately 10–20 min at a time in hour-long intervals [10]. Protection and rest are aimed to prevent further damage to the muscle, while ice decreases blood flow, bleeding, and inflammation at the damaged area. The use of ice following acute muscle injuries has been shown to be effective in decreasing pain caused by the injury, but as of yet there is no definitive proof that it leads to faster healing and a quicker return to sports [10,17,25,37]. Compression and elevation both aid in decreasing blood flow and swelling in the injured area. […] Some centers believe that NSAIDs should be contraindicated due to the increased risk of local bleeding and the potential for slower healing of the injury. Therefore, their use is controversial [35]. […] Practitioners should attempt to avoid prescribing anti-inflammatory medications for patients with quadriceps tendon tears because they have been shown to impair tendon healing.”

I included the last part of that quote in order to illustrate that there’s actually quite a lot of stuff most people probably don’t know about these kinds of things, and that this lack of knowledge may easily lead to behavioural strategies post-injury which may adversely affect outcomes. Behavioural strategies which adversely affect injury outcomes may well from an opportunity cost perspective include the failure to adopt injury-minimizing behavioural protocols, and it’s certain there are some of these in the book which most people do not know anything about. In the specific case here it may be an idea to have in mind that it might well be better to use acetaminophen/paracetamol rather than, say, aspirin in an acute muscle injury context. I could easily include other examples as well from the book of ‘things athletes would benefit from knowing but mostly don’t’, here’s another one:

“If possible, a patient who has suffered a quadriceps contusion should immediately have the knee put in 120° of flexion for approximately 10 min. This has been reported to compress the injury to limit hemorrhage and muscle spasm. Research has shown that patients who are put in 120° of flexion immediately following a quadriceps contusion return to normal range of motion more quickly than those who do not and have a lower chance of developing myositis ossificans [2–4,27].”

A few more observations from the book, first a little bit of stuff about what the future may hold:

“The use of treatment modalities based on biologicals is a popular topic for musculoskeletal disorders. Many studies have evaluated platelet- and growth factor-enriched plasma for tendon pathology; results however vary substantially between studies and affected pathology [2,8,28]. […] A number of growth factors released by platelets, such as PDGF, VEGF, IGF-1, TGF beta, and FGF, promote repair in various soft tissue models. With the results of enriched plasma for other musculoskeletal pathologies in mind, it is a promising future treatment option. As with enriched plasma modalities, other upcoming treatment options also lack any evidence for the here-described pathology. One of them is mesenchymal stem cell (MSC) therapy: regeneration of healthy muscle tissue involves infiltration of tissue- and vascular-derived cells into the wound area, releasing a cascade of mediators (GFs, BMPs, cytokines, and neuropeptides). The hypothetical benefit of MSC therapy lies in the molecular approaches by which MSC, along with genetically modified cells and gene therapy, can synthesize and deliver the desired growth factor in a temporarily and spatially orchestrated manner to the site of injury [1,9].
The current lack of knowledge can be regarded as a contraindication because it is unsure whether the used modalities will enhance regeneration of functional musculous tissue or the formation of scar tissue. This is a serious concern and should be studied meticulously before it is ready to be used in daily clinical practice.”

And lastly a few words on how little we actually know at this point (having including the stuff above I felt that I had to include the stuff below as well):

“We are still faced with a dearth of scientific knowledge on muscle injuries. Despite the growing number of publications over the last three decades, our current knowledge on etiology, prognosis, and therapy is based on only 2,000 published injuries, of which the majority is acute hamstring injuries. If we restrict ourselves to level 1 trials, then there are less than 300 injuries examined. The progress our research has led to for the individual athlete is limited: compared to three decades ago the injury and re-injury rate have not been changed. […]

The device “prevention is better than cure” certainly holds for muscle injuries […] Nonetheless, to date the evidence for preventative measures is limited to only one high-level study, in which the Nordic hamstring exercise was shown to be effective in the prevention of hamstring injury in football [17].
To be able to direct preventative measures to those players at risk for a specific muscle injury, risk factors associated with the injuries need to be identified. Unfortunately, studies published to date on risk factors for muscle injuries have methodological limitations, as they use univariate approaches and have too small sample sizes to detect small to moderate associations. Muscle injuries in sports occur from a complex interaction of multiple risk factors. This multifactorial nature should be taken into account when studying risk factors for muscle injuries by using appropriate multivariate statistical approaches [3,13]. In addition, sample sizes should be sufficient to study associations of risk factors with injury risk. As clearly depicted by Bahr and Holme [3], to detect moderate associations up to 200 injured subjects are needed. Taking hamstring injuries in football as an example, with a seasonal injury prevalence of 17 %, a sample size of over 1,000 players is needed to study the risk factors with moderate associations. Risk factor studies in the other less prevalent muscle injuries will of course need even larger numbers of athletes. […] Our main limitation is that as an individual sports physician, we deal with a too limited number of muscle injuries to justify an experience-based approach. For example, in professional football, with 15 muscle injuries per team per season [7], our most experienced sports physicians will have had managed just 450 muscle injuries in his/her 30 years’ career. As a consequence, to gain expertise and to answer the most important and simple questions, we need to collaborate. Faced with our short research history, a worldwide muscle injury registration system should start today rather than tomorrow.”

From a brief google it incidentally seems as if that last part is a topic which has received some attention lately – see e.g. this and this.

April 28, 2014 Posted by | books, health, medicine | Leave a comment

Metabolic Risk for Cardiovascular Disease

Here’s what I wrote on goodreads:

“The book is well sourced and actually does a good job of covering much of the material. But the editor has done a poor job, and as a result the book seems very sloppy compared to similar scientific publications. There are multiple spelling errors and typos along the way, and it frankly seems as if the book was ‘published too fast’, before all the errors could be corrected. At first I punished this severely when I rated it by only giving the book 2 stars, but I realized this was too harsh. There’s a lot of interesting stuff included in the book.”

Here’s the kind of thing I’m talking about:

“Numerous cardiovascular abnormalities may be encountered in obese subjects (Table 6.4) it is not written properly in the PDF files that I have but this version seems correct. Health service usage and medical costs associated with obesity …”

That comment was one of a kind (fortunately), but there are a lot of errors and typos. At one point they talk about a marginally insignificant finding with an associated P-value of 0.52. This kind of stuff makes you look sloppy. The book is a Wiley-Blackwell publication and you kind of expect a bit more from books like these.

I’ve dealt with many of the topics covered in the book before (e.g. here, here and here, Khan Academy, etc.). I got the book in part to have a book in which I knew I could easily find a reference if/when I needed one, so that I wouldn’t have to look around a lot, and I think it’ll serve that purpose reasonably well. I gave the book 3 stars on goodreads. The book deals with many of the things you’d expect a book like this to cover; lipid and lipoprotein metabolism, insulin resistance and its role in cardiovascular disease, the obesity epidemic, hypertension, type 2 diabetes and the metabolic syndrome, tobacco use and cardiovascular disease and the role of physical exercise and nutrition, among other things. There was some interesting stuff in the book, but not a lot which was all that surprising. I really liked parts of chapter 11 on diabetes management and cardiovascular risk reduction; the chapter went over some reviews and a few major studies well known to people who’re interested in these things (ACCORD, ADVANCE), and the interpretation of the data by the author was somewhat different from interpretations I’ve seen in the past. One main point in the chapter is that lowering of Hba1c may be more effective in preventing cardiovascular events/disease progression among patients without overt cardiovascular disease; the argument being that lowering of blood glucose may protect vessels from getting damaged, but once they’re damaged lowing of Hba1c may not do much difference because it’s basically too late (in part because glycemic control may play a greater relative role in the early course of the disease process, compared to other factors, than it does in the later stages, where other mechanisms may conceivably take over to a greater extent – he doesn’t spell this out explicitly but I’d be surprised if he has not been thinking along those lines). In terms of previous trials looking at the link between glycemic control and cardiovascular disease (CVD), researchers have usually looked disproportionately at diabetics with manifest CVD; this is understandable as these patients are high risk. But such applied selection mechanisms in the past may mean (among other things) that these studies may have been underpowered to find the effects they were looking for. This is an interesting line of argument I have not seen before. If you’re wondering why this is important, it’s important because whereas the link between small-vessel disease and glycemic control is incontrovertible and has been for a long time, the link between macrovascular complications (CVD, etc.) and glycemic control has long been questionable, with a lot of mixed findings. Study selection designs and similar mechanisms may help partially explain why previous studies have not been able to establish a clear relationship. There are of course other complicating factors as well. As I think I’ve said before, until it’s perfectly clear to me that glycemic control and macrovascular disease are unrelated (or at least until we know in more detail how they are related), I’ll pretend that better glycemic control may have a protective effect on both small and large blood vessels. Note that the reason why this is important is also that diabetics make up a huge proportion of all heart disease patients; in Denmark the Danish Endocrine Society noted in a report published a few years ago (I can no longer find it online, unfortunately) that roughly half of all Danish patients with chronic ischaemic heart disease, AMI or heart failure have diabetes (of course a lot of them didn’t know that they did, but that’s a different discussion).

I’ve added some observations from the book below as well as a few comments:

a general rule is that CVD risk approximately doubles for each 20mmHg increment of systolic BP and 10mmHg increment of diastolic BP above 115/75mmHg […] a substantial excess risk of stroke death among those who are overweight or obese may be largely accounted for by a higher blood pressure [31].”

“Despite the fact that obesity has been shown to be an independent risk factor for CVD, many studies have reported that obese patients with established CVD have a better prognosis than do patients with ideal bodyweight; the socalled “obesity paradox.” […] The improved survival of obese individuals is paradoxical principally because of the assumption that excessive weight is always and invariably injurious. As a matter of fact, among patients with congestive heart failure, subjects with higher BMI are at decreased risk for death and hospitalization compared with patients with a “healthy” BMI [2]. Further, obesity was associated, in a prospective cohort study, with lower all-cause and cardiovascular mortality after unstable angina/non-ST-segment elevation myocardial infarction treated with early revascularization [89]. The obesity paradox may reflect the lack of discriminatory power of BMI to adequately reflect body fat distribution [20,87,90]. Since BMI measures total body mass, i.e. both fat and lean mass, it may better represent the protective effect of lean body mass on mortality. This negative confounding may have been under-appreciated in prior studies that did not adjust for measures of abdominal obesity. It is possible that the favorable prognosis implications associated with mildly elevated BMI might actually reflect intrinsic limitations of BMI to differentiate adipose tissue from lean mass. The lack of specificity of BMI could dilute the adverse effects of excess fat with the beneficial effects of preserved or increased lean mass [91]. […] Another issue to consider is that normal-weight patients may have a significantly higher percentage of high-risk coronary anatomy compared with obese patients [97]. […] Another limitation in most studies reporting an obesity paradox in patients with CVD is that non-intentional weight loss, which would be associated with a poor prognosis, is not assessed as BMI is measured only at the beginning of the study. Patients who have decompensated heart failure may lose weight because of extensive caloric demands associated with the increased work of breathing […] the excess health risk associated with a higher BMI declines with increasing age. An explanation for the lack of a positive association between BMI and mortality at older ages is that, in older persons, higher BMI is a poor measure of body fat and may simply represent a measure of increased physical activity with preserved lean mass. Sarcopenic obesity, which is defined as excess fat with loss of lean body mass, is a highly prevalent problem in the older individual. […] in view of the importance of body fat distribution, one could argue that, instead of targeting bodyweight per se, one should pay more attention to the WC [waist circumference] and conservation of lean mass as a critical goal in intervention programs [23].”

“Self-reported diabetes mellitus is often used in studies, but that approach underestimates the true prevalence of diabetes mellitus, and may misclassify a sizable fraction of the participants. […] it has been estimated that the lifetime risk of T2DM for persons born in the USA in 2000 is approximately 33% for men and 39% for women [9].”

“Summary analyses have reported that about 65% of deaths among diabetic patients are from vascular or heart disease, 13% are from diabetes itself, 13% are from neoplasms, and the rest are from other causes [14]. Most data concerning diabetes and death in adults are concerned with T2DM, and the limited data on mortality associated with type 1 diabetes mellitus have suggested that approximately one-third are from diabetes itself, one-third are from kidney disease, and one-third are from cardiovascular disease [15,16].” [I should note that some of these numbers sound wrong to me, but for now I’ll just report the numbers. I may have a closer look at the studies later. Note that ‘deaths from diabetes’ is a variable which is incredibly hard to get right in general; everybody dies, but diabetics die faster – deaths incontrovertibly ‘directly attributable’ to diabetes like DKA or hypoglycemic coma don’t make up all the ‘excess deaths’.] Researchers have investigated the effect of diabetes on life expectancy. An Iowa study showed that estimated life expectancy was 59.7 years at birth for diabetic men and 69.8 years in diabetic women, and it was estimated that diabetes reduced the lifespan by 9.1 years in diabetic men and 6.7 years in diabetic women [17]. From US national survey data it has been estimated that men known to have diabetes at age 40 years will lose 11.6 life-years and similarly affected women will lose 14.3 life-years [9].” [Again, for now I’ll just report the numbers…]

“The Centers for Disease Control reported that there were 8 million diabetic American adults with CVD in 1997 and the number increased to more than 11 million in 2007 […] reports suggest that diabetic patients continue to experience CVD at a high rate and are surviving, which has resulted in an increased prevalence of diabetic patients with CVD [21]. […] Fewer diabetes complications such as mortality, renal failure, and neuropathy have been observed for adult T1DM patients in the Pittsburgh Epidemiology of Diabetes Complications Study over recent years. On the other hand, risk of proliferative retinopathy, overt nephropathy, and clinical CAD have not declined over the long-term follow-up interval of 30 years [97]. […] Overall 1-, 2-, and 5-year survival after myocardial infarction in a population-based Swedish cohort was 94%, 92%, and 82%, respectively, in non-diabetic patients and 82%, 78%, and 58%, respectively, in diabetic patients.” [I.e., the proportion of diabetics who can expect to survive one year after an MI corresponds to the proportion of non-diabetics who can expect to survive five years.]

“In the mid-1990s there was considerable interest in the potential benefit of antioxidant nutrients and CVD risk reduction [100–103]. Since that time a series of randomized controlled intervention trials have failed to demonstrate a benefit of vitamin E or other antioxidant vitamin supplementation on CVD risk [104, 105]. The most recent work focusing on vitamins C and E confirm these earlier trials [106]. At this time the data do not support a recommendation to use antioxidant vitamins for the prevention or management of CVD. […] The three major dietary omega-3 polyunsaturated fatty acids (PUFAs) are alphalinolenic acid (ALA, 18:3n-3), eicosapentaenoic acid (EPA, 20:5n-3), and docosahexaenoic acid (DHA,22:6n-3). The later two fatty acids are sometimes referred to as very-long-chain n-3 fatty acids. […] a number of studies have reported an inverse association between dietary n-3 fatty acids, CVD and stroke risk [41]. Intervention data have demonstrated that EPA and DHA, but not ALA, benefit cardiovascular outcomes in primarily and secondary prevention studies [42] […] Of note, the relationship between arrhythmea and EPA and DHA has recently been questioned [45]. The major source of ALA in the diet is soybean and canola oils, whereas the major source of EPA and DHA is marine oils found in fish.”

