Social and regional inequalities in the incidence of and outcomes after hip fracture in England

Abstract

Background

Hip fracture risk varies by levels of deprivation and geography; however, it is not known whether social inequalities in hip fracture incidence have changed over time in England. Furthermore, the impact of deprivation on clinical outcomes after hip fracture is not established. I examined the effect of area-level social deprivation on hip fracture incidence and outcomes among older men and women in England.

Methods

In the incidence study, hip fractures were identified using Hospital Episode Statistics (HES) (2001/02-2014/15) and Office for National Statistics (ONS) mid-year population estimates (2001-2014). The Index of Multiple Deprivation (IMD) was used to measure area-based deprivation (Q1 -least deprived; Q5 - most deprived). Age-adjusted incidence rate ratios (IRR)were calculated, stratified by gender, deprivation quintiles and region.

In the outcomes study, hip fractures were identified using HES data linked to the National Hip Fracture Database (04/2011-03/2015) and ONS mortality data. Deprivation was measured using the IMD. Associations between deprivation and 30-day mortality and emergency 30-day readmission were examined, adjusted for age, gender and comorbidity. Mean length of stay (LOS) in NHS acute and rehabilitation hospitals (‘superspell’) and total NHS bed occupancy within 1-year post-fracture were calculated.

Results

Over 14 years, 747,369 index hip fracture admissions were identified. Incidence was higher in more deprived areas, particularly among men: IRR Q5 vs. Q1 1.50 [95% CI 1.48,1.52] in men, 1.17 [1.16,1.18] in women. Age-standardised incidence increased for men across all deprivation quintiles from 2001-2014. Among women, incidence fell more amongst those least deprived compared to most deprived (year by deprivation interaction p<0.001). Age-standardised hip fracture incidence was highest in the most deprived areas in the North of England compared to the Midlands and the South for both women and men. 

Over 4 years, 218,907 index hip fracture admissions were identified. Greater deprivation was associated with higher 30-day mortality (adjusted OR 1.23 [1.17,1.30], p><0.001). Among survivors, mean superspell LOS was longer in the most deprived versus least deprived quintile (Q5: 24.4 [SD 21.7] days, Q1: 23.3 [22.1], p><0.001). Emergency 30-day readmission was higher in those most deprived compared to least deprived (adjusted OR 1.27 [1.22,1.32], p><0.001). A similar trend was observed when assessing mean total NHS bed occupancy.

Conclusions

Deprivation is a stronger relative predictor of hip fracture incidence in men than women. Absolute inequalities in hip fracture incidence are greatest in the North of England. Furthermore, increasing deprivation is associated with higher 30-day mortality and, among those who survive, greater healthcare utilisation after hip fracture. These study findings highlight the need for greater focus on addressing social inequalities in hip fracture care in England.