Survival of the fattest: unexpected findings about hyperglycaemia and obesity in a population based study of 75-year-olds

OBJECTIVE: To study the relationship between body mass index (BMI) and mortality among 75-year-olds with and without diabetes mellitus type 2 (DM) or impaired fasting glucose (IFG). DESIGN: Prospective population-based cohort study with a 10-year follow-up. PARTICIPANTS: A random sample of 618 of the 1100 inhabitants born in 1922 and living in the city of Västerås in 1997 were invited to participate in a cardiovascular health survey; 70% of those invited agreed to participate (432 individuals: 210 men, 222 women). OUTCOME MEASURES: All-cause and cardiovascular mortality. RESULTS: 163 of 432 (38%) participants died during the 10-year follow-up period. The prevalence of DM or IFG was 41% (35% among survivors, 48% among non-survivors). The prevalence of obesity/overweight/normal weight/underweight according to WHO definitions was 12/45/42/1% (14/43/42/1% among survivors, 9/47/42/2% among non-survivors). The hazard rate for death decreased by 10% for every kg/m(2) increase in BMI in individuals with DM/IFG (HR 0.91, 95% CI 0.86 to 0.97; p=0.003). After adjustment for sex, current smoking, diagnosed hypertension, diagnosed angina pectoris, previous myocardial infarction and previous stroke/transient ischaemic attack, the corresponding decrease in mortality was 9% (HR 0.92, 95% CI 0.86 to 0.99; p=0.017). These findings remained after exclusion of individuals with BMI<20 or those who died within 2-year follow-up. In individuals without DM/IFG, BMI had no effect on mortality (HR 1.01, 95% CI 0.95 to 1.07; p=0.811). The HR for BMI differed significantly between individuals with and without DM/IFG (p interaction=0.025). The increased all-cause mortality in individuals with DM/IFG in combination with lower BMI was driven by cardiovascular death. CONCLUSION: High all-cause and cardiovascular mortality was associated with lower BMI in 75-year-olds with DM/IFG but not in those without DM/IFG. Further studies on the combined effect of obesity/overweight and DM/IFG are needed in order to assess the appropriateness of current guideline recommendations for weight reduction in older people with DM/IFG.