The Association Between Dementia Status, Co-morbidity, and Osteoporosis Treatment: A Population-Based Nested Case-Control Study

Abstract

BACKGROUND: Increasing age and a diagnosis of dementia both dramatically increase the risk of serious osteoporosis related sequela. At the same time, there remains an overall low rate of osteoporosis treatment particularly in older, frail adults despite the availability of effective antiresorptive treatments such as bisphosphonate drugs. In addition, the frequency in which community dwelling persons with dementia are treated with osteoporosis medications has not been well described. Furthermore, existing literature does not adequately delineate whether the low treatment rates are simply age related variations (in treatments) or due to the presence of co-morbid conditions, particularly dementia.

METHODS: We performed a retrospective population based nested case-control study using de-identified administrative healthcare data from a Canadian province (pop. 4.1 million). We included patients 65 years and older, with a diagnosis of osteoporosis, who had continuous prescription drug coverage during the study period of 1991 to 2007. A multivariate logistic regression model was assembled to examine the relationship between osteoporosis medication dispensation and dementia status while controlling for age, sex, co-morbidity, and residence.

RESULTS: We included 39,452 patients in the osteoporosis cohort; the mean age of the sample was 80.1 years (SD 7.5; range 65–104 years), 79% of the subjects were female, and 34% had a dementia diagnosis. Only 5% of the sample had no co-morbid conditions; the majority of patients (52%) had at least 3 co-morbid conditions (SD 2.0; range 0–12 conditions). When stratified by dementia status, there were significant differences in age, frequency of co-morbidity, and residence by health region (P < .001) and sex (P < .05). Almost half of the total osteoporosis cohort were dispensed an osteoporosis medication during the study period (43%; P < .001). Those who had been dispensed drug treatment were more often younger, female, and had no diagnosis of dementia (P < .001). Drug dispensation was directly related to the frequency of co-morbid conditions; those with 4 or more conditions were dispensed treatment significantly more often (54%) than those with fewer co-morbid conditions (P < .001). A diagnosis of dementia was a significant negative predictor of drug dispensation (adjusted OR = 0.55; 95% CI = 0.44–0.69). Increasing age, male sex, and a more remote residence were all associated with a significant decrease in the likelihood of treatment. Increasing co-morbidity was significantly associated with receiving treatment (adjusted OR = 3.30; 95% CI = 2.88–3.78).

CONCLUSION: Despite the wide availability of osteoporosis medications, our findings suggest that the majority of older adults with a diagnosis of dementia, but not necessarily fewer co-morbid conditions, are not receiving treatment to prevent progression of the disease including fragility fractures.