Over many years of intermittently reading various medical journals I began to notice that the arguments in favor of public health measures were often based solely on society-wide benefits. The effects are usually small on an individual basis and only matter when applied to a large population. As reported in the popular press, the recommendations are simplified to do this because it will make you live longer.
The arguments in the journals tend to go like this:
- The relative risk of behavior X resulting in disease Y is 1.07, which means that 2,057 more individuals are expected to die per year from disease Y in the US and resulting an additional $158M over ten years in medical costs. Public health recommendations should be issued to discourage behavior X. Taxation or other disincentives should be considered.
- Diagnostic test W reduces deaths from disease Z with an odds ratio of 0.89. With (target) population-wide screening with test W, 3,700 fewer individuals will die per year from disease Z and there will be an increase of QALYs (quality-adjusted life-years) by 29,000. Universal screening will cause 580 deaths and 5,700 complications, 1,400 of which will be severe. Primary care physicians should advise their patients in the target population to be screened by test W.
What is missing is any consideration of an individual’s risk tradeoffs. A relative risk of 1.07 in example #1 means one is 7% more likely to contract disease X. A person may decide that a 7% increase in the chance of getting some obscure disease is not worth giving up behavior Y. An odds ratio of 0.89 in example #2 means the chances of dying from disease Z is reduced by about 11%. Someone might reasonably decide that modest reduction is not worth the pain, expense, or hassle of getting diagnostic test W.
The reasoning applies to all manner of public health recommendations: medications, vaccines, dietary guidelines, diagnostic tests, and other therapeutic interventions. Just because it saves money or trouble for the healthcare system or for the nation as a whole doesn’t mean it’s right for you.
Regarding disease screening specifically, in an op-ed a professor of medicine at Dartmouth notes that
Screening the apparently healthy potentially saves a few lives (although the National Cancer Institute couldn’t find any evidence for this in its recent large studies of prostate and ovarian cancer screening). But it definitely drags many others into the system needlessly — into needless appointments, needless tests, needless drugs and needless operations (not to mention all the accompanying needless insurance forms).
This process doesn’t promote health; it promotes disease. People suffer from more anxiety about their health, from drug side effects, from complications of surgery. A few die. And remember: these people felt fine when they entered the health care system.
Furthermore, there is the replication crisis in medicine, which means that many of the results that lead to public health recommendations turn out to vanish upon closer inspection. The science is not settled, as we’ve seen with the government’s dietary guidelines. As John Ioannidis pointed out in a video in that linked post, odds ratios have a funny way of converging on 1 (no correlation) when other researchers attempt replication. This image, captured from his talk, shows what happens to odds ratios as investigators dig deeper:
Everyone’s situation is unique. None of this is meant to imply that any particular test, medication, procedure, or dietary recommendation is not appropriate for a given person. However, one should be aware that the criteria used for making these recommendations do not necessarily align with the value of any specific intervention for the individual.
The more thoughtful and careful physicians take a patient’s circumstances into account but are also bound by the best practices set by professional organizations and governments. Even if it’s true that something is good for public health, that doesn’t mean it’s good for you. Next time your doctor recommends a ‘routine’ test or prescribes a drug because the guidelines say so, you might want to ask how it will help you.