Professors live longer than doctors: Immortality bias in survival analysis (Editorial)

Oscar winners live almost 4 yr longer than less successful film stars. Although criticized, this sort of finding still features frequently on the Internet and public media and also in medical literature.

In 2001, an article was published based on 235 Oscar winners, 527 nominees (non-winners), and 887 performers who were never nominated (controls).1 Controls were selected from performers who were in the same film as the nominees and were also the same sex and approximately the same age as the nominees. In the primary analysis, survival was measured from performers’ date of birth. Each performer was classified as a winner or non-winner from the outset. One reported analysis used winner as a time-dependent covariate to reflect the fact that all started out as non-winners but that some changed status over time. A basic analysis showed that the total life span of winners was 3.9 yr longer than controls and this was statistically significant. A slightly more sophisticated analysis that accounted for the dynamics of moving from non-winner to winner status found less dramatic differences.