Skip Navigation LinksHome > January 2011 - Volume 22 - Issue 1 > Body Mass Index and Mortality: Is the Association Explained...
Epidemiology:
doi: 10.1097/EDE.0b013e3181fce2a2
Genetics: Original Article

Body Mass Index and Mortality: Is the Association Explained by Genetic Factors?

Carlsson, Sofiaa; Andersson, Tomasa; de Faire, Ulfb,c; Lichtenstein, Pauld; Michaëlsson, Karle,f; Ahlbom, Andersa

Supplemental Author Material
Collapse Box

Abstract

Background: Numerous studies have shown that higher body mass index (BMI) is associated with higher mortality. We investigated the extent to which this association might be explained by genetic factors.

Methods: We used data from the Swedish Twin Registry on twins born 1886–1958 who answered a questionnaire in 1969/1970 or 1972 (n = 44,258). Information on mortality from all-causes (n = 14,217), cardiovascular disease (CVD; n = 9009), and coronary heart disease (CHD; n = 3564) was obtained by linkage to the national Causes of Death Registry for the years 1972–2004. The association between BMI and mortality was studied without control for genetic factors in cohort analyses and with control for genetic factors in co-twin control analyses.

Results: In cohort analyses, there was a clear dose-response relationship between BMI and mortality. Hazard ratios per 1 unit increase in BMI in subjects with BMI ≥18.5 were 1.05 (95% confidence interval = 1.05–1.06) for all-cause mortality, 1.07 (1.07–1.09) for CVD mortality, and 1.09 (1.08–1.10) for CHD mortality. Similar results were seen in co-twin control analyses of dizygotic twins. However, within monozygotic twins, BMI was associated with death from CHD (OR = 1.06; 1.00–1.12), whereas the association with all-cause mortality (1.01, 0.98–1.04) and CVD mortality (1.02, 0.98–1.06) was weak.

Conclusions: Our findings indicate that there is an association between high BMI and mortality from CHD that is not explained by genetic confounding. However, a large part of the association between BMI and other causes of death may be explained by genes rather than by a causal link between these factors.

© 2011 Lippincott Williams & Wilkins, Inc.

Twitter  Facebook

Login

Article Tools

Share