Bone Abstracts

Purpose: To determine the association between socio-economic status (SES) and risk of hip fracture.

Methods: Retrospective cohort study. We used a population database which contains primary care and hospital inpatient records of over >5 million people. Anyone registered in this database in 2009–2012 and resident in an urban area was eligible. Main measures: a validated SES composite index (accounting for proportion of unemployed, insufficiently educated, temporary workers, manual workers, and insufficiently educated youngsters) was estimated for each area based on census data. Outcome measure: hip fracture as coded (ICD-10) in 2009–2012. Statistics: Zero-inflated Poisson models were fitted to study the association between SES quintiles and hip fracture rates after adjustment for age, gender, obesity, smoking and alcohol drinking.

Results: Compared to the most deprived, wealthy areas had a older population (46.83 (18.49) vs 43.29 (17.59), mean years (S.D.)), had more women (54.8 vs 49.1%), as well as a lower percentage of obese (8.4 vs 16.2%), smokers (11.9 vs 16.9%) and alcoholics (1.3 vs 1.5%). The most affluent areas reported a higher incidence of hip fracture compared to the most deprived (Age–sex-adjusted incidence 38.57 (37.14–40.00) and 34.33 (32.90–35.76)/10 000 persons-year respectively). When compared to the wealthiest, deprived areas had lower hip fracture rates (unadjusted RR 0.71 (95% CI 0.65–0.78)), although age-gender (RR 0.90 (95% CI 0.85–0.95)) adjustment and further adjustment for obesity prevalence (RR 0.96 (95% CI 0.90–1.01)) significantly attenuated this association. Further adjustment for smoking and alcoholism did not make a difference (RR 0.96 (95% CI 0.91–1.02)).

Conclusion: Subjects living in wealthy areas had a 30% increased risk of hip fracture. Differences in age-gender composition, maybe due to a previously described higher mortality associated with deprivation, and a higher prevalence of obesity explain the observed risk reduction in these deprived areas. This information should be used for health-care planning and commissioning.