the popularity paradox


Imag0261

Popped into town to see @mgtmccartney talk at Conway Hall about some of the reasons behind the paradox that screening a population for disease, if not properly thought through, can do more harm than good.

I just wanted to make a note of one of the many issues discussed: what she calls the popularity paradox. Brief background: since no screening test is perfect, alongside the group of people for whom a genuine disease is detected and who are genuinely benefited by resulting treatment, there will be a group of false positives??(accidentally telling someone they have the disease when they don't), and a group of overdiagnoses (e.g. forms of cancer that are so slow-growing that they will probably never cause a symptom and without screening the patient would have died of old age without ever realising they had it). Patients in the groups get harmed with needless worry, needless treatment, and whatever side-effects those treatments have. Whether a particular form of screening is worthwhile depends on the relative proportions of the different groups, etc etc, go read it all in the book, there's a discount code in the picture.

The point was that as a particular form of screening gets promoted and pushed indiscriminately, spreading beyond symptomatic patients or high-risk populations, the proportions shift as the "harmed" groups grow faster than the "helped" groups. But since those??overdiagnosed patients will never know that they were overdiagnosed and harmed, everybody treated will assume that they were in the group of patients who genuinely needed treatment. Where previously you get people who were tested because they went to their GP with symptoms — and therefore say "getting those symptoms checked out saved my life" — and people who were tested because they were at risk — and therefore say "knowing I was at risk saved my life" — you now get a lot more people saying to their friends "screening saved my life, make sure you get checked." (And, since nobody knows that they're in the overdiagnosed group, nobody says "I went through the trauma of needless invasive treatment that permanently affects my quality of life because of this" — the harm seems worth it if you believe your life was saved.)

Anyway, the story reminded me of the anecdata for bicycle helmets. Tell people that they need a medical intervention and they'll be convinced it saved their life.

sent from my phone – 'scuse brevity and typos.

Leave a comment

Your email address will not be published. Required fields are marked *