Weight-bearing exercise and resistance training have been shown to increase femoral bone density. In addition, regular physical activity appears to reduce the occurrence of falls by strengthening muscles and improving balance. Feskanich and associates conducted a prospective 12-year study to determine the extent to which intensity and duration of walking and other leisure activities reduce the risk of hip fracture.

In 1986, questions about specific types of physical activity were added to the survey that participants in the Nurses' Health Study were asked to complete every two years. The study by Feskanich and colleagues included data from 61,200 postmenopausal registered nurses (40 to 77 years of age). None of the women had previous hip fracture, heart disease, stroke, osteoporosis, or cancer. The primary outcome measure was hip fracture resulting from low or moderate trauma (e.g., a fall from chair height, a slip on ice).

The study participants were asked how much time they spent, on average, engaging in one or more of seven common forms of physical activity: walking, jogging, running, bicycling, racquet sports, lap swimming, or aerobic activity. Walking was subdivided according to pace: easy (less than 2 miles per hour [mph]), average, brisk, and very brisk (4 mph or faster); “unable to walk” was also a category. Each activity was assigned a metabolic equivalent (MET) score, with 1 MET being the energy expended during quiet sitting. For each participant, scores for MET-hours per week were calculated from the reported activities. The calculations were then averaged from questionnaires sent to each woman every two years.

The median total activity for the study participants was seven MET-hours per week, which is equivalent to the energy expended while walking at an average pace for 2.3 hours per week. Minimal or no leisure-time physical activity was reported by 19 percent of the women (compared with 29 percent of the general U.S. population). Walking accounted for 66 percent of the total reported MET-hours per week.

Activity and body mass index (BMI) inversely and independently correlated with hip fracture risk. Women with 24 MET-hours per week had a 55 percent lower occurrence of hip fracture than those reporting three MET-hours per week. The risk of hip fracture declined by 6 percent for each MET-hour per week (dose-dependent reduction). Heavier women had a lower fracture risk in every activity category.

Hormone replacement therapy conferred the greatest risk reduction in women with the lowest level of activity (less than three MET-hours per week).Very active women (24 MET-hours per week or more) received no added benefit from hormone therapy.

When the impact of walking alone was analyzed, the authors found a dose-response risk reduction in hip fracture risk with a longer duration of walking. Walking for four hours a week or more was associated with a risk reduction of 41 percent. Furthermore, the brisker the pace, the greater the risk reduction.

An assessment of sitting or standing as measures of inactivity showed that standing for 10 hours a week or more also was associated with a significantly lower risk of hip fracture.

The authors concluded that physical activity derived from leisure-time activities or exercise is associated with a significantly reduced risk of hip fracture. A higher BMI also is associated with a reduced frequency of hip fracture, with the authors noting that increased activity could reduce the fracture risk even further in these women. Based on the authors' findings, hormone replacement therapy appears to protect inactive women, but not active women, against hip fracture risk. To maintain risk reduction, all exercise regimens must be maintained.