Lower Mortality Associated With Overweight in the U.S. National Health Interview Survey: Is Overweight Protective?

It is still debatable whether overweight has protective or detrimental effects on survival. The focus of the ongoing debate is on possible confounding bias due to factors such as preexisting illness and smoking. We aimed to assess the association between overweight and mortality and to examine confounding effects of various factors including smoking and preexisting cancer, cardiovascular disease (CVD), diabetes, asthma, bronchitis, and kidney disease on the overweight–mortality association in adults. The data were extracted from the public-use National Health Interview Survey (NHIS) 1997 to 2009. Mortality data up to December 31, 2011 were linked to 131,813 with normal weight and 120,217 overweight adults. We assessed the association between overweight and mortality using Cox proportional hazard model with adjustments for various sets of confounding factors—age, sex, smoking, race, survey year, diabetes, CVD, cancer, asthma, bronchitis, and kidney disease. During the period from the original surveys to December 31, 2011, 22,513 (11,815 normal weight and 10,698 overweight) adults died. Normal weight and overweight groups differed in the characteristics of age, sex, smoking, and preexisting diseases. After adjusting for age and sex, the risk of dying was lower for overweight than normal weight adults (hazard ratio [HR], 0.82; 95% confidence interval [CI]: 0.80–0.85). Lower mortality risk associated with overweight remained after further adjusting for smoking and preexisting diseases such as diabetes, CVD, cancer, asthma, bronchitis, and kidney disease (HR, 0.80; 95% CI: 0.78–0.82). We observed a similar pattern for men and women, and for those free from preexisting diabetes, hypertension, and CVD. In conclusion, overweight adults have a lower mortality risk than normal weight adults. Our findings do not support that the lower mortality in overweight adults is due to confounding effects of smoking and preexisting diseases.