Sex- and age-specific associations between income and incident major osteoporotic fractures in Canadian men and women: a population-based analysis

We investigated sex- and age-specific associations between income and fractures at the hip, humerus, spine, and forearm in adults aged a parts per thousand yen50 years. Compared to men with the highest income, men with the lowest income had an increased fracture risk at all skeletal sites. These associations were attenuated in women. Associations between income and hip fractures are contested, even less is known about other fracture sites. We investigated sex- and age-specific associations between income and major osteoporotic fractures (MOF) at the hip, humerus, spine, and forearm. Incident fractures were identified from administrative health data for adults aged a parts per thousand yen50 years in Manitoba, Canada, 2000-2007. Mean neighborhood (postal code area) annual household incomes were extracted from 2006 census files and categorized into quintiles. We calculated age-adjusted and age-specific sex-stratified fracture incidence across income quintiles. We estimated relative risks (RR) and 95 % CI for income quintile 1 (Q1, lowest income) vs. income quintile 5 (Q5) and tested the linear trend across quintiles. We identified 15,094 incident fractures (4736 hip, 3012 humerus, 1979 spine, and 5367 forearm) in 2718 men and 6786 women. For males, the RR of fracture for the lowest vs. highest income quintile was 1.63 (95 % CI 1.42-1.87) and the negative trend was statistically significant (p < 0.0001); individual skeletal sites showed similar associations. For females, the RR of fracture for the lowest vs. highest income quintile was 1.14 (95 % CI 1.01-1.28), with a statistically significant negative trend (p = 0.0291); however, the only skeletal site associated with income in women was the forearm (Q1 vs. Q5 RR 1.09, 95 % CI 1.01-1.28). Compared to men with the highest income, men with the lowest income had an increased fracture risk at all skeletal sites. These associations were attenuated in women. For men, these effect sizes seem large enough to warrant public health concern.