“The lipoproteins are defined by their density, for example, very low density (VLDL), low-density (LDL), and high-density (HDL). In this instance, “density” is mostly related to the triglyceride and cholesterol content; the more lipids in a lipoprotein the lower its density, as measured by how readily it floats toward the top of a tube during ultracentrifugation. TG-rich lipoproteins transport an energy source, triglyceride, to muscle and adipose tissue for use and storage. TG-rich lipoproteins also contain cholesterol, and can deliver the cholesterol to peripheral tissues and the arterial wall. LDL is a transporter of primarily cholesterol from the liver to peripheral tissues. HDL also functions to transport cholesterol but in the reverse direction as VLDL and LDL, from peripheral tissues to the liver. Lipoproteins also are required to transport fat-soluble vitamins.”

Relatively consistent evidence indicates that increasing the carbohydrate content of the diet at the expense of fat results in dyslipidemia [7–9]. The majority of the evidence suggests that carbohydrate-induced hypertriglyceridemia results from an increased rate of hepatic fatty acid synthesis [10,11] and subsequent production of hepatic triglyceride-rich particles, very-low-density lipoprotein (VLDL) […] Within the context of a stable bodyweight, replacement of dietary fat with carbohydrate results in higher triglyceride and VLDL cholesterol concentrations, lower HDL cholesterol concentrations and a higher (less favorable) total cholesterol to HDL cholesterol ratio [16–21]. […] Sedentary individuals characterized by visceral adiposity are at particularly high risk for carbohydrate-induced hypertrygliceridemia [9]. […] Studies performed in the mid 1960s demonstrated that changes in dietary fatty acid profiles altered plasma total cholesterol concentrations in most individuals […] Many studies have since confirmed these early observations using a variety of different experimental designs [29]. When carbohydrate is displaced by saturated fatty acids, LDL cholesterol concentrations increase, whereas when carbohydrate is displaced by unsaturated fatty acids LDL cholesterol concentrations decrease, with the effect of polyunsaturated fatty acids greater than monounsaturated fatty acids […] When carbohydrate is displaced by saturated, monounsaturated or polyunsaturated fatty acids, HDL cholesterol concentrations are increased, with saturated fatty acids having the greatest effect and polyunsaturated fatty acids having the least effect.”

“Some agents affect HDL and TG in the same direction. Drinking alcoholic beverages and postmenopausal estrogen treatment raise HDL and TG. Testosterone lowers HDL and TG. Since we do not have a way as yet to evaluate the function of HDL in reverse cholesterol transport [one of the chapters spends a significant amount of time on that one – there’s a lot more to be said about that stuff than what’s in the wiki article], we cannot be confident that these or any changes in HDL concentration affect atherosclerosis in the direction expected from the relation of HDL concentrations and CHD risk [59,65]. There is also no clear relation between genetic variants in enzymes or transporters in HDL metabolism that cause either very low or high HDL cholesterol concentrations and CHD [74].” [HDL is usually termed ‘good cholesterol’, but in reality it’s much more complicated than that. We are very sure by now that high ‘anything which is not HDL’ is bad for you, though – in fact:]The combination of VLDL cholesterol and LDL cholesterol, named “non-HDL cholesterol” [2], or perhaps better “atherogenic cholesterol,” is a measurement that generally predicts CVD better than LDL-C [LDL-Cholesterol].”

February 16, 2014 Posted by | books, diabetes, health, medicine | Leave a comment

Some more health economics stuff

The Oxford Handbook of Health Economics (here’s a previous post about the book) has a lot of stuff. Here are some more observations which might be of interest:

User charges:

“Out-of-pocket payments play a dominant role in LMICs [low- and middle-income countries] where they cover about 50 percent of health care expenditures. [They] are less important in the high-income countries [but] there seems to be a tendency toward an increase of patient cost-sharing in countries where it traditionally has played a minor role […] This is not only explained by a concern to fight moral hazard and overconsumption, but it also reflects the increasing pressure on the public financing part of the system.” [In ‘low-income-countries’ out-of-pocket expenditures in 2008 made up on average 67.4 % of total expenditures in health, whereas the corresponding numbers for ‘lower middle income countries’, ‘upper middle income countries’ and ‘high income countries’ were 46.8%, 30.2% and 14.4% respectively. The global average was 22.5%. Note that total out-of-pocket expenditures incurred in high-income countries (in e.g. dollars) may make up a much larger share of the total global out-of-pocket expenditures than you might believe from those numbers alone – recall that high income countries spend approximately 100 times as much money on health per person than do low-income countries (low income countries spent on average $23 on health per capita in 2008, whereas high income countries spent $2414 per capita).]

“User charges do have a negative effect on health care consumption. The evidence is overwhelming for co-payments in the developed insurance systems. […] The evidence is almost equally strong for the effects of user charges in LMICs. Introducing or increasing user fees has almost always and everywhere led to a decrease of utilization […] both in developed health insurance systems and in LMICs, the evidence suggests that the decrease in utilization may have negative effects on the quality of care […] Most studies find that cost sharing leads to a decrease in the utilization of essential medication, defined as medication that is necessary to maintain or improve health. Often adherence to a regimen of maintenance medication goes down with patients skipping doses or stretching out refills. With a few exceptions […], higher cost-sharing for, and therefore lower utilization of, prescription drugs, has led to greater use of inpatient and emergency medical services by chronically ill patients [effects like these, I should point out, may well make cost-sharing a less than ideal cost-saving mechanism; emergency services are incredibly expensive compared to ‘routine management’] […] cross-price effects are also significant. Again, the evidence for the developed countries and the LMICs goes in the same direction. Two- or three-tier plans for prescription drugs in the US, introducing differentiated cost sharing for different categories of drugs, have clear effects on the pattern of drug use. […] user charges are a strongly regressive component in the health care financing structure of developed countries […] A large majority of studies suggest that user charges lead to a stronger reduction in utilization among the poor than among the rich (James et al. 2006).”

Insurance and the demand for medical care:

two main empirical findings from research to date are these: (1) the aggregate or average consumer demand curve, whether Marshallian (uncompensated) or Hicksian (compensated), slopes downward and to the right. (2) Demand curves are significantly price responsive at all consumer income levels. These conclusions are at variance with common perceptions of medical care demand by non-economists, who traditionally have asserted that non-poor consumers only use medical care when they have to do so because they are sick or are ordered to do so by their physician, and that only lower income households would restrain their demand for needed care because of cost sharing.”

“When the consumer has price-sensitive demand for care, the influence of deductibles on spending is complex because a deductible in effect faces the consumer with a two part block tariff: full price up to a certain level of spending, and then low or zero marginal price. Since the marginal price is different depending on whether the deductible is covered or not, the consumer has to consider the distribution of expected expenses […] While the actual analytics of demand responsiveness are complicated by a deductible […] the main intuitive finding is obvious: the lower the deductible the higher the demand for care, other things equal.”

Hospitals:

“A traditional discussion about choosing the “right” (desired) hospital output is associated with the role of quality and the trade-off between lower costs and higher quality. This trade-off is based on the assumption that higher quality implies more costs. This is likely to be so in efficient hospitals. However, inefficient hospitals may have room to improve simultaneously in both dimensions.”

“Whenever hospitals are funded by case payments they prefer to receive more patients for treatment while hospitals funded by capitation (to treat people in a defined catchment area) will invest more in keeping patients treated at primary care level when clinically feasible. […] several health systems make referral by a GP a necessary condition to visit a specialist. […] Gatekeeping determines to a considerable extent the demand faced by the hospital. Moreover, referrals to the hospital depend on both the incentives faced by GPs and on the formal relationship between primary care and hospitals.”

One of the main issues in measuring economies of scale (productivity, in general) in hospitals is the role of quality. More efficient hospitals are more likely to have a lower marginal cost of providing quality, and accordingly they may supply a higher quality level in equilibrium (which is likely to raise costs and mask their efficiency advantage). […] Under regulated prices, quality is the main “competitive tool” of hospitals and it is used intensively. Whenever both price and quality are available instruments to the hospital, the effort to attract patients is spread over both of them. […] response to an unexpected demand surge for hospital services is more likely to be met by early discharges to free up capacity than rationing admissions.”

The economics of the biopharmaceutical industry:

“The US research-based biopharmaceutical industry invests 15-17 percent of sales in R&D, and the R&D cost of bringing a new compound to market is estimated at over $1bn. […] The cost of developing an approved new medical entity (NME), measured as a discounted present value at launch, [was] $138 million in the 1970s […] the global nature of pharmaceutical R&D raises issues of appropriate cross-national price differentials and cost sharing. National regulators have incentives to free-ride, driving domestic prices to country-specific marginal cost, leaving others to pay for the joint costs of R&D. The long R&D lead times – on average roughly twelve years from drug discovery to product approval – make the incentives for short run free-riding by individual countries particularly acute because negative effects will be delayed for years and hard to attribute. […] In practice, the ability of pharmaceutical firms to price-discriminate is undermined by government policies […] the design of each country’s price regulatory system affects not only its prices and availability of drugs but also availability in other countries through price spillovers in the short run, and through R&D incentives in the long run. […] North America accounted for 45.9 percent of global pharmaceutical sales in 2007, compared to 31.1 percent for Europe

“The theoretically optimal insurance/reimbursement contract for drugs must deter both insurance-induced over-use by patients and excessive prices by manufacturers, while paying prices sufficiently to reward appropriate R&D, taking into account the global scope of pharmaceutical sales. […] An important conclusion is that patient cost sharing alone cannot simultaneously provide optimal incentives for efficient use of drugs, control of patient moral hazard and optimal provider incentives for R&D. In addition, given the global nature of pharmaceutical utilization, creating optimal R&D incentives require appropriate price differentials across countries […] generally, regulatory systems that induce price convergence across countries are likely to reduce social welfare. […] Overall, countries that use direct price controls do not consistently have lower prices than countries that use other indirect means to constrain prices”

In the US, generics now account for almost seventy percent of all prescriptions but only about 16 percent of sales, due to their low prices. Although US prices for on-patent drugs are on average 20-40 percent higher in the US than in other industrialized countries, US generic prices are lower […] many middle and low income countries have relatively high generic prices […] and uncertain generic quality. […] Empirical studies of generic entry has shown, not surprisingly, that generic prices are inversely related to number of generic competitors […]; generic entry is more likely for compounds with large markets […], [and in] chronic disease markets”

“the FDA is required by statute to consider risks and benefits to patients. Costs […] is beyond the FDA’s purview. […] Currently the US lags other countries in the use of comparative and/or cost-effectiveness as an input to reimbursement decisions.”

February 10, 2014 Posted by | books, economics, health, health care, medicine | Leave a comment

The Oxford Handbook of Health Economics (2)

Oxford Handbook

Here’s the first post about the book. If you want to know what I’ve been doing over the last few days, look at the red thingy. I’ve read roughly 700 pages so far. Book-blogging takes time, so I’ve been emphasizing reading over blogging.

This book covers a lot of stuff. There’s a lot more in there than I can justify covering here. That said, in a way I also feel that it’s necessary to note how little stuff the book actually covers: In more than a few chapters I’ve added remarks such as, ‘this topic is covered in much more detail in Juth and Munthe‘, or ‘for a much more comprehensive review, see Goldstein’s book‘. The book also covers stuff covered in greater detail here, here and here; as mentioned before I know a bit about these things already, though I haven’t felt like I’ve really had a great overview of the material. Having read books like this one, this one or perhaps this one may help understand some issues presented in specific chapters better, but you don’t really need to have done that – in most cases the chapters can stand on their own. I should mention that in one specific chapter (about addiction) I basically wrote in the margin at one point that the authors didn’t seem to know what they were talking about here, and that they should familiarize themselves with the medical- and neuroscientific research on the stuff they’d written about there (addiction) before writing any more stuff on that (specific) subtopic. It wasn’t a big part of the chapter though, and that has only happened once; most chapters are great, and none are what I’d really term ‘weak’ – I’m currently at either four or five stars on goodreads, probably a little bit closer to five than four. I should note that I have had similar ‘these guys don’t seem to know a lot about what the non-economists have found out about this stuff’- experiences as I had when reading the addiction chapter previously a few times when covering labour economics topics during my coursework; sometimes it seems to me that economists who’re very fond of their models (and the models of their antecessors) don’t really have a clue what’s really going on because they refuse to learn what people in other fields have already found out (perhaps because they assume that related work will not help them in their model-building efforts? (…if so, I think they’re wrong)) – it always bothers me.

Anyway, some observations from the book below:

Infectious diseases:

“From an economist’s perspective, infectious diseases are distinguished from many other health issues by the central role played by externalities.1 Control of infectious diseases yields both positive externalities (prevention and treatment can delay or reduce spread of infection to uninfected individuals) and negative externalities (overuse of treatment can lead to drug resistance, which has global consequences for treatment effectiveness). […] vaccination, an important tool in the prevention of infectious diseases, presents a classic public goods problem. Society gains from individual vaccination because of herd immunity, but this value is not recognized by individuals, who have an incentive to free-ride on vaccination by other individuals. […] disease reporting and eradication efforts are also public goods. […] a country’s incentives to control a freely moving disease like malaria are determined as much by its ability to stop the inflow of infected individuals as by the ability to control the disease within its own borders. Reducing malaria in a country could have transboundary benefits by incentivizing infection control in its neighboring countries as well. This principle also applies more generally to the challenge of global disease eradication. […] Eradication is a binary public good: the maximum benefits are achieved when the disease is completely gone.”

“Together, all infectious diseases account for more than 25 percent of premature death globally.”

“In sum, obtaining accurate information about potential epidemics is as much about reporting incentives as it is about detection technology.”

“from an economic perspective, disease burden may be a poor criterion to use for allocating treatment resources.”

Mental health:

“(OECD) nations commonly spend between 5 percent and 14 percent of their health dollar on mental health care […] this implies that OECD countries devote between 0.3 percent and 1.1 percent of their national incomes to treatment of mental disorders.2 […] It is important to note that the patterns of spending on mental health care are different from those observed in international comparisons of health care spending. […] there is more variation in mental health spending levels across nations than there is for health care. […] The commitment by OECD countries to promote community-based treatment and inclusion of people with mental disorders into the mainstream of society while also accepting the responsibility for public protection creates a policy tension that […] shapes public mental health spending. […] there have been notable reductions in the inpatient psychiatric capacity in virtually all OECD countries [since the 1960s]. […] [There is] growing variation in how each society sees the function of the psychiatric hospital. […] in France and the United States, two countries that spend similar shares of GDP on mental health care, France allocates roughly 80 percent of mental health spending on inpatient care (Verdoux 2007) and the United States about 36 percent (Mark et al. 2007). […] Mental health spending in the US as a share of total health spending has declined from nearly 11 percent in the 1970s to 6.2 percent in 2003”

“Cost-effectiveness evaluations of evidence-based treatments for depression suggest that they produce gains in Quality Adjusted Life Years (QALYs) at levels comparable to other medical treatments […] rates of treatment for the mental disorders, with some of the strongest effectiveness of care evidence, such as depression and anxiety disorders, are quite low […] mental health services are frequently funded and/or supplied by several bureaucratic departments all operating under fixed budgets. […] There may therefore exist opportunities for cost shifting. That is, strict rationing of mental health services may be seen as an opportunity to expand monies available for general medical care while allowing people with mental disorders to obtain care from the social care sector. […] recently the creation of combined trusts (mental health and social care) has tried to use organizational design to blunt incentives to cost shift created by fragmentation in financing.”

