STATS ARTICLES 2005
Steroids and Teenage Girls
June 21, 2005
A trend in the wrong direction or a study misunderstood?
Last week’s congressional hearing on steroid abuse threw the media a curveball: could there really be an upsurge in body-building among girls in grades nine through 12? Well, that’s what a 2003 state-by-state survey by the Centers for Disease Control (CDC) seemed to show, as 5.3 percent of the respondents reported using illegal steroids.
Instead of asking whether the survey had produced credible numbers, media reports found ways to explain why these numbers made sense. An editorial in today’s Baltimore Sun warned readers that it wasn’t just a spurt in teen athleticism (thanks to Title IX) driving steroid abuse, but the lure of reducing body fat. “For young women with eating disorders, steroids can become another risky substance that they might abuse, similar to diuretics, laxatives and amphetamines.”
The Sun admitted that researchers disagreed about how widespread this problem might be,
“with one Harvard psychiatry professor suggesting that anonymous surveys probably ‘exaggerated’ steroid use by young girls. But another expert from Penn State University said that based on the survey, there could be as many as 300,000 to 400,000 high-school-age girls using anabolic steroids. Whatever the actual number, the trend seems to be going in the wrong direction.”
Well, that depends on whether the survey’s methodology provides an accurate picture of abuse, doesn’t it? In his testimony to congress, Harvard’s Dr. Harrison Pope (a member of STATS advisory board) argued that the CDC survey was so full of potential “false positive responses” that “we cannot responsibly conclude there is currently a widespread public health problem of illegal anabolic steroid use among teenage girls, unless we see new, valid data to the contrary.”
The problem is that the CDC used an anonymous questionnaire to gauge abuse. This meant that it was impossible to go back to those girls who answered “yes” to the question “During your life, how many times have you taken steroid pills or shots without a doctor’s prescription” and ask whether they had specifically taken illegal anabolic steroids.
How many girls answered “yes” to taking steroids, even though they hadn’t actually taken anabolic steroids, the muscle-building variety under congressional scrutiny? Consider the possibility that some girls might have taken steroid skin creams, or asthma inhalers containing steroids, or birth control pills, or over-the-counter adrenal steroids, which were legal until last year. And how many girls took products they mistakenly thought contained steroids?
In explaining the inherent limitations of an anonymous survey, Dr. Pope then pointed to the kind of survey – a face-to-face confidential interview – that could tell us how widespread steroid abuse is among teenage girls. The most recent study to do this, the National Household Survey in 1994, found that just .02 percent of women aged between 15 and 44 had taken anabolic steroids.
Now one may object to that figure being out of date; but other anonymous surveys of anabolic steroid use (Monitoring the Future. 2004) found that only one percent of ninth-grade girls had used steroids in contrast to the CDC’s finding of 7.3 percent of ninth-grade girls. (It should be noted that the MTF survey unlike the CDC survey explained what a steroid was). Allowing for false positives in the MTF survey, the real figure is probably much closer to the National Household Survey. And the MTF survey hasn’t shown any dramatic increase in steroid use over the past decade.
Finally, and one of the reasons he was called to testify in front of congress, Pope is the author of the only recent peer-reviewed study looking at anabolic steroid use in women. Even after two years spent trawling among hard-core gyms in Eastern Massachusetts for subjects, Pope was only able to find 17 women who had used anabolic steroids – and none of them had begun using before the age of 18.
In other words, set against numerous studies and research on anabolic steroid use in women, the CDC came up with a statistic that was anomalous. And instead of being skeptical about how the CDC arrived at this figure, the Baltimore Sun downplayed its critics.
The moral here is that journalists forget the adage “even if your mother says she loves you, check it out” at the public’s peril. This story should not have been covered as an alarm followed by a lone expert protesting in the penultimate paragraph – as it was in some publications last week; nor should it have been treated as a ho-hum case of “experts” in disagreement: the CDC’s study was either a reliable indication of steroid abuse among teenage girls or it wasn’t. And given the scientific and statistical criteria for evaluating such questions, the conclusion was clear from Pope’s testimony: it wasn’t.