Background Osteoporotic fractures are a cause of significant mortality and morbidity, with huge personal and societal costs. Anthropometric factors such as BMI, and lifestyle factors such as smoking and excess alcohol consumption are potentially modifiable and may impact fracture risk. We considered associations between relationship status and incident fracture in the Hertfordshire Cohort Study, a community-dwelling cohort of older adults. We also explored associations between anthropometric, socioeconomic and lifestyle factors relevant to bone health and relationship status.

Methods 2,997 participants (1,418 women) completed a baseline questionnaire and clinic visit to ascertain their socioeconomic, anthropometric and lifestyle factors. Participants were followed up for a mean of 15 years using UK Hospital Episode Statistics, that report clinical outcomes using ICD-10 coding; ICD-10 diagnosis codes were used to ascertain incident fractures. Relationship status (single vs currently married/cohabiting) at baseline was examined in relation to incident fracture using time-to-first-event Cox regression with adjustment for age. Baseline characteristics in relation to current relationship status were examined using logistic regression. Analyses were stratified by sex.

Results At baseline, mean (SD) age of participants was 66.2 (2.8) years; 14% of men and 27% of women were currently single. Over follow-up, 15% of women sustained a fracture and 5% sustained a hip fracture; corresponding figures in men were 7% and 2%. Single women had increased risk of incident fracture compared to women who were married/cohabiting (age-adjusted hazard ratio [95% CI]: 1.36 [1.02, 1.81], p=0.034). Compared to their married/cohabiting counterparts, single women had: higher mean age [67.1 vs 66.5 years, p<0.001]; higher mean BMI [28.1 vs 27.4 kg/m², p=0.016]; greater prevalence of current smoking [14% vs 8%, p=0.004]; and were less physically active (p=0.003). Single women were more likely to have left school before the age of 15 years (p=0.037) and to not owner-occupy their home (p<0.001). We did not observe significant differences in fracture risk according to relationship status in men (age-adjusted hazard ratio for single men compared to those married/cohabiting: 1.56 [0.97, 2.51], p=0.068) or among either sex regarding risk of hip fracture.

Conclusion These results highlight the differing risk of fracture and modifiable fracture risk factors among single women relative to their married/cohabiting counterparts. Such associations between marital status and fracture were not identified in men.