Public sector health care financing:

“In general, it can be shown to be efficient for the consumer’s cost-share to be lower when he or she incurs large health care costs, but higher with relatively low costs. This can be accomplished via a plan with an initial deductible (under which consumers are responsible for 100 percent of their health care costs in a given period of time, up to the limit of the deductible), followed by one or more intervals of partial cost sharing, perhaps up to some maximum (a “stop-loss” provision) beyond which the plan pays 100 percent of any additional costs. […a related observation from another chapter: “In the pure theory of insurance, Arrow (1963) showed that, with proportional administrative loading, optimal coverage is full coverage above a deductible” – this result is called ‘Arrow’s theorem of the deductible’ and lots of people have written stuff about that one] […] The theoretical analysis of the efficient degree of consumer cost-sharing has focused on the trade-off between the gain from more complete insurance against the associated inefficiency of over-utilization, but in practice, the appropriate degree of cost-sharing should also depend on certain other factors, in particular, on the relative costs of administering plans with different degrees of cost-sharing. […] Patient cost-sharing as a means of controlling health services utilization and aggregate health care costs is an example of what in the health economics literature is called “demand-side incentives” (that is, incentives that affect the patients who use health services). A prominent theme in the health economics literature in recent years has been that services utilization and total health care spending in a given population also depend strongly on the incentives of the providers of health services who treat the patients and advise them on what services they should utilize (“supply-side incentives”). If utilization can be effectively controlled through supply-side incentives, the case for high user fees is less strong”

“In comparing the equity and efficiency properties of the social insurance model of funding health care with the general-revenue financing model, the first point that should be made is that, for those populations for which membership in the public plan is compulsory (which may be the entire population), the contributions that the insured are required to pay toward funding the plan […] are equivalent to a tax. […] This equivalence has two important consequences. First, it means that the equity and efficiency properties of the social insurance system can only meaningfully be analyzed as part of the overall system of raising government revenue for all purposes: As previously argued, it is not meaningful to separately analyze the equity and efficiency properties of the revenue raised for some particular purpose. In this sense, therefore, social insurance funding of health care involves the same issues as those arising when funding is from general revenue. Second, once it is recognized that the contributions paid into the social insurance system is only one of many sources of government revenue, it becomes clear that it is not in general efficient to match the revenues raised from this source with a particular kind of spending (health care). If one wants to explain why many countries still try, at least to some extent, to match health care expenditures under their public plans to specific types of revenue (such as social insurance contributions), one must appeal to other factors […], not economic efficiency or equity.”

Health care cost growth:

“The dominant factor contributing to rising spending is the development and diffusion of new medical technology […] The conclusion that technology is a primary driver of cost growth is based on a wide body of literature […] Alarm over health care cost growth is typically centered on the rise in health care expenditures at the population level. Expenditures reflect both unit costs (prices) and utilization patterns (quantities). Some interventions may reduce unit prices, but, because of the utilization response, may not reduce expenditures. […] This helps explain why innovative technology often raises expenditures in the health care sector, even though it is perceived to lower cost in other industries. For example, as technology reduced unit cost in the information technology sector, spending growth in the overall sector increased 26 percent annually from 1982 to 1996 (Haimowitz 1997). Expenditures are also not limited to any particular disease. Individuals cured of one disease inevitably get another. It is possible that reductions in expenditures on one disease may increase overall spending if competing conditions are more expensive. Finally, cost growth at the population level may not reflect trends in cost growth for particular services. Efforts to constrain spending in one area may simply generate greater spending in other areas. For example, in the United States, as inpatient spending growth slowed following implementation of prospective payment systems (PPS), outpatient spending soared (Miller and Sulvetta 1992).”

“In assessing cost containment strategies it is crucial to distinguish between those interventions that affect the trajectory of cost growth versus those that affect the level. […] This distinction is important in assessing the ability of systems which are more conservative in their adoption of new technology to control cost growth. A system that adopts new technology more slowly than another system may have the same rate of cost growth if the baseline level of costs is lower. For example, if a given country has a base spending rate that is 20 percent below that of another country, it will experience the same cost growth if it utilizes a new technology 20 percent less frequently.”

“decreased utilization associated with cost sharing does not disproportionately impact necessary care, as proponents of cost sharing would hope and standard economic theory would predict. Patients apparently reduce use of appropriate and inappropriate care in similar proportions […] Consistent with this view, many recent studies suggest patients reduce use of prescription drugs when faced with modestly higher copayments […] cost sharing has been demonstrated to have disproportionately negative effects on the quality and delivery of health care among low-income populations […] Adverse events, lower adherence, and decreased management of illness are associated with increased patient cost sharing […] the longer term consequences on health associated with lower utilization of high value services have yet to be fully evaluated. […] Because cost sharing is associated with lower costs, many health care payers view cost sharing as a means to reduce growth in health care (Chernew 2004). Yet there is virtually no evidence examining the impact of cost sharing on cost growth. It is possible higher cost sharing lowers spending, but does not alter the trajectory of spending growth. […] Although the debate about the relationship between physician and hospital supply and spending and costs will continue, it is important to note that much of this literature is related to the level of costs, not the trajectory. The limited evidence on cost growth suggests that even in the most successful settings […] the share of GDP devoted to health care still rises, albeit at a somewhat slower rate than in other markets.”

“Many observers have noted that the health care expenditures of individuals with chronic disease are much greater than expenditures of individuals without such disease […] The share of obese Medicare beneficiaries increased from 9.4 percent in 1987 to 22.5 percent in 2002 […] For this reason, some believe that initiatives aimed at improving health will save money. […] [However] most preventive services are not cost saving from a societal perspective. […] In general […] evidence of […] savings associated with disease management and pay for performance is weak. […] it is likely too optimistic to assume that better health will substantially lower the trajectory of health care spending. Health care costs were growing rapidly well before the epidemic of obesity and health care cost growth among the healthy persists. […] Because healthier beneficiaries live longer, and may demand a range of quality of life improving services, it would not be prudent to assume that better health, as desirable as it is, will substantially slow cost growth.”

February 6, 2014 Posted by | books, health, health care, medicine | Leave a comment

The Oxford Handbook of Health Economics

I don’t know in how much detail I’ll cover this book in the week to come but there’s some interesting stuff in there and I figure I might as well start ‘work-blogging’ a bit again. It is my goal to read at least 100 pages/day in this book over the next week’s time, meaning I should be finished at the end of next week (the book has 937 pages, excluding the index at the end – I’ve read 165 pages so far). 100 pages per day isn’t actually all that much and I may decide to work more than that, but the book is occasionally a little technical and I set that goal also because I knew it would be achievable (plus it’s not the only work I’ll be doing). I’ll probably cover plenty of stuff here which will end up not being directly relevant to my work, but I don’t think any of you will object strongly to this – almost none of the stuff I cover here on the blog has anything to do with my university activities anyway. I’m in the very rare situation that stuff which I might have read anyway because I find it interesting actually turns out to be relevant to my work/studies.

I don’t know how difficult the material covered in this book will be to understand for people without a background in economics; it’s much easier to make correct assumptions about such things if you’re covering stuff which is technically not within ‘your field’ either (e.g. psychology textbooks). Fortunately unlike Mas-Colell this is not a (not very well…) disguised math textbook, so I think it actually makes sense to cover some of the stuff from the book here on the blog – it made pretty much no sense to cover the stuff in Mas-Colell which was a big part of why I didn’t do it. I don’t know to which extent it’ll be necessary to add links and comments etc. along the way to aid understanding and I will probably underestimate the need – if you don’t get what they’re talking about even though I seem to be assuming that you ought to get it, just ask questions in the comments and I’ll try to clarify. I have done work on some of the topics covered in the book before, so it’s not like I don’t know anything about these things. Only one chapter so far has been what I’d term ‘model-heavy’, and although I probably won’t talk much about that one there should still be plenty of stuff in the book which it makes sense to cover here. I should note that I’m very aware of the fact that time spent covering the book is time not spent reading, and although I may derive a small benefit from covering the material here I won’t let blogging interfere with the reading goal. So I don’t know how much I’ll blog in the week to come – we’ll see.

Below some observations from the first 7 chapters/160 pages:

“Health care spending has been capturing a growing share of the GDP in all OECD countries between 1970 and 2008. In the median OECD country, health care spending was 5.1 percent of GDP in 1970 and increased to 9.1 percent by 2008 […] A persistent outlier in health care spending has been the US, where health care spending increased from 7.0 percent of the GDP in 1970 to 16.0 percent in 2008. […] Denmark was also an early outlier, actually spending a greater percentage of its GDP on health than the US (7.9%) in 1970. By 2008, Denmark had similar health care spending levels as the median OECD country, a result of a very slow rate of growth in health care spending between 1970 and 2008.” [Actually one might argue that this last part shouldn’t be news to ‘regular readers’ as I’ve covered the Danish numbers before here on the blog in a post in Danish. Note that the data in that post seem to indicate that there may have been a structural shift around the year 2000 or so – the cost share has risen relatively fast since that time, compared to earlier. Denmark is above 11% of GDP now.]

“Growth rates in health care spending are […] compared using the average annual growth rates of health care spending adjusted for inflation and population growth. Using this measure, the average annual rate of health care spending growth was 3.9 percent in the median OECD country from 1970 to 2008 […] The rate of health care spending growth was higher than inflation in every OECD country during the overall time period. […] In 2008, approximately three-quarters of the health spending was from public funds in the median OECD country, while the remaining quarter was from private funds. […] Of the private health care dollars in the median OECD country, 72 percent were out-of-pocket expenses in 2008. Private health insurance is responsible for a small proportion of health spending in most OECD countries. […] Across the OECD, about 15 percent of all tax revenue is devoted to health care –  a proportion that is steadily increasing.”

“Three sectors of health care represents over half of the total health care spending in most OECD countries: inpatient hospital care, outpatient medical services, and pharmaceuticals. […] Inpatient hospital spending declined rapidly during the period from a median of 48.5 percent in 1970 to 32.3 percent in 2008. […] During the same time period, the length of stay for inpatient care has fallen by approximately 58 percent. […] The median OECD country has seen a slight increase in the share of pharmaceutical expenditures from 17.5 percent […] to 13.8 percent [sic; I’m sure those numbers were mixed up during editing.]  […] Real GDP per capita increased by 120 percent from 1970 to 2008 in the median country of the OECD. […] health spending per capita increased by 314 percent […] after controlling for inflation and population growth […] Health spending grew an average of 1.8 percent per year faster than GDP in the median OECD country […] The percent of the population over the age 65 [increased] 40 percent in the median OECD country. […] Life expectancy increased by 8.8 years […] while fertility rates declined by 31 percent. […] Every OECD country had an increase in the number of physicians per 1000 capita between 1970 and 2008 […] The median OECD country had a 223 percent increase.” […but note also that: “on the whole, healthcare wages does not drive health spending growth”]

“Chronic diseases is creating a growing burden on health care spending. […] within the US, 85 percent of health spending was attributed to people with chronic diseases in 2006 (Anderson 2007).”

“Low and middle-income countries (LMICs) […] account for 84 percent of the world’s population, 90 percent of the world’s disease burden […], 24 percent of the world’s GDP, and only 13 percent of global health expenditure. […] The lower the country income level, the higher tends to be the share of out-of-pocket payments […] and the lower the share of revenues (e.g. tax, insurance premiums) which flows through financing agents. […] Even in a country like India, 83 percent of total expenditures on health is from private sources, and of this 94 percent is from out-of-pocket payments. […] [there is] only 0.3 physicians and one nurse per 1000 people in low-income countries. SSA has the lowest density of physicians (one doctor for every 5.000 people), and South Asia of nurses (one nurse for every 1430 people).” [I was very surprised when looking at the data in this chapter; everybody knows that South-Saharan Africa sucks, but South Asia do almost as badly on a lot of health metrics – and on some of them they do even worse than SSA.] […] “Within overall low coverage levels, there are considerable within-country inequalities by socioeconomic group […] In low-income countries, children from the highest wealth quintile have double the measles immunization coverage of the lowest wealth quintile, and there is a seven-fold difference between highest and lowest wealth quintiles in presence of a skilled birth attendant at birth” [note that this latter difference likely translates into a lot of dead babies; infant mortality rates are high these places.]

“There is some evidence from developing countries to suggest that while the public share of revenue may increase as countries grow richer, the public share of [health care] provision shrinks.”

“In low-income countries […] resource limitations make it difficult to provide universally even a limited package of high priority interventions […] The commonly recommended solution is to target resources to the poorest, but there is little evidence so far that such targeting can be done effectively, or that it is cost-effective relative to broader approaches”

“In sum, education is strongly related to health, with both reserve causality and direct effects. However, the extent to which the correlation between education and health reflects direct causality, reverse causality, or omitted factors is not known. Although mechanisms by which health affect educational attainment are well-understood, how education affects health is not. […] it seems unlikely that any one mechanism alone can explain the effect of education on health.” [I wrote a review paper on this topic a while back, coming to some roughly similar conclusions. I won’t cover this stuff in detail here because it’d just be review, but if you want to know more you’re welcome to ask. I incidentally think I may have written about this topic on the blog previously, but I’m not certain how detailed my coverage was.]

“Both income and wealth have strong independent correlations with health, net of education and other measures of SES. Assessing causality is difficult, however. Income and wealth improve access to health inputs (such as medical care and food), but health improves one’s ability to participate in the labour market and earn a decent wage. Illness also raises health care spending, thus reducing wealth […do note here that: “onset of a new illness reduces household wealth by far more than the household’s out-of-pocket health expenditures […] A large share of this reduction in wealth is attributable to a decline in labor earnings.]. Additionally, “third factors” – such as education – may determine both financial ressources and health status. Despite these caveats, many public health researchers have attributed the health-income gradient to a causal effect running from income to health. Some have even gone as far as labeling income “one of the most profound influences on mortality” (Wilkinson 1990: 412). Initial research seemed to support this view – in one such study, McDonough et al. (1997) estimated that a move from a household income of $20.000-$30.000 to a household income greater than $70.000 (in 1993 dollars) was associated with a halving of the odds of adult mortality. It was difficult to fathom that an association so large could be entirely due to omitted variables or reverse causality. However, more recent studies suggest that the direction of causality is far from clear and, furthermore, that it varies considerably by age. Among adults, the negative impact of poor health on income and wealth appears to account for a sizeable part of the correlation between financial ressources and health. […] Careful studies that look for the effect of income on health find little evidence to support this causal link in samples of older individuals in developed countries. […] a preponderance of evidence suggests that in developed countries today, income does not have a large causal effect on adult health, whereas adult health has a large effect on adult income. […] In the last two decades, economists’ most substantial contributions to this literature have involved untangling causal mechanisms.” [This is not news to me as I’ve done some work on this subject during my studies. But it seems like this often surprises people who don’t know what (at least some) economists/econometricians do.]

February 2, 2014 Posted by | books, economics, health | Leave a comment

Antibiotic Policies: Controlling Hospital Acquired Infection

I’ve read about and blogged this topic before, but this is the first academic text on the topic I’ve read. I liked the book and gave it four stars on goodreads. It’s a typical Springer publication, i.e. it’s a collection of relevant studies/papers published on the topic; there are fourteen papers/chapters included in the book. Given the nature of the book there’s some overlap across chapters, but that’s to be expected and it doesn’t really matter much. The book was published in 2011 so it’s reasonably up to date even though things are happening fast in this area.

