Debating the Science of Dead Pop Stars

 If there’s a rock and roll heaven, sang the Righteous Brothers, you know they’ve got a hell of a band. Maybe — but the real question is, do they have a health plan? According to one study, “famous” pop stars are far more likely to die prematurely than the general population — often from either drug use or depression. That finding is objectionable to some rock fans, but their protests raise a different set of problems. Numbers can be misused or twisted, but they hold wisdom. In this case, they may have more to teach us than even the study’s authors realize.

The article, published in the Journal of Epidemiology and Community Health, is entitled “Elvis to Eminem: Quantifying the price of fame through early mortality of European and North American rock and pop stars.” It’s analytically sound. They define “fame” as having been “a solo performer or group member with a listed album” in Virgin’s list of “All Time Top 1000 Albums.” Maybe that’s not a perfect definition — but, in the words of economist and systems theorist Herbert Simon, it “satisfices.”

The authors give this reason for conducting the study: “While qualitative reviews support rock and pop stars suffering high levels of stress depression, and substance use, quantitative studies of mortality amongst such individuals are almost completely absent.” The piece goes on to describe their methodology in detail, and it looks reasonable.

Their findings are interesting, although they may seem obvious (or, as analysts prefer to say, “intuitive.”). American and European pop stars (including rock, rap, and R&B) are more likely to die young than their peers, when compared to populations with similar backgrounds. The authors use valid sample sizes, and their results are statistically significant.

Statistical validity is important, whether we’re discussing dead pop stars, cancer victims, or the frequency and size of hurricanes. Any one event — a death, a storm, an accident — can be caused by many things. But when the numbers reveal a pattern, there is an underlying story to be told.

Ideologues and industry spokespeople often challenge the idea of statistical significance, using the general public’s innumeracy to cast doubt on everything from global warming to carcinogens in the environment. But the “Elvis to Eminem” study brought objections from a different quarter. Music blogger Brad Laidman wrote a critical post called “I Want My Rock Stars Dead.” It was then distributed to the email list for Rock and Rap Confidential, Dave Marsh’s music magazine.

What were their objections to the study? First, Laidman goes after the methodology with a classic anti-science line of attack: “Their study sample is too small and 75 percent of it hasn’t managed to die yet.” That’s wrong, and here’s a simple response: Even if all of a sample group hasn’t died yet, some percentage of it has. If that percentage is greater than it is for the general population, in a statistically meaningful way, then something significant is taking place.