Lifestyle factors such as diet and physical activity vary in different social and income groups, and are known to be important influences on the incidence of osteoporotic fractures. Financial and social pressures are a common concern for older people. We set out to study the effect of socioeconomic deprivation on the incidence of fracture in older people and to compare the findings with those for younger groups. The All Wales Injury Surveillance System (AWISS) is a computerized system that collects injury data from most A&E departments throughout Wales. In this population-based study of 1.8 million people living in the 445 electoral tracts covered by AWISS in south, west, and northeastern Wales, we identified all 60,106 residents who presented with a fracture in 1999 and 2000. We linked details of their fracture with published Townsend deprivation scores for the electoral tract in which they were living and calculated fracture rates by fifths of deprivation. We observed the expected pattern of increasing fracture incidence in older age groups. Fracture incidence was significantly higher in electoral wards with poorer Townsend scores, resulting from a marked effect of socioeconomic deprivation on fracture incidence among younger adults with a rate ratio of 1.64 (95% CI, 1.57 to 1.72). This effect diminished with age, and was not observed in older age groups. At ages 85 and over the rate ratio was 0.94 (95% CI, 0.87 to 1.01). Socioeconomic factors clearly play a part in the causation of fracture in younger adults. Lifestyle influences are important in older age groups, but socioeconomic deprivation does not appear to be a risk factor for the development of osteoporotic fractures in elderly people.

Fracture prevention strategies will be most cost-effective if targeted on groups of frail elderly people who are at particularly high risk of falls and fractures. Elderly people living in care homes are one potential target population, but fracture incidence in this setting remains poorly defined in many countries. We used the All Wales Injury Surveillance System (AWISS) in a population-based study of people aged over 65 years living in the city of Cardiff. We linked a postcode-based register of all sheltered accommodation and all residential and nursing homes in the city with injury data from Cardiff's single Accident and Emergency Department. Cardiff has 47,700 residents aged over 65 years with 1918 (4.0%) living in residential or nursing homes and 1868 (3.9%) in sheltered accommodation. In 1999 we identified a total of 1305 fractures including 366 hip fractures, giving a crude fracture incidence of 27.4/1000 per year and 7.7/1000 per year, respectively. Care home residents suffered 213 fractures, 95 of which were of the hip, giving a crude fracture incidence of 111.1/1000 per year and 49.5/1000 per year, respectively. People living in sheltered accommodation suffered 94 fractures, including 28 at the hip, giving a crude fracture incidence of 51.4/1000 per year and 15.3/1000 per year, respectively. People in care homes and sheltered accommodation tend to be older than those living in the community, and we adjusted for this by calculating age- and gender-standardized relative ratios for each setting. Compared with the community dwelling population, care home residents had an overall fracture risk of 2.9 (95% CI 2.5-3.3) and a hip fracture risk of 3.3 (95% CI 2.6-4.2). People in sheltered accommodation had a total fracture risk of 1.7 (95% CI 1.4-2.1), and a hip fracture risk of 1.6 (95% CI 1.1-2.4). Such figures support the potential cost-effectiveness of strategies that seek to prevent fractures in care homes and sheltered accommodation, and are of special interest to those planning intervention studies in these settings.

No abstract