Some of the authors of the studies included in the book assume that the reader possesses a level of knowledge about microbiology which goes way beyond what you’d get from reading an intro text like Hardy, and although I’ve also previously browsed one of the books you’d actually need to have read in order to understand the details (Brooks, Butel & Morse), I’ve of course long ago forgotten much of that stuff and so occasionally felt a bit lost while reading the book. There’s some good stuff in there though, and many of the chapters I did not find that hard to read although some details eluded me. It’s my impression that you probably will not get much out of the book if you’ve never read a microbiology text before (I actually feel a bit sad having to write that as the topics covered are very important in terms of future public health, and so in a way I’d really wish as many people as possible actually read this book, or at the very least familiarized themselves in some other way with the problems covered in the book).

“This book serves a twin purpose in helping to construct a more informed evidence base for coherent policy making while, at the same time, providing practical advice for health professionals in the prevention and control of HAIs.”

The quote above is from the preface of the book. The papers included in the book cover a wide variety of topics; one chapter deals with the ‘total scale’ of the problem of healthcare associated infections (HAIs), another chapter deals with (among other things) how antibiotic treatment regimes and the development of resistant strains in the community and/or health care institutions are associated, one chapter deals with the epidemiology of drug resistant strains of bacteria and how to properly categorize drug resistance (which can take on many forms), and quite a few chapters focus on specific HAIs (C. difficile, MRSA, VRE, ESBL-producing bacteria, CRE, Acinetobacter baumannii, and MDR (multi-drug resistant) Pseudomonas all get a chapter each). Many intervention studies are covered and the focus is not just on identifying the extent of the problem but also on finding ways to counter the problems; one chapter deals specifically with antibiotic stewardship, which is one of the main ways to try to stop the spread of antibiotic resistance, but many other chapters cover that topic as well in the specific setting. Another key strategic element in any intervention strategy, infection control measures (hand hygiene, patient isolation, etc.), is likewise covered in many of the chapters, and as the studies included have a very ‘evidence-based medicine approach’ to these matters important but potentially embarrasing problems like compliance problems on the part of health care providers [it’s harder to convince doctors to wash their hands than it is to convince nurses..] are not overlooked. The book is not US-centric; countless international studies are included, and a specific chapter is reserved to dealing with MDR infections in low-resource health care settings. The institutional setting is important and is covered in a few chapters, and included in that discussion are observations related to how things like reimbursement methodology may impact health care provider behaviours and how faulty incentive structures on the institutional level may aggravate the problems with resistance development e.g. by failing to address collective action problems in this area.

As might be inferred from the comments above, there’s way too much stuff in there for it to make sense for me to cover it all here. However I have added some observations from the book below, emphasizing some important points and observations along the way and adding a few comments here and there.

“What is required is tackling of the problem at its root cause, namely the gross over use of antibiotics.” […let’s just start out with that one, so that people will not falsely assume that this aspect is not covered in the book.]

“In broad terms, there are two means by which patients can develop multi-resistant infections—they can either develop their own resistant pathogen, or they can acquire someone else’s strain.
Emergence of new resistant pathogens is directly related to antimicrobial selection pressure either via the mutation of new resistance genes or the alteration of bacterial ecology (e.g. in the gut) that facilitates the transfer of naturally occurring or emergent resistance genes from one bacterial class to another […] antibiotic use in food production can have the same effect as direct human antibiotic misuse, since it can select for both resistant pathogens (e.g. fluoroquinolone-resistant Campylobacter in chicken meat) or resistance genes such that food consumption results in either direct fecal colonisation or acquisition of resistance genes by routine gut flora [3,4]. Antibiotic stewardship is therefore not simply a hospital issue.” […]

“The global burden of healthcare associated infections (HAI) is currently unknown, despite international efforts to fill this gap in our knowledge. Where the size of the burden of HAI has been quantified, the greatest impact is in those countries with least resources to measure and manage them. […] 3.5–10.5% of hospitalised patients in industrialised countries may experience HAI (E.C.D.C. 2008), while greater than 25% of hospitalised patients in developing world nations may be affected (W.H.O. 2005). […] While in 2000, 70 countries did not screen donated blood for HIV, hepatitis B or hepatitis C, currently the risk of bacterial infection from transfusion is greater than the risk of acquiring these viruses. Reuse of contaminated needles or syringes during injections in limited resource settings poses a major threat for transmission of infection, accounting for an estimated 21 million hepatitis B infections, 2 million hepatitis C infections and over 95,000 HIV infections. […] Of the 8.8 million deaths in children under the age of 5 years, infectious diseases account for 5.5 million (63%) […] clinicians in developing countries tend to diagnose and prescribe medication empirically. People with undetected resistance then receive antibiotics to which their isolate is not susceptible. For example, one study in western Kenya found that more than half of the patients treated empirically for bacterial diarrhea were given ineffective antibiotics. Among patients with shigella, this number exceeded 80% (Shapiro et al. 2001). […] In developing countries, antibiotics are a scarce resource, and most clinics and hospitals can barely afford common first-line agents, much less second and third-line alternatives […] variation in prices of antibiotics is considerable. The wholesale price differential between amoxicillin and co-amoxiclav, for example, is on the order of a factor of 20 (Forster 2010). This means that where resistant bacteria necessitate the use of co-amoxiclav, only 5% of the patients can be treated for the same budget as with amoxicillin. […] In coastal Kenya, resistance to chloramphenicol, amoxicillin, cotrimoxazole, and gentamicin in Gram-negative sepsis is common, and susceptibility remains only to two rarely used drugs, ciprofloxacin and cefotaxime. The cost of treating a 15 kg child with sepsis would be $0.38–2.30 for gentamicin and chloramphenicol versus $73–108 for the effective drugs […] In Thailand, only 9% of antibiotics administered in a teaching hospital were appropriate to the patient’s condition, and 36% of patients were given antibiotics without evidence of an infection […]

“The underused vaccines that could have the biggest effect on antibiotic use in hospitals are against Streptococcus pneumoniae and Haemophilus influenzae type b. To these should be added one of the new vaccines against Rotavirus, the main cause of dehydrating diarrhea, which kills 400,000–500,000 infants and children in developing countries annually. Even though Rotavirus is, in fact, a virus, reducing its incidence will reduce antibiotic use. The most appropriate treatment for rotavirus and other causes of watery diarrhea is oral rehydration therapy, but since antibiotics are used inappropriately in many cases, reducing the number of cases will reduce antibiotic use.” [..vaccines against viruses may help decrease the number of bacteria resistant to antibioticsyep, this stuff is complicated..]

“HAI are recognised as among the most common adverse outcomes from hospitalisation in the US; approximately 1.7 million HAI are reported across the US each year, which are associated with around 99,000 deaths per year. Around a third of HAI are urinary tract infections, one fifth are surgical site infections, 15% are pneumonia and 14% are bloodstream infections (C.D.C. 2010).” […] Estimates in Europe are that approximately 4.1 million patients per year experience HAI, and that attributable deaths are of the order of 37,000 per year (E.C.D.C. 2005–2010).” [These estimates are somewhat uncertain and I’m not sure how much you should read into the fact that they differ in the way that they do, with fewer but more lethal HAIs in the US. Before you read a lot into it, you should certainly note that there is huge regional variation in the data here.] […]

“Surgical prophylaxis is a common area of overuse as shown in many publications. Measured by total DDDs [defined daily doses], it can amount to around one third of a hospital’s total antibiotic use. This illustrates the potential for ecological damage although surgeons often ask whether 24 h or even single dose prophylaxis can really select for resistance. The simple answer is yes, but of course much of the problem is extension of prophylaxis beyond the perioperative period, often for several days in critical patients, perhaps until all lines and drains are removed. There is no evidence base in favour of such practices.” […]

“Since 2002, increasing rates of CDI [Clostridium difficile Infection] with a more severe course, higher mortality (from 4.7 to 13.8%) and more complications (from 7.1 to 18.2%) have been reported in Canada […] Of all patients who develop CDI in the hospital setting, approximately 80–90% have used antibiotics in the previous 3 months. […] MRSA can survive for months in hospital environment […] and it can be isolated on clinical equipment, as well as on general surfaces especially close to patient’s area, such as curtains, beds, lockers and over-bed tables […] Before contact precautions are implemented, MRSA carriers may have already contaminated their environment with MRSA. […] Cross-transmission between patients may occur via HCWs [health care workers’] hands after touching contaminated environmental surfaces […] One study showed that 10% of HCWs fingertips were contaminated with MRSA after contact with MRSA positive patient’s environment […] There is now reasonable evidence that rates of MRSA, C.difficile, VRE and multi resistant Gram-negatives can be reversed by modulating use of key agents such as cephalosporins and quinolones […] The real problem for the future, of course, is how to do this without “squeezing the balloon”, transferring the resistance selection pressure to other classes of agents. This highlights another paradox, that of current antibiotic policies which tend to lead to a lack of diversity of use of different classes of antibiotics. Diversity of use is probably one of the best strategies to delay emergence of resistance, although a lack of choice of truly different drug classes makes its implementation problematic. Moreover, the holy grail, and the most difficult thing is to achieve total reduction in prescribing while not compromising patient outcomes. Again, this isn’t something current strategies are good at achieving.” […]

“ESBL-producing bacteria are not only present in hospitals from endemic nosocomial sources but are introduced into the hospital from other health care facilities (particularly high rates occur in care of the elderly homes […] but also from individuals coming
from the community (Ben-Ami et al. 2006). […] This community carriage is an important facet of ESBL control [Again, what happens outside the hospitals matter a great deal…] […]

“Carbapenems have the broadest antimicrobial spectrum of any beta-lactam antibiotic and are frequently used as first-line agents for the treatment of severe infections caused by multiresistant Gram-negative bacteria […] The emergence and spread of carbapenem-resistant Enterobacteriaceae (CRE) are therefore a major concern for patient safety and public health. Infections due to CRE may lead to increased likelihood of treatment failure and growing reliance on third-line agents and combination therapy, with doubtful therapeutic efficacy and increased potential for toxic side-effects […] It also increases the cost of treatment […] CRE differ from most other multidrug-resistant bacterial pathogens in that there is no reliable treatment available (Schwaber and Carmeli 2008). […] two cases of panresistant CRE were recently reported from a hospital in New York […panresistant strains are basically untreatable, US.] […] Patients with CRE infection are at high risk of treatment failure and adverse outcomes, including increased mortality and morbidity, longer length of hospital stay, and higher treatment costs when compared to infections caused by susceptible strains. Several studies have reported high percentages of crude in-hospital mortality— some over 50%—among patients infected with CRE […] the magnitude of the excess mortality directly attributable to CRE is difficult to quantify […] Overall, uncertainties persist in individual patient-level analyses regarding which prior antibiotic exposures are most important as risk factors for acquisition of, transmission of and infection with CRE. Similarly, ecologic studies using aggregate datalevel analyses do not show a clear-cut picture.” […]

“Antibiotic policies are crucial but they cannot be effective without active infection control program[s]. A hospital with a strong infection control program without an antibiotic stewardship component would tackle transmission of multi-resistant organisms such as VRE but would not prevent individual patients from getting colonised or infected with resistant microbes. On the other hand, strong antibiotic stewardship would be expected to control the menace of multi-resistant organism but in absence of an infection control program, transmission of organisms (even if not multiply resistant) would be easy and would adversely affect patient care.” […]

“A retrospective, risk-adjusted, cohort study of 80 patients with Acinetobacter bacteraemia conducted in Korea demonstrated that those infected with imipenemresistant strains had a significantly higher 30-day cumulative mortality rate than those infected with imipenem-susceptible strains (57.5% versus 27.5%) […] This was mainly due to a higher rate of inappropriate antimicrobial therapy. […] Carbapenems are the mainstay of treatment for severe infections. However, carbapenem-resistant A. baumannii strains have emerged worldwide. […] A considerable proportion of multi-drug resistant A. baumannii strains are susceptible only to polymyxins, which prompted the use of an old antibiotic in recent years. […] Polymyxins are polypeptide antibiotics that act as detergents on the bacterial cell wall. They were introduced in 1940 but they were abandoned in the 1980s due to the occurrence of nephrotoxicity and neurotoxicity. […] Reported nephrotoxicity ranges between 8 and 36%. […] Reported neurotoxicity ranges between 7 and 29%, with oral and perioral paresthesias, visual disturbances and polyneuropathy […] [So basically what has happened is that doctors have been forced to restart using drugs they threw away 30 years ago because those drugs caused kidney failure and severe nerve damage. These old drugs are currently the only drugs that work against some MDR infections, and no new drugs are even close to being developed at this point]. […]

P. aeruginosa is the second most common cause of health-care associated pneumonia, of hospital-acquired pneumonia and of ventilator-associated pneumonia (VAP). It is also reported as the cause of 9% of hospital-acquired urinary tract infections (UTIs). […] It is estimated that the rate of colonization and/or infection by MDR P. aeruginosa is 0.5 episodes/1,000 patient-days in the general ward and 29.9 to 36.7/100 patients in the ICU (Agodi et al. 2007; Peňa et al. 2009). […] infections by MDR P. aeruginosa have a significant impact on mortality. A retrospective study of our group in non-neutropenic hosts in the general ward disclosed 22.2% mortality of infections by MDR P. aeruginosa compared to 0% of infections by susceptible isolates […]. For ICU infections caused by MDR P. aeruginosa mortality ranges between 22% and 77%; this ranges between 12% and 23% when ICU infections are caused by susceptible isolates (Shorr 2009).” […]

Antibiotic effectiveness can be viewed as a shared resource in which current use depletes future value and imposes costs on society in the form of longer hospitalization, higher mortality rates, and the diversion of resources into the provision of newer and more expensive drugs. In making treatment decisions, prescribers should weigh the favorable effects of applying antibiotics to improve a patient’s health against the negative consequences for the public and future drug effectiveness (Laxminarayan 2003b). However, clinicians usually ignore the future therapeutic risks to society associated with antibiotic use and instead focus on the direct benefits of antibiotic treatment to their patients. […] In the absence of a good pipeline of new drugs, it is the balance between the individual patient and society as a whole, otherwise known as the ecological perspective, that has to be clearly established and debated. We need to get clever, quickly. […] In the long term, new antibiotics are needed […] However, as a gap of 10–15 years has been identified (European Centre for Disease Prevention and Control and European Medicines Agency 2009), immediate action is needed to conserve the power of the available arsenal.”

November 27, 2013 Posted by | books, health, health care, medicine | 13 Comments

Health Online 2013 (Pew)

“Thirty-five percent of U.S. adults say that at one time or another they have gone online specifically to try to figure out what medical condition they or someone else might have.

These findings come from a national survey by the Pew Research Center’s Internet & American Life Project. Throughout this report, we call those who searched for answers on the internet “online diagnosers”.

When asked if the information found online led them to think they needed the attention of a medical professional, 46% of online diagnosers say that was the case. Thirty-eight percent of online diagnosers say it was something they could take care of at home and 11% say it was both or in-between.

When we asked respondents about the accuracy of their initial diagnosis, they reported:
41% of online diagnosers say a medical professional confirmed their diagnosis. An additional 2% say a medical professional partially confirmed it.
35% say they did not visit a clinician to get a professional opinion.
18% say they consulted a medical professional and the clinician either did not agree or offered a different opinion about the condition.
1% say their conversation with a clinician was inconclusive.

