The Association Between Older Age, Co-Morbidity, and Treatment Status of Incident Osteoporotic Fractures: A Population-Based Nested Cohort Study

Abstract

BACKGROUND: Despite strong evidence-based rationale for both the primary and secondary prevention of osteoporosis, there remains an overall low prevalence of osteoporosis treatment in older adults. Furthermore, there is some question whether low treatment rates in older adults are simply age related variations (in treatments) or due to the presence of co-morbid conditions. Therefore, we sought to examine the association between older age, co-morbidity, and the use of osteoporosis medications following an incident osteoporosis related fracture.

METHODS: We performed a retrospective nested cohort analysis using de-identified administrative healthcare data from the province of British Columbia, Canada (pop. 4.1 million). We included patients 65 years and older, who had continuous enrollment in the provinces’ prescription drug plan, with a study-defined osteoporosis-related fracture during the study period of April 1, 1999 to March 31, 2002. A multivariate logistic regression model was used to examine the association between the dependent variable - osteoporosis medication dispensation within six months of index fracture and the predictor variables - age, sex, co-morbidity, fracture site, year of fracture, health region, and osteoporosis treatment prior to the index fracture.

RESULTS: After exclusion criteria were applied, we identified 11,870 consecutive patients who had been hospitalized with 12,025 incident (study-defined) fractures during the study period. The mean age of the sample was 81.1 years (SD 7.7; range 65–104 years), and 74% of the subjects were women. The majority of patients (99%) sustained one fracture (range 1 to 4); the fractures were predominately of the hip (63%) followed by fractures of the wrist (17%), pelvis (9%), vertebra (7%), and ribs (4%). The majority of subjects had no co-morbid conditions or only one (63%); 31% had two to three co-morbidities, and 6% had four or more co-morbid conditions. Overall, there was a low rate of osteoporosis treatment before the incident fracture (15% treatment); this rate improved to 19% at six months post fracture. Those receiving treatment after the index fracture were significantly younger, more often female, and had fewer co-morbid conditions (P < .001). The use of an osteoporosis medication prior to the index fracture was the strongest predictor of post-fracture treatment (adjusted OR = 15.89; 95% CI = 9.69–26.04). Increasing age, more than one co-morbid condition, and male sex were all associated with a significant decrease in the likelihood of dispensing osteoporosis drugs when compared to younger and healthier women.

CONCLUSION: Despite the wide availability of osteoporosis medications, our findings suggest that the majority of older adults, many of who have at least one co-morbid condition, are not receiving treatment to prevent the progression of the disease and to prevent further fractures.