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.] […]
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.
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.”
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:
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.
No comments yet.