Women are more likely than men to go online to figure out a possible diagnosis. Other groups that have a high likelihood of doing so include younger people, white adults, those who live in households earning $75,000 or more, and those with a college degree or advanced degrees.”

The quotes above are from a Pew report, Health Online 2013, published earlier this year. Below I’ve added some more data from the report, as well as a few comments. You can click the tables to view them in a higher resolution.

“Looking more broadly at the online landscape, 72% of internet users say they looked online for health information of one kind or another within the past year. […] 77% of online health seekers say they began at a search engine such as Google, Bing, or Yahoo. Another 13% say they began at a site that specializes in health information, like WebMD. Just 2% say they started their research at a more general site like Wikipedia […] 39% of online health seekers say they looked for information related to their own situation. Another 39% say they looked for information related to someone else’s health or medical situation. […] As of September 2012, 81% of U.S. adults use the internet and, of those, 72% say they have looked online for health information in the past year. [Incidentally, according to this Pew report, the number of online Americans is actually 85%, but it’s in that neighbourhood… Note that 72% of 81% is just 58% (they say 59% in the report later, probably due to rounding) – so almost half of all Americans don’t look for health information online. That’s a lot of people.] […]

Online diagnosis and follow-up

Females are more likely to be online diagnosers, as are young people, whites, rich people, and college-educated individuals (when we compare the females with males, the young people with the old, the white people with the non-white, etc. See also the remarks in the update..). Note that education is basically a step-function here; the more education you get, all else equal the more likely you are to try to diagnose yourself online. Note also that some of these differences are really huge; roughly 10 percent of people without a HS diploma answered that they’d looked online to diagnose a condition during the last year, whereas half of all college-educated individuals answered in the affirmative.

Insured vs uninsured

A potentially important thing to have in mind when comparing the numbers for insured and uninsured individuals is that internet usage and health insurance status probably covary; I believe it’s likely that uninsured people are also less likely to use the internet. Low-income individuals with short educations are much less likely to be online, independent of age (see the link above).

“Twenty-six percent of internet users who look online for health information say they have been asked to pay for access to something they wanted to see online. […] Of those who have been asked to pay, just 2% say they did so. [I was very surprised that that number was strictly larger than zero…] Fully 83% of those who hit a pay wall say they tried to find the same information somewhere else. Thirteen percent of those who hit a pay wall say they just gave up. […] Respondents living in lower-income households were significantly more likely than their wealthier counterparts to say they gave up at that point. Wealthier respondents were the likeliest group to say they tried to find the same information elsewhere.”

Educ in detail

Do remember when looking at the numbers above that health status and education are related variables; lower educated people are more likely to be in poorer health than are higher educated people on average, in part because of lifestyle choices (I’ve written about these differences before – see e.g. this post (and note that there’s a lot of stuff in those links – and that I have a lot more links for you if you don’t find them satisfactory, as I’ve done academic work in this field and am quite familiar with the literature on the links between education and health.)). Yet even when conditioning on online status (low-educated individuals are less likely to be online), individuals with low educations are still, all other things being equal, much less likely than are the college educated to look online for many types of health information.

Update: To illustrate how much trouble you might get into if you don’t have in mind the differences in internet adoption rates across social strata, I decided to add a few more numbers. The numbers are from the Offline Adults report, to which I also link above:

Offline demographics

People without a high school diploma are roughly 10 times as likely not to use the internet as are people with a college degree; 41% of people without a HS diploma don’t use the internet – 4% of college-educated don’t. For individuals with an income below $30k, one in four don’t use the internet, whereas roughly 5% of those with an income north of $50k don’t. It’s very safe to say that not all subgroups included in some of the specific types of response data above are equally representative of the groups from which they are derived. Note also that potential drivers of the relevant intragroup differences here may be very important if one were to try to find ways to ‘bridge the information gap’; for example if some of the low-educated individuals who don’t use the internet can’t read, finding ways to provide them with internet access may not make much difference.

I should point out here that based just on the observations above it’s impossible to say anything about the details of what drives these results. It’s not clear e.g. how big a role the age variable plays when it comes to the contribution from income and education; old people on a pension have much lower incomes (but higher net savings) than most people who’re still active in the labour market (link), and older people are also significantly less likely to have college degrees and more likely to not have a high school diploma. The significance tests they report which are meant to indicate whether or not e.g. the results for people with an income of $30-50k are different from the results for people with incomes below $30k don’t take stuff like that into account, they’re just of a ‘let’s ignore everything else and compare the numbers’-kind and so can’t really be trusted. Maybe income doesn’t matter once you’ve taken age and education into account. I’m not saying this is the case, but given the data you can’t say if that’s true or not. Disentangling the ‘pure partial effects’ would be nice, but that’s likely to be a lot harder than it looks; multicollinearity is likely a problem, and some of the correlated regressors display non-linear relationships (e.g. income-age – see the link above). Be careful about which conclusions you draw.

October 14, 2013 Posted by | data, demographics, health | Leave a comment

Adipose tissue and cancer (2)

I’ve read roughly two-thirds of the book by now – I like it, pretty much every page contains new stuff which I didn’t know anything about and it’s quite interesting. Some more stuff from the book below, as well as some comments. As always you can click images to view them in a higher resolution.

“Obesity increases the incidence of many cancers, such as breast, prostate, and colon cancer. However, endometrial cancer is the mostly tightly linked with obesity. Estimates suggest that nearly 40 % of cases of endometrial cancer can be attributed to obesity. […] Obese women have a threefold higher risk of developing endometrial cancer than lean women [6]. […] every increase in BMI of 5 kg/m^2 increases a woman’s risk of the developing of endometrial cancer by approximately 60 % (relative risk, 1.59; 95 % confidence interval [CI], 1.50–1.68) [7]. Endometrial cancer in obese women is more likely to have lower risk features such as endometrioid histology and low/intermediate grade. […] An elevated waist-to-hip ratio, reflecting a preferential deposition of adipose in the abdomen, increases the risk of developing endometrial cancer by 220 % [10]. […] Among the population as a whole, obesity increases the risk of death from endometrial cancer. In a study of 900,000 prospectively followed healthy patients, 57,145 individuals died of cancer over 16 years. The relative risk of death from endometrial cancer in this population was 6.25 for women with a BMI >40 and 2.77 with a BMI between 35 and 39 [6].”

“As a component of adipose tissue in obese individuals, immune cells, and specifically macrophages, secrete a variety of growth, survival, and proangiogenic factors, as well as bioactive molecules that enable tumor growth and contribute to the remodeling of the tumor microenvironment to facilitate metastases. Furthermore, reactive oxygen and nitrogen species released by activated macrophages are mutagenic and accelerate oncogenic mutations that contribute to cancer risk and progression [30,33]. So, not only does inflamed visceral adipose tissue provide an ideal milieu for the growth of metastatic endometrial cancer but proinflammatory factors also secreted by infiltrating adipose immune cells mediate systemic effects on tumor progression at distant sites, including the endometrium.”

Cancer prevention
“Taken together, current evidence suggests that through a variety of mechanisms, weight loss and physical activity reduce proproliferative signaling and counteract environmental conditions that support the initiation and progression of endometrial cancer.” […as the figure above illustrates, endometrial cancer is far from the only cancer type where behavioral factors play a large role – US.]

“Multiple epidemiologic studies demonstrate that women who use combination estrogen and progesterone oral contraceptives (OCP) decrease their risk of endometrial cancer by 50 % [78–80]. While there is no data to support a decreased efficacy in endometrial cancer protection in obese women, there are studies that suggest that obese women have a slightly decreased contraceptive efficacy compared to thin women [81].”

Relative cancer risk by BMI
“At the cellular level, overweight and obesity are characterized by the increase in number and size of adipocytes. A lean adult has 35 million adipocytes, each containing 0.4–0.6 μg of triglycerides, whereas an extremely obese person has 125 million adipocytes, each containing 0.8–1.2 μg of triglycerides [27]. Traditionally, adipocytes have been viewed solely as energy depots, but after the discovery of leptin in 1994 and extensive research in the field in the last decades, it has been established that the adipose tissue is an active endocrine organ. The adipocyte is a major source of secreted proteins …”

A really important point which has been repeated, explicitly or implicitly, again and again in this book, and which I thought I should emphasize here using ‘non-textbook language’, is that fat cells aren’t just inactive cells that ‘hang around’ doing nothing. They do a lot of stuff while they’re ‘hanging around’. And when you have a lot of them hanging around in the wrong places, many of the things they’re doing are really quite bad for you. As you’ve probably already inferred, the book goes into a lot more detail about mechanisms and how these things work in detail (to the extent that we even know what’s going on in the first place), but if you don’t remember much from the posts about this book this is at least, I think, one of the key points you should try to remember; adipose tissues are active tissues and they – and the secretions derived from them – play a major role in a variety of contexts, including some contexts which are highly relevant to e.g. cancer pathogenesis. There’s still a lot we don’t know because this stuff is complicated; I link to leptin above, which has been intensively studied and is also relatively intensively covered e.g. in chapter 5 of the book, and the wiki link about adipokines mentions a few others – but I should note here that there are more than 50 different types of adipokines that we know of at this point. Different types of cancer start out in different types of tissues and a diverse set of mechanisms are involved in the disease processes, and so it seems likely that different types of adipokines play different roles in different types of cancers. There are still a lot of things which are not clear, but as they put it in the conclusion of chapter 5: “There are strong epidemiological, molecular, and clinical evidences showing associations between adipokines and the incidence and clinical outcome of cancer.” It should be noted that work on this stuff is not limited to work on just ‘human data’ – lab-work using rodents, which is covered in chapter 6 of the book, has added some details and some interesting observations regarding potential mechanisms of action, and such animal models seem to support ‘a causal link’ of some sort between body weight and the development of specific types of cancers in a number of important (…to humans…) cases, including breast cancer and colon cancer. However the precise mechanisms of action are still far from clear, as they note in their conclusion in chapter 6:

“As detailed here, overweight and/or obesity is associated with an elevated risk of several cancers; however, it is clear that a common disease mechanism was not identified. Although the current literature hypothesizes at least three major components such as sex hormones, insulin-related pathologies, and adipokines, these components cannot explain every aspect of clinical features/disease courses. But as models improve both for obesity and various cancers, hopefully it will become easier to identify mechanisms of action for the relationship of body weight and cancer.”

October 4, 2013 Posted by | books, cancer, health, health care, medicine | Leave a comment

Adipose tissue and cancer

I’ve read the first third of this book, and it’s been a quite interesting read so far. Some parts have been easier to read than others and occasionally it gets a bit technical, but overall it’s a quite readable book for someone with my background and I’m certainly learning some new stuff by reading this.

Some observations from the book:

“obesity and metabolic syndrome are linked to various chronic diseases [6,7] including cardiovascular disease, type II diabetes, and the focus of this chapter, cancer. Importantly, not all obese individuals develop the metabolic dysregulation usually associated with obesity and metabolic syndrome, and these “metabolically healthy obese” individuals do not have elevated cancer risk. An estimated 30 % of obese individuals in the USA are metabolically healthy [8]. Conversely, some nonobese individuals can develop the metabolic perturbations usually associated with obesity, and these individuals appear to be more prone to chronic diseases including cancer [9]. Thus, an emerging hypothesis is that the obesity-related metabolic perturbations, and not specific dietary components or increased adiposity, are at the crux of the obesity–cancer connection.” […]

“Evidence-based guidelines for cancer prevention urge maintenance of a lean phenotype [10]. Overall, an estimated 15–20 % of all cancer deaths in the USA are attributable to overweight and obese body types [11]. Obesity is associated with increased mortality from cancer of the prostate and stomach in men; breast (postmenopausal), endometrium, cervix, uterus, and ovaries in women; and kidney (renal cell), colon, esophagus (adenocarcinoma), pancreas, gallbladder, and liver in both genders [11]. While the relationships between metabolic syndrome and specific cancers are less well established, first reports from the Metabolic Syndrome and Cancer Project, a European cohort study of ~580,000 adults, confirm associations between obesity (or BMI) in metabolic syndrome and risks of colorectal, thyroid, and cervical cancer [12].”

“During obesity, adipose tissue responds to the excess energy by increasing adipocyte size (hypertrophy) and enhancing adipocyte proliferation (hyperplasia) [14]. Adipocyte size strongly correlates with insulin resistance and secretion of proinflammatory cytokines [3]. Moreover, location of the adipose tissue also determines risk for metabolic diseases. […] Healthy adipose tissue must be able to rapidly respond to excess energy intake by inducing adipocyte hypertrophy and hyperplasia, remodeling of the extracellular matrix, and enhanced neovascularization to nourish the adipose tissue. In pathological states such as insulin resistance associated with obesity, rapid adipocyte hypertrophy occurs with restricted angiogenesis resulting in cellular hypoxia, and thereby resulting in local inflammation [15]. Macrophages surrounding necrotic adipocytes phagocytize fatty acids, which are released from the adipocyte. This produces bloated, lipid overburdened macrophages, which is characteristic of chronic inflammation and often observed in obese individuals [14]. […] inflammation is a recognized hallmark of cancer, and growing evidence continues to indicate that chronic inflammation is associated with increased cancer risk [75–77]. Several tissue-specific inflammatory lesions are established neoplastic precursors for invasive cancer, including gastritis for gastric cancer, inflammatory bowel disease for colon cancer, and pancreatitis for pancreatic cancer [78,79].”

“When lipid storage capacity in adipose tissue is exceeded, surplus lipids often accumulate within muscle, liver, and pancreatic tissue [16]. As a consequence, hepatic and pancreatic steatosis can develop; both have been positively associated with insulin resistance and ultimately lead to impairment of lipid processing and clearance within these tissues [16]. […] The term nonalcoholic fatty liver disease (NAFLD) refers to a disease spectrum that includes variable degrees of simple steatosis, nonalcoholic steatohepatitis (NASH), and cirrhosis [19,20]. Simple steatosis is benign, whereas NASH is defined by the presence of hepatocyte injury, inflammation, and/or fibrosis, which can lead to cirrhosis, liver failure, and hepatocellular carcinoma. […] NASH occurs in 20 % of cases of NAFLD and ~5–20 % of NASH cases progress to cirrhosis; 80 % of cryptogenic cirrhosis cases present with NASH [22]. Of this group, ~0.5 % will eventually progress to hepatocellular carcinoma […] In Western populations, overnutrition/obesity is the most common cause of NAFLD” […] NAFLD has evolved in parallel to the obesity pandemic as the most prevalent liver disease worldwide. Whereas the fact that chronic liver inflammation as observed in nonalcoholic steatohepatitis (NASH) finally leads to the development of hepatocellular carcinoma is well accepted [123], its association with increased formation of adenomatous polyps and CRC has just recently been established [124,125].”

“Hyperglycemia, a hallmark of metabolic syndrome, is associated with insulin resistance, aberrant glucose metabolism, chronic inflammation, and the production of other metabolic hormones such as IGF-1, leptin, and adiponectin [37]. […] In metabolic syndrome, the amount of bioavailable IGF-1 increases […] Elevated circulating IGF-1 is an established risk factor for many cancer types [38,39].”

VEGF [Vascular Endothelial Growth Factor], a heparin-binding glycoprotein produced by adipocytes and tumor cells, has angiogenic, mitogenic, and vascular permeability-enhancing activities specific for endothelial cells [83]. Circulating levels of VEGF are increased in obese, relative to lean, humans and animals, and increased tumoral expression of VEGF is associated with poor prognosis in several obesity-related cancers [84]. The need for nutrients and oxygen triggers tumor cells to produce VEGF, which leads to the formation of new blood vessels to nourish the rapidly growing tumor and may facilitate the metastatic spread of tumors cells [83].”

