Mortality and Morbidity Liaison Committee--ISCS body mass index study.
OBJECTIVE: This study used the Impairment Study Capture System (ISCS) to examine the relationship between mortality and body mass index (BMI) in an insured population, particularly BMI in isolation from other risk factors. BACKGROUND: Large-scale studies of build in an insured population have traditionally been done on policies issued at standard premium rates. Insured mortality experience on elevated BMI is scarce or outdated. Increasingly competitive underwriting of build throughout the years has influenced what has been issued standard, and therefore, the relative experience of substandard to standard business. METHODS: We studied 241,966 policies submitted through the ISCS between 1989 and 2003 with actual height and weight and a code signifying abnormal build. Actual BMI were derived for these insureds. The average BMI was 35.0. Standardized mortality ratios (SMR) were computed using the 2001 Valuation Basic Table (VBT) as the expected basis. The average duration of exposure was 2.5 years. Results were stratified by underwriting factors of interest. RESULTS: Standardized mortality ratios rose quite modestly as BMI increased up until reaching severe obesity. Ratios for nonsmoker policies where elevated build was the only impairment saw SMR of 265% at BMI < 18.5, 130% at BMI 30.0-34.9, 160% at BMI 35.0-39.9 and 239% at BMI > or = 40.0. Ratios where other impairments were present tended to be higher in moderately obese ranges and lower at extremely obese ranges. No underwriting factor of significance impacted the pattern of ratios as BMI increased. CONCLUSIONS: While an average of 3 years may not be long enough to see the true manifestations of obesity, excess mortality is exhibited at low and high BMI ranges, especially when seen in isolation. The study is not a direct comparison of obese to non-obese subjects, but it is a point of reference for how obese insureds have fared vs standard issued policies. The relatively favorable experience may have more to do with the construction of the (standard) VBT table than any mitigating effect of modest obesity.