Clueless Americans think they’ll never get sick

Cassie Murdoch points to a report from a corporate survey:

Sixty-two percent of U.S. employees say it’s not likely they or a family member will be diagnosed with a serious illness like cancer, a survey indicates.

The Aflac WorkForces Report, a survey of nearly 1,900 benefits decision-makers and more than 6,100 U.S. workers, also indicated 55 percent said they were not very or not at all likely to be diagnosed with a chronic illness, such as heart disease or diabetes.

Here are some actual statistics:

The American Cancer Society, Cancer Facts & Figures 2012, said 1-in-3 women and 1-in-2 men will be diagnosed with cancer at some point in their lives, and the National Safety Council, Injury Facts 2011 edition, says more than 38.9 million injuries occur in a year requiring medical treatment.

The American Heart Association, Heart Disease & Stroke Statistics 2012, said 1-in-6 U.S. deaths were caused by coronary heart disease, Tillman said.

And some details on the survey:

The survey conducted in January and February by Research Now. The first 3,151 worker interviews were nationally representative, while the remaining 3,000 interviews were conducted among the Top 30 designated market areas.

Did these people really say they that neither they nor a family member will have a serious illness? Is this for real? What were they thinking? I’m used to seeing wacky survey findings, but this one is ridiculous.

19 thoughts on “Clueless Americans think they’ll never get sick

  1. The question is ridiculous. You might as well ask if they will ever die. Ever and never are greatly abused in surveys. The question is what the time frame for diagnosis is. A reasonable time frame for this may be during their work life. The numbers may still be higher than people believe but they are much less than certain.

    • I was going to say this as well. The statistics on 1 in 2 or 1 in 3 are also ridiculous, what fraction of those people got cancer after age 70 say? Biologists will tell you that you have pretty much a 100% chance of getting cancer if you live long enough… but many people die of other causes before they get cancer. Many people over age say 80 who die of some non-cancer cause and are autopsied will probably have some sort of cancer, perhaps a slow growing one like prostate.

      Surveys would be better if they said something like “how likely do you think it is that you will be diagnosed with a serious or life threatening illness requiring hospitalization before you retire?” or maybe “in the next 10 years” or “before your children have graduated from college” or some other meaningful timeframe for people’s realistic planning horizons.

      • I guess my point being that if you don’t specify an explicit planning horizon you won’t get the respondent to use a default of “in your lifetime” you’ll get an implicit default of “in whatever time frame you feel like considering at the moment” which will be very different for different people and at different times of the day or depending on what life stage those people are in etc.

        • Given the context of a survey about valuation of employer-provided health benefits, I would suspect people are assuming “while you work for this employer” or at most “before you are 65 and on medicare.” The other badly interrogated part is the meaning of “likely”. For example, if “likely” means “more likely than not” and 1/6 of deaths are due to cardiovascular disease, you aren’t likely to die from it. There is no reason to draw a line from the fraction reporting “likely” to the percentage of people who get the disease, except under assumptions of very precise individualized knowledge of their risks.

      • Daniel noted,

        “Surveys would be better if they said something like “how likely do you think it is that you will be diagnosed with a serious or life threatening illness requiring hospitalization before you retire?” or maybe “in the next 10 years” or “before your children have graduated from college” or some other meaningful timeframe for people’s realistic planning horizons.”

        Or, as my estate-planning attorney noted, “that you will be hit by a bus when you leave my office? Or next year?”

        That is, no one knows when they are going to need serious medical care.

        The question needs to be put in a way that will get the respondent to think about this.

        Young people may think that they are immortal. But even young people will have experience of other young people experiencing life-threatening situations. The answer to such a question has to reflect this experience.

  2. So, from your posting there are 62% and 55% of folks saying the don’t think diagnosis of a serious or chronic illness is likely; does that mean 38% and 45% said they think it is likely? The 45% and 38% thinking it is likely isn’t too far off from the actual statistics you cited. Maybe they’re not quite so clueless.

    • The question refers to “you or a family member”. Obviously, the probability of a family member having a serious or chronic illness is going to be a function of the size of a family unit and, except for single people, will be considerably higher than the corresponding probabilities for an individual.

      The main issue would seem to concern the reasoning behind the optimistic assessments on the part of the respondent population. Some obvious possibilities come to mind:

      1. Is there an implicit time horizon (e.g., 5 or 10 years)?
      2. Do they think that major and chronic illnesses are much less common than they actually are?, or,
      3. Do they think that these things are more likely to happen to other people?

  3. Well, hold on a second. If the American Cancer society says 1/3 of women and 1/2 of men will be diagnosed with cancer at some point in their lives, then roughly 58% of Americans will never receive such a diagnosis, right? I’d say that’s remarkably close to 62% — especially given that the survey seems limited to employees, who generally receive better health care than the unemployed.

    Yes, yes, the question seems to have been about the likelihood of oneself or a family member being diagnosed with a serious illness “like cancer”, but inclusion of the word “cancer” probably anchored the responses to cancer alone.

    Sounds like these clueless Americans are smarter than you think.

  4. Based upon the amount that I hear of U.S. folks of employable age being concerned about health benefits I suspect that the survey report isn’t reflective of relevant beliefs. Perhaps it’s picking up a near-costless superstitious response.

  5. Wishful thinking is part of the human condition. Nobody ever dies, nobody ever gets sick – well other people do, just not me or my family.

  6. @Lord: But it could also be a sign of people under-estimating the risk of rare but very bad risks. That is the more concerning angle, for questions like “can we save for retirement” or “can we self-insure against major medical risks”. Part of the argument for such models involves good risk modeling on the part of individuals. If we have evidence that they have cognitive bias towards under-estimating risk or short time frames, that is worthy of note.

  7. Cancer is mainly a disease of old age and smokers. Coronary problems run in families and usually occur after age 50. The Pareto Principle also applies to medical insurance claims– 20% of subscribers are responsible for 80% of claims. So it’s perfectly reasonable for a working 30 year old, not to expect a serious illness over the next 30 years. So I agree completely with Lord– it’s a stupid question. Since all people are mortal, they must eventually die of something. Everybody knows that, I’m sure the people who answered the question had some time frame in mind.

  8. If the survey wasn’t anonymous (and as it was done in the workplace) then perhaps these employees are saying to their bosses that they or their family aren’t going to cost the company a lot of money through getting sick and claiming on company health insurance and so they are worth continued employment (even if they have no idea of it being true).

    It’s not like people want to put a target on themselves and say “I’m costly, fire me!”

    It would be interesting to see if those people who said they or their family may get some chronic disease are still as likely to be employed in the company in three years as those who didn’t.

  9. I don’t have the numbers to hand, but my impression of the cancer suite of illnesses is that many cancers are completely survivable given the right treatment. So maybe another issue was that respondents were using the availability heuristic, which meant that for example – keeping with the cancer idea – they only thought about lung, breast, prostate, and bowel cancer and forgot about all the lymphomas etc. Re diabetes, I would expect the rates to go ever upwards given trends in obesity.

  10. Having seen people die of cancer in their fifties, and with no effective treatments, I think a few of you are too credulous regarding the optimistic won’t-happen-to-me mindset.

Comments are closed.