“Epidemiological studies indicate that obesity represents a significant risk factor for the development of various cancers such as prostate and breast cancer, leading cancers in the Western world. An impressive body of evidence, however, also indicates that the risk of colorectal adenoma, and cancer (CRC) is increased in subjects with obesity and related metabolic syndrome [2,3]. […] Colorectal cancer is the second leading cancer death in the Western world and its death rate correlates with body mass index [5]. […] Recent CRC screening studies suggest that obesity and an increased body mass index are a significant additional risk factor for the development of colonic polyps with evidence that advanced adenomas arise in men almost a decade earlier than in women [7]. […] menopausal status appears to modify the relationship between BMI and colon cancer with a strong association between BMI and colon cancer risk seen in premenopausal but not postmenopausal women [21]. […] being obese prior to being diagnosed with colon cancer increases your risk of dying from the disease [29–32]. […] more and more studies are now demonstrating the location of body fat tissue is the best predictor of all-cause and colorectal cancer mortality […] colon cancer survival may be less likely for patients who are […] too thin at diagnosis [34].”

“In a meta-analysis of 52 studies (24 case–control and 28 cohort studies) examining the link between physical activity and colon cancer, a significant 24 % reduced risk of colon cancer in people who were most active compared with the least was found [48]. This supports other reviews of the association between physical activity and colon cancer in the Asian and European populations [49,50]. […] Physical activity also appears to affect disease outcome and recurrence after diagnosis and treatment with the greatest effect on colon cancer incidence [53]. […] new well-controlled clinical trials on obesity prevention and obesity treatment are necessary before therapeutic implications of WAT [White Adipose Tissue] reduction on cancer predisposition are completely understood. One of the possibly important considerations is the number of adipocytes and the accompanying stromal/vascular cells in WAT increasing in obesity and remaining increased even upon subsequent weight loss, which occurs via adipocyte size reduction. The pool of ASC [Adipose Stem Cells] is likely to remain intact and could contribute to cancer onset or progression despite calorie restriction and reduced adiposity.”

“There is general agreement that obesity is associated with an increased incidence of breast cancer in postmenopausal women (reviewed in [14–17]). […] The European Prospective Investigation into Cancer and Nutrition (EPIC) study [18], which had 57,923 postmenopausal participants, is of particular interest because of its large size, its prospective design, and the observations made concerning exogenous estrogens as a confounder. The results showed that a long-term weight gain was related to an increase in risk, but only in those who were not taking hormone replacement medication: compared with women with a stable body weight the relative risk for women who gained 15–20 kg was 1.5 with a confidence interval of 1.60–2.13. As reported by others, adiposity ceased to be a risk factor in current replacement therapy users, who were already at a high risk for breast cancer compared with nonusers. […] Preexisting obesity and postoperative weight gain are associated with poor prognosis in both premenopausal and postmenopausal breast cancer patients. […] A pivotal review of the literature by Chlebowski et al. [25] found that in 26 out of 34 studies individual studies, totaling 29,460 women, obesity was related to an increased risk of recurrence or reduced survival.”

“Daling et al. [29] have provided a major contribution to our understanding in the relationships between body fat mass and tumor biomarkers of progression in young breast cancer patients. In their study, not only was a combination of obesity and an absence of ER expression in premenopausal breast cancer patients aged younger than 45 years associated with an increased risk of dying from the disease, but those with BMI values in the highest quartile were more likely to have larger tumors of high histologic grade. This observation is particularly significant because it implies that large tumors in overweight/obese women grow at a faster rate than tumors of similar size from leaner women, rather than simply arising from delayed diagnosis due to palpation difficulty in obese women.”

“Wolf et al. [36] and Schott et al. [37] suggested that up to 16 % of breast cancer patients have diabetes, and that T2D may be associated with a 10–20 % excessive risk of breast cancer. […] There is ample epidemiological evidence that diabetes contributes to breast cancer risk [17,36–40]. […] Overall survival in cancer patients, with or without preexisting diabetes, has shown diabetes to be associated with an increased all-cause mortality risk. […] The Danish Breast Cancer Cooperative Group, with 18,762 newly diagnosed T2D cases, found that the recurrence with metastases was 46 % higher in obese women with a BMI of 30 kg/m^2 or greater beyond the first 5 years.”

The relationship between obesity and prostate cancer is a complicated one. […] The explanation for this confusion may rest, at least in part, in the reports that obesity as a positive risk factor for prostate cancer relates specifically with the aggressive phenotype [56–60] […] a meta-analysis by Discacciati et al. [61] of the results from 25 studies that examined disease stage and BMI showed not only a positive relationship between obesity and advanced prostate cancer but also a decrease in the risk for localized disease. The association between obesity and an aggressive prostate cancer phenotype is reflected in the relationship between the BMI and prostate cancer mortality rate. For example, in one large retrospective cohort study by Andersson et al. [62] […] there was a significantly larger prostate cancer mortality rate in the higher BMI categories”

Two studies have been reported in which meta-analysis was used to examine previously published investigations into the relationship between diabetes mellitus and prostate cancer risk [66,67]. […] [The first] meta-analysis showed that there was an inverse relationship between diabetes and prostate cancer risk, which translated to a 9 % reduction in risk. […] The overall conclusion […in the second meta-analysis] was the same: diabetic men have a significantly decreased risk of developing prostate cancer (RR = 0.84; 95% CI, 0.76–0.93). […] Gong et al. [69] reported a large prospective study of diabetes and prostate cancer from the USA after the two meta-analyses described above had been published that also took account of potential confounding by obesity. Men with diabetes had a 34 % lower risk of prostate cancer compared with men without diabetes that was not affected by adjustment for the BMI […] In contrast to these results, recently published studies have found that the presence of diabetes is positively associated with prostate cancers of high-grade [71–73] and late-stage tumors [72] ], a reversal in the observed relationship that needs to be considered in the context of the duration of the presence of T2D and the detection of prostate cancer by prostatic-specific antigen screening.”

October 1, 2013 Posted by | books, cancer, diabetes, health, health care, medicine | Leave a comment

Gender, Physical Activity, and Aging (2)

I’ve finished the book. I gave it four stars. There’s a lot of research covered here, lots of ground covered, and most of it is well written. When findings are conflicting we are told so, and the reasons for discrepancies across studies are often investigated in some detail. Conclusions drawn are often of a tentative nature, and the uncertainties involved are often emphasized. Given the amount of material covered, detailed coverage of specific studies and their limitations is somewhat limited, but stuff like differences in study design across studies and consequences of these are occasionally emphasized; the lack of longitudinal studies in specific areas of research covered are for example a few times highlighted as a potential problem regarding which causal inferences to draw from the research. Sometimes the controls included in a specific study are mentioned, sometimes they’re not, and that bothered me a little a few times as it was for example unclear in some cases whether a specific study had controlled for initial health status or not; the inclusion of this variable might not deal completely with the endogeneity problem (exercise affects health status, but health status also affects exercise ability – or at the very least perceived exercise ability), but without including it in some specific cases you’re quite likely to get into trouble, and it seems very probably that some older studies have overlooked this problem. In the book the results of quite a few RCTs are covered, which is nice – but there are also a lot of cross-sectional studies, epidemiological studies of various kinds, survey-based stuff… Natural experiments are almost (? can’t remember now if the proper word is ‘completely’…) absent in this work, but then again I guess this book was written before IV estimation really took off, and besides such methods would be most useful in areas which although related to the areas covered in the book are still sufficiently different in scope for it to (arguably) make sense not to include that kind of stuff in the book.

In general the stuff’s not easy to read if you don’t know a lot of stuff about biology, physiology and related fields, and I felt more than a few times that I was in over my head (though some chapters were a lot easier than others; because of my knowledge of diabetes and the physiological consequences of adiposity chapter 10 was an easy read) – but stuff like study methodology should not cause you problems as most of the -designs applied are quite uncomplicated. The stuff included about health impacts of health interventions (e.g. exercise) deals mainly with ‘idealized’ high-compliance settings; they do talk a bit about the adherence-/compliance problems which are observed when doing intervention studies, but more focus on this aspect might have been a good idea – at least I think the book sells the idea that exercise is good for the individual much better than it does the idea that large-scale policy interventions undertaken in order to improve health that way might be a good idea. I believe quite a bit of work has been done in that area since then, so you can find out about that stuff elsewhere.

I haven’t ‘punished’ the authors for not defining all terms applied in the work and for including stuff I was supposed to know but didn’t – it’s not their fault I’m an ignorant fool. But you should have in mind when interpreting my rating that a person like me is borderline too ignorant about stuff like immunology, microbiology, and cardiology to read and understand all the stuff in the book, despite having, among other things, read textbooks dealing with all three areas before. I got a lot of stuff out of this book, but I’d have gotten more out of it if I’d have known more about some of the things they cover.

I have included some observations from the book below. I’ve mostly stayed clear of the technical stuff:

“It has been suggested that declines in functional capacity with age reflect age-related reductions in physical activity. Inactivity has been estimated to account for 50% of the age-related loss in function. […]

The ability to move with purpose and to remain independent with increasing age depends to a large degree on retaining an adequate functional capacity in the neuromuscular system. This system, which governs the generation and control of muscle force, typically undergoes a substantial decline in functional capacity with age […] Beginning at approximately 30 years of age, human muscular strength declines at a rate of 10 to 15% per decade1 […] Because the activities of daily living usually do not require maximal muscular efforts, the gradual loss of strength for most individuals does not become functionally significant until after 55 to 60 years of age. […]

Evidence is accumulating that progressive resistance exercise is an efficacious, nonpharmacological treatment for age-related losses in muscle mass, quality, and function. The benefits of regular physical activity throughout the life span have been a major focus of exercise physiologists for many decades. […] Taken together, these results suggest that, if sustained over a lifetime, regular physical activity has a significant protective effect on muscle strength, function, and oxidative capacity. Benefits appear to relate not only to function, but also to longevity.9,10 […]

Evidence from both animal and human models suggests that there are age-related differences in the susceptibility of skeletal muscle to exercise-induced damage and ability for post-damage repair. One such difference is in the amount of work required to induce muscle damage. […] Human studies have […] tended to demonstrate a greater degree of exercise-induced muscle damage in elderly than in younger adults. […] Most animal and human studies agree that recovery from acute exercise-induced muscle damage is impaired in older subjects. […]

evidence is mounting that estrogen may be a key hormone for maintaining muscle strength in women.96 […] although gender differences in tissue antioxidant potential, partially due to estrogen, may exist, their importance in influencing exercise-induced peroxidative muscle damage is still undefined. […] whether females are afforded greater protection from postexercise muscle inflammatory damage and/or are delayed in muscle healing rates as a consequence of their higher estrogen levels is not yet firmly established. […]

The elderly appear to be more susceptible to exercise-induced muscle damage than younger adults. In addition, following such damage, there seems to be a relative impairment in muscle repair and adaptation in the sedentary elderly. Nevertheless, evidence suggests that the ability of older muscle to adapt to a resistance training program remains robust, and the physiological mechanisms associated with muscle repair and hypertrophy are still able to function even in older adults. […] regular resistance exercise may be one of the best intrinsic methods of protecting muscles from exercise-induced muscle damage and helping to normalize the rate and quality of muscle repair processes and adaptation to muscular activity in older individuals. […]

Over half of the older population is afflicted by arthritis, and approximately one-third of postmenopausal women will experience an osteoporotic fracture in their lifetime.1 […] distinguishing between disease processes and disease consequences of arthritis and normal aging is complex. […] More than 80% of people over the age of 75 years have clinical osteoarthritis and more than 80% of people over the age of 50 have radiologic evidence of the condition. Before the age of 50 years, the prevalence of osteoarthritis in most joints is higher in men than in women. […] The prevalence of rheumatoid arthritis increases with age and is found in about 10% of adults older than 65 years of age. […] The association between obesity and osteoarthritis has been evident for many years, but whether obesity was a cause or a consequence of the disease was unclear. It now appears that obesity not only predates osteoarthritis, but also increases the rate of disease progression, especially in women and in those with osteoarthritis of the knee.26,27 […] Occupational or recreational activities associated with excessive, repetitive, or high-impact joint loads are risk factors for osteoarthritis. In contrast, moderate physical activity such as running decreases the risk for osteoarthritis, at least in men.28 […] General exercise and physical activities are not harmful to the arthritic process, as was once thought. […] Despite many interstudy differences of methodology, there is consistent support for the notion that exercise and physical activity can and should play a role in both prevention and rehabilitation. […]

The one fairly robust finding across studies of exercise adherence is past exercise behavior. […] The fact that current exercise behavior appears to be best predicted by past exercise behavior is not surprising, but it is disconcerting, given that those individuals who most need to exercise are also those who are least likely to exercise. […]

Based upon […] systematic reviews of randomized investigations, the effect of exercise on bone mass appears to be a gain of approximately 1% per year, regardless of menopausal status. Data comparing male and female subjects are limited […]

Aging, per se, appears to have only a small effect on glucose intolerance and insulin resistance. The majority of insulin resistance that develops in older women and men is explained by increased adiposity, particularly in the abdominal region. […] life style appears to be a much stronger determinant of insulin resistance than aging per se. […]

Overall, exercise training in later life reduces the magnitude of catecholamine and pituitary hormone responses to a given bout of exercise, and increases the resistance to physical stress. Given that stress hormones are in general immunosuppressive, appropriate exercise training may be beneficial for the aging immune system, raising the threshold level of immunosuppressive (i.e., strenuous and/or prolonged) physical exercise. […] Strenuous exercise increases concentrations of various proinflammatory and anti-inflammatory cytokines, naturally occurring cytokine inhibitors, and chemokines.43 Given that aging is associated with increased inflammatory activity, strenuous exercise may induce the cytokine cascade more markedly in aging adults than in young peers. […] The results seem to imply that older adults should adopt a more cautious approach to strenuous exercise. […]

Human immune function undergoes adverse changes with aging. The T cells, which have a central role in cellular immunity, show the largest age-related differences in distribution and function. The underlying causes include thymus involution and continuous attrition caused by chronic antigenic overload. Immune function is apparently sexually dimorphic; women have more vigorous immunologic activity than do men, thus reducing their risks of infection. However, the same mechanisms make women more susceptible to various autoimmune diseases. The sexual dimorphism in immune function may become less apparent with aging, although it persists into later life […]

In Canada, by the age of 65 years, 37% of women and 31% of men have noted some limitations in their physical activity.28 As early as 55 years of age, 2% of men and 10% of women are unable to carry their groceries, and in those over the age of 80 years, the prevalence of this particular handicap rises to 20% of men and 30% of women.81 […] When interpreting functional changes, it is often difficult to disentangle what is a consequence of normal aging from the effects of disuse and chronic disease. One U.S. study estimated that as much as half of age-related decline in functional capacity was self-imposed, due to an accumulation of body fat and a failure to take adequate physical activity.32 […] in Denmark, Avelund et al.5 noted a substantial inverse relationship between levels of habitual physical activity and the loss of functional capacity. […]

Although an age-related loss of function can in itself cause disability, chronic disease is the usual source of impairment. […] A major part of the age-related loss of functional capacity, with the associated social and economic costs of prolonged disability, is due to adoption of an inappropriate lifestyle. […] The apparent prevalence of various disease conditions varies according to the diagnostic criteria applied. For example, aging is inevitably associated with a progressive increase in blood pressure, but when a certain arbitrary level of pressure is surpassed, hypertension is diagnosed […] Similarly, when the age-related decrease in bone mineral content reaches an arbitrary figure, clinical osteoporosis is diagnosed. […]

Many elderly people have multiple disorders, and it is then difficult to assess the contribution of specific conditions to the reported level of disability. In young and middle old age, the main causes of disability are chronic disease and a restriction of mobility, but in the oldest old mental deterioration and a loss of the special senses become important sources of impairment.62 […]

the main social costs associated with aging are incurred in the final year of life, as heroic attempts are made to prolong the survival of sedentary and severely disabled individuals.9,17 Regular physical activity decreases the risk of chronic disease and thus the scope for heroic treatment; it increases healthy life expectancy.17 […] there is little evidence that regular physical activity prolongs survival into advanced old age. What it does is to avert premature death, at a time when an individual is contributing to society rather than drawing upon its resources. The survival curves for active and sedentary individuals converge around the age of 80 years […]

The muscle force and aerobic power required to undertake many of the tasks important to the independence of an elderly person (for example, rising from a chair or climbing a flight of stairs) are almost directly proportional to an individual’s body mass. Thus, a 10% reduction in body mass will effectively increase muscle strength and maximal aerobic power by some 10%, equivalent to a 10-year reversal of the effects of aging.”

