Effect of diabetes and hypertension on obesity-related mortality

Background. Obesity is increasing and, along with it, greater mortality resulting from the overweight condition. Weight-reduction surgery is recommended for many obese patients based on a perceived risk of greater obesity-related mortality. However, many of the studies cited to justify this have aggregated patient groups together, making it difficult to apply their findings to individual patients. Methods. The,combined National Health Interview Survey database from the years 1986-1994 that has been linked to the National Death Index was analyzed. Patients were stratified for their body mass index (BMI) category, the presence or absence of diabetes and hypertension, gender, and race. The contribution of hypertension and diabetes to mortality was determined by Cox proportional hazards modeling. The absolute magnitude that the effect gender, race, BMI, hypertension, and diabetes had on mortality was determined by examination of stratified age adjusted death rates. Results. We analyzed 662,443 records from individuals interviewed between 1986 and 1994. Of these, 49,391 had died in the follow-up period (mean follow-up, 7.2 years; range, 1-14 years). Cox proportional hazards modeling-adjusting for age, race, and gender-revealed that hypertension (hazard ratio = 1.35 [95 % CI 1.34-1.35], P < .0001) and diabetes (hazard ratio = 2.29 [95 % CI 2.28-2.29], P < .0001) increased mortality independent of body weight. Serious obesity was associated with an increased mortality by Cox modeling (hazard ratio = 1.72 [95 % CI 1.71-1.73], P < .0001); however, when assessed by stratified age-adjusted mortality rates, this increase was numerically small because of the relatively low mortality rate for those without hypertension or diabetes. The presence of hypertension or diabetes substantially increased age-adjusted mortality rates. Much of the increased mortality observed in diabetic patients was attributable to complications of diabetes and not necessarily from atherosclerosis. Conclusion. Because weight-loss surgery successfully cures hypertension or diabetes in most obese patients, it should be considered for obese patients having these diseases. Based on the greatly elevated mortality associated with diabetes, surgery may be justified for obese diabetic patients with BMIs lower than the currently accepted criteria. In contrast, for the morbidly obese without diabetes or hypertension, mortality is only slightly increased from obesity alone. In terms of mortality, the benefits of weight reduction resulting from weight-loss surgery are less clear if there is no coexistent diabetes or hypertension.