Context: Low bone mineral density (BMD) has been associated with increased all-cause mortality. Cause-specific mortality studies have been controversial.
Objective: The aim of the study was to investigate associations between BMD and all-cause mortality and in-depth cause-specific mortality.
Design and setting: We studied two cohorts from the prospective Rotterdam Study (RS), initiated in 1990 (RS-I) and 2000 (RS-II) with average follow-up of 17.1 (RS-I) and 10.2 (RS-II) years until January 2011. Baseline femoral neck BMD was analyzed in SD values. Deaths were classified according to International Classification of Diseases into seven groups: cardiovascular diseases, cancer, infections, external, dementia, chronic lung diseases, and other causes. Gender-stratified Cox and competing-risks models were adjusted for age, body mass index, and smoking.
Participants: The study included 5779 subjects from RS-I and 2055 from RS-II.
Main outcome measurements: We measured all-cause and cause-specific mortality.
Results: A significant inverse association between BMD and all-cause mortality was found in males [expressed as hazard ratio (95% confidence interval)]: RS-I, 1.07 (1.01-1.13), P = .020; RS-II, 1.31 (1.12-1.55), P = .001); but it was not found in females: RS-I, 1.05 (0.99-1.11), P = .098; RS-II, 0.91 (0.74-1.12), P = .362. An inverse association with chronic lung disease mortality was found in males [RS-I, 1.75 (1.34-2.29), P < .001; RS-II, 2.15 (1.05-4.42), P = .037] and in RS-I in females [1.72 (1.16-2.57); P = .008], persisting after multiple adjustments and excluding prevalent chronic obstructive pulmonary disease. A positive association between BMD and cancer mortality was detected in females in RS-I [0.89 (0.80-0.99); P = .043]. No association was found with cardiovascular mortality.
Conclusions: BMD is inversely associated with mortality. The strong association of BMD with chronic lung disease mortality is a novel finding that needs further analysis to clarify underlying mechanisms.