August 4, 2013 Posted by | books, data, demographics, health, medicine | 2 Comments

Gender, Physical Activity, and Aging

“In the past few years, research on age-related changes in biological function, physical capacity, and the training responses of women has grown. The time is thus opportune to present a succinct summary of these investigations, exploring the interesting issues of potential gender differences in both the course of aging and responses of the elderly to physical activity. This book undertakes this task, drawing upon the knowledge of leading experts in exercise gerontology.”

I’ve read books dealing with related stuff before, but from different angles not involving much stuff about gender differences.

The book spends a lot of time on gender-differences related to the aging process and how this stuff relates to stuff like physical activity and other environmental factors as well as genetics. Some of the chapters are quite easy to read, others are significantly more technical – chapter four occasionally contained stuff which was at least borderline beyond me (unless you’ve learned by heart stuff like what’s covered here and here you’ll either be at least somewhat lost or you’ll need to look up some stuff along the way), but I read on anyway (occasionally rewatching a Khan video…) and even if some of the finer details sometimes elude you you’ll probably learn some stuff. I thought the first parts of chapter 3 were quite weak compared to the stuff on those topics I’ve read elsewhere (can’t remember where – Whitbourne perhaps? Razib Khan?). On a general note I believe I’m leaning towards a 3 star goodreads rating at this point – I’ve read roughly half the book. There’s a lot of stuff and it’s a well documented book (aside from the intro chapter and chapter 5, all chapters so far have had more than 100 references – chapter 3 e.g. has more than 250 references), but some of the topics covered I don’t find to be very interesting, and e.g. the stuff on motivation in chapter 2 is way too theoretical to be of any use in an applied setting (so why are they wasting time writing about it?). I added a ‘pure speculation’ note in the margin at one point. But again, there’s a lot of good stuff – some stuff from the first half of the book:

“On average, a larger fraction of total body mass is fat and bone structure is lighter in women than in men; likewise, most older individuals accumulate body fat […] Gender differences in muscle strength and maximal aerobic power become much smaller if values are expressed per unit of lean body mass. A lean mass adjustment may be appropriate when comparing specific aspects of muscle and cardiac function between women and men, young and old, or indigenous and modern populations. But during most activities of normal daily life a person must displace the entire body mass rather than just lean tissue; if the type of activity to be performed requires a size adjustment, then total body mass is the most appropriate unit of reference.25 […] Fat is a poor conductor of heat, and if the layer of subcutaneous fat is increased […] then the person concerned must direct an increased fraction of total cardiac output to skin rather than working muscles when carrying out heavy physical work in a warm environment.27 This reduces the external power output for a given maximal oxygen intake […]

From a physiological point of view, there is an important gender difference in average blood hemoglobin concentration. […] The lower average figure in females reflects mainly physiologic influences […] Given the predominantly biological basis of the hemoglobin
differential, the gender gap seems likely to be small after menopause;37 at this stage, values remain relatively constant in women, but tend to decrease in men. […] The maximal oxygen content of unit volume of blood is stoichiometrically related to the hemoglobin concentration.20 Thus, for each liter of blood pumped by the heart, a woman necessarily transports approximately 10% less oxygen than a man, at least prior to menopause. […]

Over a 3-month period, even a moderate training program can augment muscle strength and maximal aerobic power by 20% or more, 25,26 — equivalent to a reversal of approximately 20 years of normal aging.26 It is thus important that any gender comparisons of the aging process be based on individuals who begin with a similar initial training status and are pursuing similar patterns of habitual physical activity. […] laboratory-measured indices of physical fitness (such as treadmill endurance time) bear a closer relationship to health outcomes than do questionnaire assessments [my emphasis, US] […]

One U.S. study attributed as much as a half of variance in the age-linked decrease in relative aerobic power to a combination of a decrease in habitual physical activity and an increase in body mass.12 In order to see the true influences of gender and aging upon the primary variable, it is thus desirable to focus attention on a population sample where habitual physical activity and body fat content have remained constant over the individual’s life span. […]

it has […] been recognized for many years that the aging process actually shows some curvilinearity. […] For example, women show a sharp acceleration in the rate of bone mineral loss during the five years around the age of menopause;18 in men, there is an accelerating loss of lean tissue after retirement […] In women, functional losses accelerate with the distinct end point of menopause, but in men the process begins at a later age, and develops more gradually […]

Two robust findings from population-based surveys are that older adults are less active than younger adults, and older women are less active than older men. […] Physical activity is associated with physical function, even in those living with chronic disease52 and is inversely related to disability in women.53 A low level of physical activity is considered a risk factor for functional decline in older adults.54,55 […]

Older adults have an average of 11.4 contacts per year with a physician or health care provider,1 […]

randomized controlled exercise trials with older adults have shown that exercise adherence is comparable or superior for home-based compared to group-based exercise.122-124 […] older adults are more likely to cite the health benefits of physical activity as a motivation for exercise than are younger adults. […] perceived exercise benefits or outcome expectations are positively associated with physical activity participation in older women.82,96,98,109,128 […]

Official statistics show that women have had a substantially longer average survival than men in all countries of the world throughout the past century […] Lifestyle and other environmental factors plainly modulate any underlying effect of constitution, since gender discrepancy increased very substantially in almost all countries during the first two thirds of the 20th century (Table 3.1). In the U.S., for example, the gender differential increased from 5.8% in 1900 to 9.6% in 1990. […] Gains in life expectancy during the present century have been much larger in women than in men.206,236 Thus, a large fraction of the elderly population and an even larger fraction of those who are very old are now women. […] Although women currently have a much longer average survival than men, many aspects of function such as aerobic power and muscular strength deteriorate at a similar absolute rate in the two sexes […] Thus, women face a substantially longer period than men when their level of function is insufficient to undertake instrumental and other activities of daily living43,210,224 […] Indeed, their active life expectancy112 may be no greater than that of their male counterparts. […] In the U.S., the total number of disabled life years averages 10.8 years for men and 14.0 years for women,223 and in some countries the gender discrepancy in chronic disability is even larger.” […]

“the major effect of genotype is probably in terms of increasing an individual’s susceptibility to various causes of disability and
premature death. […] In general, women have fewer material resources than men during old age.132 They are much more likely to be living on their own,16 and thus tend to receive less psychological support than men.42,149,167 They are also more likely to be disabled. In general, disability increases social isolation […] Growing evidence suggests that physical activity helps to maintain social contacts, enhances mental health, and sustains cognitive ability in the very old.210 […]

Gender differences in the prevalence of cigarette smoking have in the past explained much of the shorter average life span of the male. Indeed, after allowing for a somewhat greater incidence of traumatic deaths among men, there is almost no gender difference in life expectancy among nonsmokers.155 Now that the prevalence of smoking also shows little sex difference, a progressive equalization of average life span between men and women can be anticipated.173 [my emphasis, US] […]

Particularly in the final years of life, the greatest dividend from regular physical activity may be an increase in the individual’s functional capacity and thus quality of life […] rather than an extension of life span.209 […] Because women generally have a smaller functional margin than men, they are likely to show a larger gain in quality of life as a result of participating in a regular physical activity program […]

Specific genes can now be identified that increase the risk of various chronic diseases, affecting an individual’s quality of life and survival prospects. Furthermore, the frequency of occurrence of such genes in some instances differs between the two sexes. Nevertheless, various environmental challenges exert a powerful influence, both in their own right and as the reason why adverse genetic characteristics become manifest. The gender difference in survival thus seems determined almost entirely by environmental factors, with cigarette smoking playing a dominant role. […]

there is typically a ten-year delay in the onset of coronary symptoms for women compared to men […]

4.1
Here’s a quote from chapter 4 I picked because I figured it’d convey part of why this chapter may be hard to understand for some readers:

“With advancing age, strategies used to augment cardiac output during exercise shift from codependence on catecholamine-mediated inotropic, chronotropic, and volumetric means, to greater dependence on changes in ventricular end-diastolic volume via the Frank–Starling mechanism. […] The key signaling proteins involved in both the amplification and integration of extracellular signals in the myocyte (from the sarcoplasmic reticulum to the intracellular effectors) include β-adrenergic receptors, G proteins, and adenyl cyclase. […] Variability in the R-R interval, used as an index of parasympathetic tone, is augmented with aerobic training …”

Back to the ‘not-that-hard-to-understand-stuff’:

Exercise training which increases and augments the ventilation threshold by 10 to 15% may increase the time to fatigue by as much as 180% when exercising at a fixed intensity. Improvement in submaximal aerobic exercise performance is believed to result in part from change in skeletal muscle metabolism. These changes in submaximal performance have a profound effect on the ability of older people to function in daily life.99 [my emphasis, US] […] Aging of the cardiorespiratory system is not due to breakdown in a single step of oxygen conductance from the atmosphere to exercising muscles. Rather, there are physiological changes in each of the series of resistors. […] it is apparent that aging affects each step in the delivery of oxygen, often with differing effects across genders. Many of these changes are slowed or reversed by exercise training […]

It has been widely demonstrated that greater physical activity is associated with a reduced all-cause mortality in men. However, this relationship has been studied less frequently in women. […] Physical activity and longevity were investigated by Paffenbarger et al.1 in 17,000 male Harvard alumni aged 35 to 74 years. With physical activity assessed by questionnaire in the follow up, it was estimated that those expending greater than 8.4 megajoules per week in exercise (walking, stair climbing, sports play) had a 25 to 30% lower mortality rate than those with lower weekly energy expenditures. Paffenbarger and colleagues2,3 also showed that physical activity participation (in the form of moderately vigorous sports play) initiated in middle-age was independently associated with a 23% lower all-cause death rate …”

August 1, 2013 Posted by | biology, books, data, demographics, health, medicine | Leave a comment

Calculated Risks: Understanding the Toxicity of Chemicals in our Environment (ii)

I finished the book – I didn’t expect to do that quite so soon, which is why I ended up posting two posts about the book during the same day. As Miao pointed out in her comment a newer version of the book exists, so if my posts have made you curious you should probably give that one a shot instead; this is a good book, but sometimes you can tell it wasn’t exactly written yesterday.

This book may tell you a lot of stuff you already know, especially if you have a little knowledge about biological systems, the human body, or perhaps basic statistics. I considered big parts of some chapters to be review stuff I already knew; I’d have preferred a slightly more detailed and in-depth treatment of the material. I didn’t need to be reminded how the kidneys work or that there’s such a thing as a blood-brain barrier, the stats stuff was of course old hat to me, I’m familiar with the linear no-threshold model, and there’s a lot of stuff about carcinogens in Mukherjee not covered in this book…

So it may tell you a lot of stuff you already know. But it will also tell you a lot of new stuff. I learned quite a bit and I liked reading the book, even the parts I probably didn’t really ‘need’ to read. I gave it 3 stars on account of the ‘written two decades ago’-thing and the ‘I don’t think I’m part of the core target group’-thing – but if current me had read it in the year 2000 I’d probably have given it four stars.

I don’t really know if the newer edition of the book is better than the one I read, and it’s dangerous to make assumptions about these things, but if he hasn’t updated it at all it’s still a good book, and if he has updated the material the new version is in all likelihood even better than the one I read. If you’re interested in this stuff, I don’t think this is a bad place to start.

I found out while writing the first post about the book that quoting from the book is quite bothersome. I’m lazy, so I decided to limit coverage here to some links which I’ve posted below – the stuff I link to is either covered or related to stuff that is covered in the book. It was a lot easier for me to post these links than to quote from the book in part because I visited many of these articles along the way while reading the book:

Aflatoxin.
No-observed-adverse-effect level.
Teratology.
Paraquat.
DDT.
Methemoglobinemia.
Erethism.
DBCP.
Diethylstilbestrol.
Polycyclic aromatic hydrocarbon.
Percivall Pott.
Bioassay.
Electrophile.
Clastogen.
Mutagen.
Dose–response relationship.
Acceptable daily intake.
Linear Low dose Extrapolation for Cancer Risk Assessments: Sources of Uncertainty and How They Affect the Precision of Risk Estimates (short paper)
Delaney clause.
OSHA.

Do note that these links taken together can be somewhat misleading – as you could hopefully tell from the quotes in the first post, the book is quite systematic and the main focus is on basic/key concepts. To the extent that specific poisons like paraquat and DDT are mentioned in the book they’re used to ‘zoom in’ on a certain aspect in order to illustrate a specific feature, or perhaps in order to point out an important distinction – stuff like that.

July 30, 2013 Posted by | biology, books, health, medicine | Leave a comment

Calculated Risks: Understanding the Toxicity of Chemicals in our Environment

So what is this book about? The introductory remarks below from the preface provide part of the answer:

“A word about organization of topics […] First, it is important to understand what we mean when we talk about ‘chemicals’. Many people think the term refers only to generally noxious materials that are manufactored in industrial swamps, frequently for no good purpose. The existence of such an image impedes understanding of toxicology and needs to be corrected. Moreover, because the molecular architecture of chemicals is a determinant of their behaviour in biological systems, it is important to create a little understanding of the principles of chemical structure and behavior. For these reasons, we begin with a brief review of some fundamentals of chemistry.

The two ultimate sources of chemicals – nature and industrial and laboratory synthesis – are then briefly described. This review sets the stage for a discussion of how human beings become exposed to chemicals. The conditions of human exposure are a critical determinant of whether and how a chemical will produce injury or disease, so the discussion of chemical sources and exposures naturally leads to the major subject of the book – the science of toxicology.

The major subjects of the last third of this volume are risk assessment […] and risk control, or management, and the associated topic of public perceptions of risk in relation to the judgments of experts.”

