Saturday, May 07, 2005
The Great Big Fat Hype Path to Policy
This WSJ OpinionJournal article discusses the general phenomena of how hype becomes expensive policy by treating the "Great Big Fat (or Obesity)" crisis as : " a case study in how public policy gets formulated in a highly advanced, highly educated and not least, highly neurotic society." As the article points out : "The day the munching died is March 9, 2004, when the Journal of the American Medical Association gave its imprimatur to a study by the U.S. Centers for Disease Control, announcing that "obesity" had caused 400,000 deaths in 2000, a whopping 33% increase from 1990. "
And with that "official imprimatur", those who would protect us from ourselves ( a real bipartisan cause!) moved into high gear :"Back in March, the great American campaign to control obesity was no longer an issue. It was a done deal. The government was onboard. The medical community was onboard. And of course the food industry and plaintiffs lawyers had launched parallel campaigns to cash in on fat people. But .....
It turned out that the CDC's arithmetic had "methodological flaws." After recombing the data dump, the CDC announced last month that the new number of obesity-related deaths annually is not 400,000 but . . . 26,000. Most intriguingly, though, the new study found that 86,000 "overweight" people lived longer than people of normal weight.
This is confusing--and that's the point. Science, of its nature, is always confusing. Medicine is uncertain. But public-policy formation in the U.S., especially as concerns health policy or the environment, whether obesity or the melting of the polar ice caps, admits to very little confusion. We claim to know. But in fact we usually don't know."
Importantly, The CDC's conclusions about mortality were not based on sound scientific theory or experimentation that could be replicated. Instead, it was based almost wholly on statistical associations. Some things simply may not lend themselves to sound scientific analysis, and good statistical analyses can yield useful insights. But non-replicable findings, based solely on statistical studies, should sound a warning bell telling us to be very cautious in moving to policy formulation. As the article points out, this new-found clarity on the numbers does not obviate the issue; it does allow us to discuss the issue in rational terms, rather than accept the hyped conclusion as religious belief beyond questioning.
The same caution applies to many fields of science- emotion based policy (global warming comes to mind here) : "Public officials will always ride in the slipstream of an evident crisis. But there is a cautionary tale here. The informational world we inhabit has become a volatile mixture of news, rumor and often incomplete science. This or that threat, need or cause comes at us constantly. But there may be a limit to how often politicians can lower a bucket into the well of public credibility, asking people to alter their behavior and pay handsomely for the privilege--as here, or climate change or fuel alternatives. There might not be much left when the authorities most clearly must ask people" to deal with a real threat.
The CDC's director called it's 400,000 obesity death miscalculation a "lesson in humility." It's a lesson more government agencies and "advocacy" agents should learn. And, even if they don't, the rest of us can learn to be a lot more skeptical of the next "Great Crisis" that is supported only by statistical association or simulation.
And with that "official imprimatur", those who would protect us from ourselves ( a real bipartisan cause!) moved into high gear :"Back in March, the great American campaign to control obesity was no longer an issue. It was a done deal. The government was onboard. The medical community was onboard. And of course the food industry and plaintiffs lawyers had launched parallel campaigns to cash in on fat people. But .....
It turned out that the CDC's arithmetic had "methodological flaws." After recombing the data dump, the CDC announced last month that the new number of obesity-related deaths annually is not 400,000 but . . . 26,000. Most intriguingly, though, the new study found that 86,000 "overweight" people lived longer than people of normal weight.
This is confusing--and that's the point. Science, of its nature, is always confusing. Medicine is uncertain. But public-policy formation in the U.S., especially as concerns health policy or the environment, whether obesity or the melting of the polar ice caps, admits to very little confusion. We claim to know. But in fact we usually don't know."
Importantly, The CDC's conclusions about mortality were not based on sound scientific theory or experimentation that could be replicated. Instead, it was based almost wholly on statistical associations. Some things simply may not lend themselves to sound scientific analysis, and good statistical analyses can yield useful insights. But non-replicable findings, based solely on statistical studies, should sound a warning bell telling us to be very cautious in moving to policy formulation. As the article points out, this new-found clarity on the numbers does not obviate the issue; it does allow us to discuss the issue in rational terms, rather than accept the hyped conclusion as religious belief beyond questioning.
The same caution applies to many fields of science- emotion based policy (global warming comes to mind here) : "Public officials will always ride in the slipstream of an evident crisis. But there is a cautionary tale here. The informational world we inhabit has become a volatile mixture of news, rumor and often incomplete science. This or that threat, need or cause comes at us constantly. But there may be a limit to how often politicians can lower a bucket into the well of public credibility, asking people to alter their behavior and pay handsomely for the privilege--as here, or climate change or fuel alternatives. There might not be much left when the authorities most clearly must ask people" to deal with a real threat.
The CDC's director called it's 400,000 obesity death miscalculation a "lesson in humility." It's a lesson more government agencies and "advocacy" agents should learn. And, even if they don't, the rest of us can learn to be a lot more skeptical of the next "Great Crisis" that is supported only by statistical association or simulation.
