breast and prostate cancer screening: useless?

Gerd Gigerenzer, a famous German psychologist and since 1997 director of the Center for Adaptive Behavior and Cognition in Berlin, recently published a study on breast and prostate cancer screening.

The astonishing results:

• Without screenings, 5 out of 1.000 women die of breast cancer over the age of 50. If women go to preventive medical checkups regularly, the number decreases only to 4 out of 1.000.
• For men, there are several studies from the US and Europe showing that prostate cancer screenings don’t have any effects on mortality.
• Most people dramatically overestimate the usefulness of screenings (by the factor of ~ 50) – so did I. Only 0.8% have an adequate estimate of screening effects, and I guess those were depressed patients …
• Around 125 out of 1.000 women get a false positive breast cancer diagnosis (they’re told that they probably have cancer, which is actually not the case), and tissue is extracted for further testing. Around 4 out of 1.000 women undergo a surgery due to a wrong diagnosis.

Gigerenzer states that he doesn’t want to criticise preventive checkups in general, but that people need to be better informed about the (non-existent …) effects of screenings.

My opinion:

• This probably was a cross-section design, asking men and women over 50 whether they had had screenings in the past, and correlating these findings with the current states of cancer (yes, no).
• Apart from a extremely costly longitudinal study over the course of 20 years, that’s a proper way to conduct a study like this, as far as I’m concerned.
• BUT there might be a major point of criticism here: the quality of retrospective examinations. The screenings the study talks about probably took place in the last 20-40 years, and I’d argue that there is at least a chance that the reliability of cancer screenings increased within that period of time. If so, Gigerenzer must not draw conclusions from a study examining past preventive medical checkups to current screenings.