What can I say? – it made sense to read a toxicology textbook in between the Christie novels… The book was written in the 90s, but there are a lot of key principles and -concepts covered here that probably don’t have a much better description now than they did when the book was written. I wanted the overview and the book has delivered so far – I like it. Here’s some more stuff from the first half of the book:

“the greatest sources of chemicals to which we are regularly and directly exposed are the natural components of the plants and animals we consume as foods. In terms of both numbers and structural variations, no other chemical sources matches food. We have no firm estimate of the number of such chemicals we are exposed to through food, but it is surely immense. A cup of coffee contains, for example, nearly 200 different organic chemicals – natural components of the coffee bean that are extracted into water. Some impart color, some taste, some aroma, others none of the above. The simple potato has about 100 different natural components …” […]

“These facts bring out one of the most important concepts in toxicology: all chemicals are toxic under some conditions of exposure. What the toxicologist would like to know are those conditions. Once they are known, measures can be taken to limit human exposures so that toxicity can be avoided.” […]

The route of exposure refers to the way the chemical moves from the exposure medium into the body. For chemicals in the environment the three major routes are ingestion (the oral route), inhalation, and skin contact (or dermal contact). […]

The typical dose units are […] milligram of chemical per kilogram of body weight per day (mg/kg b.w./day). […] For the same intake […] the lighter person receives the greater dose. […] Duration of exposure as well as the dose received […] needs to be included in the equation […] dose and its duration are the critical determinants of the potential for toxicity. Exposure creates the dose.” […]

Analytical chemistry has undergone extraordinary advances over the past two to three decades. Chemists are able to measure many chemicals at the part-per-billion level which in the 1960s could be measured only at the part-per-million level […] or even the part-per-thousand level. […] These advances in detection capabilities have revealed that industrial chemicals are more widespread in the environment than might have been guessed 10 or 20 years ago, simply because chemists are now capable of measuring concentrations that could not be detected with analytical technology available in the 1960s. This trend will no doubt continue …” […]

“The nature of toxic damage produced by a chemical, the part of the body where that damage occurs, the severity of the damage, and the likelihood that the damage can be reversed, all depend upon the processes of absorption, distribution, metabolism and excretion, ADME for short. The combined effects of these processes determine the concentration a particular chemical […] will achieve in various tissues and cells of the body and the duration of time it spends there. Chemical form, concentration, and duration in turn determine the nature and extent of injury produced. Injury produced after absorption is referred to as systemic toxicity, to contrast it with local toxicity.” […]

“Care must be taken to distinguish subchronic or chronic exposures from subchronic or chronic effects. By the latter, toxicologists generally refer to some adverse effect that does not appear immediately after exposure begins but only after a delay; sometimes the effect may not be observed until near the end of a lifetime, even when exposure begins early in life (cancers, for example, are generally in this category of chronic effects). But the production of chronic effects may or may not require chronic exposure. For some chemicals acute or subchronic exposures may be all that is needed to produce a chronic toxicity; the effect is a delayed one. For others chronic exposure may be required to create chronic toxicity.” […]

“In the final analysis we are interested not in toxicity, but rather in risk. By risk is meant the likelihood, or probability, that the toxic properties of a chemical will be produced in populations of individuals under their actual conditions of exposure. To evaluate the risk of toxicity occurring for a specific chemical at least three types of information are requred:
1) The types of toxicity the chemical can produce (its targets and the forms of injury they incur).
2) The conditions of exposure (dose and duration) under which the chemical’s toxicity can be produced.
3) The conditions (dose, timing and duration) under which the population of people whose risk is being evaluated is or could be exposed to the chemical.

It is not sufficient to understand any one or two of these; no useful statement about risk can be made unless all three are understood.” […]

“It is rare that any single epidemiology study provides sufficiently definitive information to allow scientists to conclude that a cause-effect relationship exists between a chemical exposure and a human disease. Instead epidemiologists search for certain patterns. Does there seem to be a consistent association between the occurence of excess rates of a certain condition (lung cancer, for example) and certain exposures (e.g. to cigarette smoke) in several epidemiology studies involving different populations of people? If a consistent pattern of associations is seen, and other criteria are satisfied, causality can be established with reasonable certainty. […] Epidemiology studies are, of course, only useful after exposure has occurred. For certain classes of toxic agents, carcinogens being the most notable, exposure may have to take place for several decades before the effect, if it exists, is observable […] The obvious point is that epidemiology studies can not be used to identify toxic properties prior to the introduction of the chemical into commerce. This is one reason toxicologists turn to the laboratory. […] “The ‘nuts and bolts’ of animal testing, and the problems of test interpretation and extrapolation of results to human beings, comprise one of the central areas of controversy in the field of chemical risk assessment.” […]

Toxicologists classify hepatic toxicants according to the type of injuries they produce. Some cause accumulation of excessive and potentially dangerous amounts of lipids (fats). Others can kill liver cells; they cause cell necrosis. Cholestasis, which is decreased secretion of bile leading to jaundice […] can be produced as side effects of several therapeutic agents. Cirrhosis, a chronic change characterized by the deposition of connective tissue fibers, can be brought about after chronic exposure to several substances. […] ‘hepatotoxicity’ is not a very helpful term, because it fails to convey the fact that several quite distinct types of hepatic injury can be induced by chemical exposures and that, for each, different underlying mechanisms are at work. In fact, this situation exists for all targets, not only the liver.”

July 30, 2013 Posted by | biology, books, health, medicine | 2 Comments

A few Cochrane reviews and links

I’ve spent the last few days at my parents’ place and haven’t had much time for blogging due to social obligations. I read The Murder on the Links the day before yesterday and I’ll finish Lord Edgware Dies later today – I’ll probably blog the books tomorrow. For now I’ll just post a few Cochrane reviews and a couple of links:

i. Abstinence-only programs for preventing HIV infection in high-income countries (as defined by the World Bank). (link to the full paper here)

“Abstinence-only programs are widespread and well-funded, particularly in the United States and countries supported by the US President’s Emergency Plan for AIDS Relief. On the premise that sexual abstinence is the best and only way to prevent HIV, abstinence-only interventions aim to prevent, stop, or decrease sexual activity. These programs differ from abstinence-plus designs: abstinence-plus programs promote safer-sex strategies (e.g., condom use) along with sexual abstinence, but abstinence-only programs do not, and instead often highlight the limitations of condom use. An up-to-date review suggests that abstinence-only programs do not affect HIV risk in low-income countries; this review examined the evidence in high-income countries.

This review included thirteen randomized controlled trials comparing abstinence-only programs to various control groups (e.g., “usual care,” no intervention). Although we conducted an extensive international search for trials, all included studies enrolled youth in the US (total baseline enrollment=15,940 participants). Programs were conducted in schools, community centers, and family homes; all were delivered in family units or groups of young people. We could not conduct a meta-analysis because of missing data and variation in program designs. However, findings from the individual trials were remarkably consistent.

Overall, the trials did not indicate that abstinence-only programs can reduce HIV risk as indicated by behavioral outcomes (e.g., unprotected vaginal sex) or biological outcomes (e.g., sexually transmitted infection). Instead, the programs consistently had no effect on participants’ incidence of unprotected vaginal sex, frequency of vaginal sex, number of sex partners, sexual initiation, or condom use.”

ii. Healthcare financing systems for increasing the use of tobacco dependence treatment.

The short version:

“Apart from providing counselling and drug treatment, strategies that reduce or cover the costs of accessing or providing these treatments could help smokers quit.

We found eleven trials, eight of which involve financial interventions directed at smokers and three of which involve financial interventions directed at healthcare providers.

Covering all the costs of smoking cessation treatment for smokers when compared to providing no financial benefits increased the proportion of smokers attempting to quit, using smoking cessation treatments, and succeeding in quitting. Although the absolute differences in quitting were small, the costs per person successfully quitting were low or moderate. Financial incentives directed at healthcare providers did not have an effect on smoking cessation.”

From the paper:

Summary of main results:

With very high to modest levels of consistency, we detected a statistically significant positive effect of full financial interventions targeting smokers with regard to abstinence from smoking compared to provision of no financial intervention at six months follow-up or more (all abstinence measures: RR 2.45, 95% CI 1.17 to 5.12). The effect of full financial interventions was also extended to favourable outcomes on the use of smoking cessation treatments: the pooled effect of full coverage compared with no financial intervention on the use of smoking cessation treatments was highly significant for each treatment type (NRT, bupropion, and behavioural interventions).Despite the observation of multiple favourable effects of full as compared to no financial intervention, when full coverage was compared to partial coverage, results showed no significant effect on smoking cessation or quit attempts. […]

Five studies presented data on cost effectiveness. When full benefit was compared with partial or no benefit, the costs per quitter ranged from $119 to $6,450. [the $6,450 estimate is an outlier in that group; the other estimates are all much lower, at or below $1500/quitter – US] […]

In this review, covering the full cost to smokers of using smoking cessation treatment increased the number of successful quitters, the number of participants making a quit attempt, and the use of smoking cessation treatment when compared with no financial coverage. As the majority of the studies were rated at high or unclear risk of bias in three or more domains, and there was variation between the settings, interventions and participants of the included studies, the results should be interpreted cautiously. The differences in self-reported abstinence rate, number of participants making a quit attempt and use of smoking cessation treatments were modest.”

iii. Psychosocial and pharmacological treatments for deliberate self harm.

“Deliberate self-harm is a major health problem associated with considerable risk of subsequent self-harm, including completed suicide. This systematic review evaluated the effectiveness of various treatments for deliberate self-harm patients in terms of prevention of further suicidal behaviour. […]

Main results:

A total of 23 trials were identified in which repetition of deliberate self-harm was reported as an outcome variable. The trials were classified into 11 categories. The summary odds ratio indicated a trend towards reduced repetition of deliberate self-harm for problem-solving therapy compared with standard aftercare (0.70; 0.45 to 1.11) and for provision of an emergency contact card in addition to standard care compared with standard aftercare alone (0.45; 0.19 to 1.07). The summary odds ratio for trials of intensive aftercare plus outreach compared with standard aftercare was 0.83 (0.61 to 1.14), and for antidepressant treatment compared with placebo was 0.83 (0.47 to 1.48). […]

Authors’ conclusions:

There still remains considerable uncertainty about which forms of psychosocial and physical treatments of self-harm patients are most effective, inclusion of insufficient numbers of patients in trials being the main limiting factor. There is a need for larger trials of treatments associated with trends towards reduced rates of repetition of deliberate self-harm. The results of small single trials which have been associated with statistically significant reductions in repetition must be interpreted with caution and it is desirable that such trials are also replicated.”

A few other links which are not from the Cochrane site:

iv. Plausible indeed!

v. Errors in DCP2 cost-effectiveness estimate for deworming.”Over the past few months, GiveWell has undertaken an in-depth investigation of the cost-effectiveness of deworming, a treatment for parasitic worms that are very common in some parts of the developing world. While our investigation is ongoing, we now believe that one of the key cost-effectiveness estimates for deworming is flawed, and contains several errors that overstate the cost-effectiveness of deworming by a factor of about 100. This finding has implications not just for deworming, but for cost-effectiveness analysis in general: we are now rethinking how we use published cost-effectiveness estimates for which the full calculations and methods are not public. […]we see this case as a general argument for expecting transparency, rather than taking recommendations on trust – no matter how pedigreed the people making the recommendations. Note that the DCP2 was published by the Disease Control Priorities Project, a joint enterprise of The World Bank, the National Institutes of Health, the World Health Organization, and the Population Reference Bureau, which was funded primarily by a $3.5 million grant from the Gates Foundation. The DCP2 chapter on helminth infections, which contains the $3.41/DALY estimate, has 18 authors, including many of the world’s foremost experts on soil-transmitted helminths.”

vi. Evolution, Creationism, Intelligent Design – a Gallup poll from last year. According to that poll a majority of Americans (56%) think creationism should be taught in public school science classes. One of the questions asked were: If the public schools in your community taught the theory of evolution, — that is, the idea that human beings evolved from other species of animals — would you be upset, or not?  A third of the people asked (34%) answered yes to this question. Incidentally in related news it should be noted that in a recent poll of South Korean biology teachers, 40% of them “agreed with the statement that “much of the scientific community doubts if evolution occurs”; and half disagreed that “modern humans are the product of evolutionary processes”.”

In slightly related news, according to an older poll conducted shortly before the turn of the century roughly one in five Americans asked back then didn’t know that the Earth revolves around the Sun. Other countries didn’t do any better:

“Gallup also asked the following basic science question, which has been used to indicate the level of public knowledge in two European countries in recent years: “As far as you know, does the earth revolve around the sun or does the sun revolve around the earth?” In the new poll, about four out of five Americans (79%) correctly respond that the earth revolves around the sun, while 18% say it is the other way around. These results are comparable to those found in Germany when a similar question was asked there in 1996; in response to that poll, 74% of Germans gave the correct answer, while 16% thought the sun revolved around the earth, and 10% said they didn’t know. When the question was asked in Great Britain that same year, 67% answered correctly, 19% answered incorrectly, and 14% didn’t know.”

You do have a potential ‘this is a silly question so I want to mess with the people asking it’-effect lurking in the background, but that’s probably mostly related to people giving the wrong answer deliberately. But even if many of the people asked perhaps gave the wrong answer deliberately, there’s still a substantial number of people answering that they ‘don’t know.’ I found the numbers surprising and I would love to see some updated estimates; a brief googling didn’t turn up anything.

July 28, 2013 Posted by | data, demographics, evolution, health, medicine, Psychology, religion, studies | 5 Comments

Tuberculosis

Khan Academy has, in collaboration with Stanford School of Medicine, made 15 videos about the disease (so far) with a total duration of two hours and 37 minutes – you can watch all of them here. There’s some overlap here and there, different videos covering similar stuff, and a lot of details are left out. But this is still good stuff and the videos were an enjoyable part of my day yesterday. I know I’ve covered this disease before, but given how many people have been exposed and how important it was in the past (roughly one hundred years ago one sixth of all French deaths were due to this disease), this is arguably a disease you should at least have some knowledge about. Some samples from the playlist below:

Note that the spelling is off in that video; it’s Ghon focus and Ghon’s complex, not Gohn -ll- …

A quote from the last video above: “DOT – Directly Observed Therapy – is very important.”

There are theoretical reasons why DOT may be useful/efficient, as mentioned in the video. And I’ve seen it argued elsewhere that “treating tuberculosis with the DOTS strategy is highly cost-effective” [DOTS means “directly observed therapy, short course” – which is a specific type of DOT therapy; “a comprehensive tuberculosis management programme that focuses on low-income countries.” (see the Cochrane link for more)]. But I’m also aware that there are reasons to be skeptical as well:

Authors’ conclusions

The results of randomized controlled trials conducted in low-, middle-, and high-income countries provide no assurance that DOT compared with self administration of treatment has any quantitatively important effect on cure or treatment completion in people receiving treatment for tuberculosis.

PLAIN LANGUAGE SUMMARY

Directly observing people taking their tuberculosis drugs did not improve the cure rate compared with people without direct monitoring of treatment

Tuberculosis is a very serious health problem with two million people dying each year, mostly in low-income countries. Effective drugs for tuberculosis have been available since the 1940s, but the problem still abounds. People with tuberculosis need to take the drugs for at least six months, but many do not complete their course of treatment. For this reason, services for people with tuberculosis often use different approaches to encourage people to complete their course of treatment. This review found no evidence that direct observation by health workers, family members, or community members of people taking their medication showed better cure rates that [sic] people having self administered treatment. The intervention is expensive to implement, and there appears to be no sound reason to advocate its routine use until we better understand the situations in which it may be beneficial.”

July 11, 2013 Posted by | health, health care, Khan Academy, Lectures, medicine | Leave a comment