Bisphosphonate medications have been used for more than 20 years to prevent osteoporotic fractures, based on strong evidence that these medications significantly reduce the risk of fracture within the first three to five years of treatment. However, the optimal duration of bisphosphonate therapy to minimize the risk of osteoporotic fractures isn't known. Researchers used data from Kaiser Permanente Northern and Southern California to determine whether continuing bisphosphonates for an additional two to five years further reduces the risk of hip fracture in older women.

For this retrospective study, the researchers identified women who had been treated with oral bisphosphonates for five years and were at least 60% adherent to therapy. Covariates-including body weight; smoking status; and age, race, and ethnicity-that may be associated with treatment or the risk of hip fracture were considered in the analysis. The study cohort included 29,685 women (median age, 71 years; 60% were non-Hispanic White).

The researchers used observational data to emulate the results of a five-year trial in which women in the study cohort would have been randomized to one of three regimens: (1) discontinued use of bisphosphonate medications at study entry (five years of use), (2) continued treatment for two additional years (seven years of use), and (3) continued treatment for five additional years (10 years of use). For the first two groups, the emulated trial protocol allowed for a six-month grace period for bisphosphonate discontinuation to occur.

During the five years of follow-up after study entry, 507 incident hip fractures were identified. The risk of hip fracture wasn't different for women who discontinued bisphosphonate treatment at study entry compared with those who continued for an additional five years. The results suggest there might be some benefit to continuing therapy for an additional two years, but not without the six-month grace period for the discontinuation of bisphosphonates.

The authors caution that their findings might not apply to older women at higher risk for osteoporotic hip fractures. Also, this study didn't address the optimal duration of therapy, and, as in any observational study, residual unmeasured confounding is possible.