June 6, 2007 — The Pain of Obesity

A Duke University study of its work force reveals dramatic differences in job-related injuries and claims among its most obese workers
By Betty Joyce Nash

A Duke University Medical Center study finds that the school's heaviest employees filed twice the number of workers' compensation claims than its normal-weight employees. They also racked up seven times the medical costs from job-related injury and illness. Study results were published in the April 23 issue of the Archives of Internal Medicine.

Being fat carries risks for various ailments, including stroke, cancer, diabetes, asthma, and arthritis. It also causes economic pain: The obese are more likely to be unemployed or underemployed. Careful interpretation of the results of Duke's study may shed light on the consequences and causes of obesity.

Dr. Truls Ostbye, with co-authors John Dement and Katrina Krause, examined the records of more than 11,700 Duke University employees who received health risk appraisals between 1997 and 2004 for the study. A range of occupations were included: nurses, technicians, groundskeepers, professors, and secretaries. He found that extremely obese employees — those with a body mass index (BMI) greater than 40 — filed twice as many claims as those at the recommended weight. (A BMI of 18.5 to 24.9 is considered normal, while people between 25 and 29.9 are overweight and those greater than 30 are obese.)

Extremely obese workers also had 13 times more lost workdays than employees with a normal BMI. Their average medical claims per capita were $51,019 compared to $7,503 for the non-obese. Workers in some professions were more at risk for injury, regardless of their body shape. For instance, high claims rates among nurses and aides reflect patient lifting and repositioning, according to the study.

The Duke study adds to the concerns employers have about containing health care costs. Employers weigh benefits such as drugs, weight loss surgery, and wellness programs. Some have even offered per-pound cash incentives.

John Cawley, a Cornell University health economist, is also troubled by the study results and what it means for the health of the American work force. But he says there may be problems with causality. "If I've always had a bad back, I'm going to be more sedentary; it may not have been obesity that made me disabled but an underlying health problem."

Researchers also question the use of body mass index to accurately measure obesity. Since BMI is a weight-to-height ratio, it may yield a misleading result if a person has a lot of muscle or bone mass. A football player or bodybuilder, for example, could be classified as obese.

In a large epidemiological study such as Duke's, however, it is typical to use self-reported weight and height to determine people's fitness levels, Ostbye says. Also, the major increase in workers' comp claims is in the higher BMI levels, and that's unlikely due to muscularity. So BMI is probably a good indicator of obesity.

Nevertheless, social science research would benefit from better fatness measures, Cawley and Cornell economist Richard Burkhauser point out in a National Bureau of Economic Research working paper published in April. Researchers use BMI because it's quick, easy to compute, and reasonably accurate. But obesity rates vary when calculated using percent body fat, or total body fat divided by total mass.

For example, white men and black men have similar rates of obesity using BMI. However, when percent body fat is used, obesity rates among white men are higher. "These findings challenge the conventional BMI-based wisdom about who is obese and why," Cawley and Burkhauser write.

The number of obese adults in the United States has increased by more than 50 percent since the 1970s as measured by BMI, with extreme obesity growing at a much faster rate. People are taking in more calories and burning fewer. The Duke study suggests that the costs of these choices are also adding up.