Urban-Rural Differences in Hip Fracture Mortality: A Nationwide NOREPOS Study

Abstract

Higher hip fracture incidence in urban than in rural areas has been demonstrated, but urban-rural differences in posthip fracture mortality have been less investigated, and the results are disparate. Hence, the aims of the present register-based cohort study were to examine possible urban-rural differences in short- and long-term mortality in Norwegian hip fracture patients and their potential associations with sociodemographic variables, and to investigate possible urban-rural differences in excess mortality in hip fracture patients compared with the general population. Data were provided from the NOREPOS hip fracture database, the 2001 Population and Housing Census, and the National Registry. The urbanization degree in each municipality was determined by the proportion of inhabitants living in densely populated areas (rural: <1/3, semirural: 1/3 to 2/3, and urban: >2/3). Age-adjusted mortality rates and standardized mortality ratios were calculated for hip fracture patients living in rural, semirural, and urban municipalities. A flexible parametric model was used to estimate age-adjusted average and time-varying HRs by category of urbanization with the rural category as reference. Among 96,693 hip fracture patients, urban residents had higher mortality than their rural-dwelling counterparts. The HR of mortality in urban compared with rural areas peaked during the first 1 to 2 years postfracture with a maximum HR of 1.20 (95% CI, 1.10 to 1.30) in men and 1.15 (95% CI, 1.08 to 1.21) in women. The differences were significant during approximately 5 years after fracture. Adjusting for sociodemographic variables did not substantially change the results. However, absolute 30-day mortality was not significantly different between urban and rural residents, suggesting that health-care quality immediately postfracture does not vary by urbanization. The novel findings of a higher long-term mortality in urban hip fracture patients might reflect disparities in health status or lifestyle, differences in posthip fracture health care or rehabilitation, or a combination of several factors. © 2019 The Authors. JBMR Plus published by Wiley Periodicals, Inc. on behalf of American Society for Bone and Mineral Research.

Keywords: AGING; EPIDEMIOLOGY; GENERAL POPULATION STUDIES; OSTEOPOROSIS; STATISTICAL METHODS.

Figure 1

Age‐adjusted mortality rates with 95% CIs in hip fracture patients, by sex, urbanization category, and time after hip fracture. Norwegian hip fracture patients aged 50 to 100 years. The NOREPOS Hip Fracture Database (2002 to 2013).

Figure 2

Predicted relative mortality in patients living in urban compared with rural municipalities by time after hip fracture in (A) men and (B) women. Two points in time during follow‐up and their corresponding HRs with 95% CIs are specified with reference lines: (1) when the HR reaches maximum, and (2) when the HR is no longer significant (when the lower confidence interval intersects the line corresponding to a HR of 1). Norwegian hip fracture patients aged 50 to 100 years. The NOREPOS Hip Fracture Database (2002 to 2013).

Figure 3

Standardized mortality ratios (SMRs) by category of urbanization with 95% CIs in hip fracture patients (men and women aged 50 to 100 years from the NOREPOS Hip Fracture Database, 2002 to 2013), relative to the expected sex, age, calendar year, and urbanization category‐specific mortality in the general population without hip fracture.