Hip fracture : risk factors and mortality

Abstract

The objective of this thesis was to identify factors related to risk of hip fracture and mortality after the hip fracture. We used data from a large population-based case-control study in postmenopausal women aged 50-81 years during 1993-1995 who resided in six counties in Sweden. The analysis was based on 1327 incident post-menopausal cases of hip fracture and 3262 randomly selected controls. Information on possible risk factors was collected by a comprehensive mailed questionnaire. Socioeconomic status and marital status were obtained by record linkage with census data. In addition, all subjects were followed up until December 1998 by the Inpatient and Cause-of Death Registers.

Adult weight change and height were the dominant body size risk factors. Current weight had protective effect independent of weight change only among the minority of women who did not gain weight as adults, whereas weight at age 18 appeared not to be a risk factor for hip fracture.

High leisure physical activity in the years before interview was inversely associated with risk of hip fracture. This decrease in risk was particularly pronounced in women who had gained weight during adult life. Occupational physical activity was not associated with hip fracture risk.

Decreased risk of hip fracture was found among women who were gainfully employed, who had high household income, and among women who were living in a one-family house as opposed to more crowded housing. Divorced, unmarried and widowed women had a higher risk of hip fracture than those who were married or cohabiting. However, occupational grouping and educational level were not associated with risk.

A strong association was found between the risk of hip fracture and duration of smoking, but there was no clear relation between the numbers of cigarettes smoked and fracture risk. The impact of smoking appeared to be reversible: 15 years after cessation there was no excess hip fracture risk. Alcohol consumption had a weak inverse association with risk.

Hip firacture'patients had a two fold higher mortality rate than control subjects after adjustment for factors associated with risk of hip fracture. The increased mortality did not vary by type of fracture (cervical vs. trochanteric). However, a greater relative risk of death was found among younger women and among those who did not gain weight as adult. The primary causes of death among women who died after hip fracture were cardiovascular disease and cancer.