Background/Purpose: Many bisphosphonate-treated patients discontinue therapy within the first year. Some later restart the same or a different bisphosphonate. The implications of this on total healthcare costs are unclear. In a national cohort of postmenopausal veterans, we examined the relationship between bisphosphonate switch behaviors and total healthcare costs, defined as a sum of inpatient, outpatient, and pharmacy costs from the Department of Veterans Affairs (VA) Decision Support System dataset.

Methods: We identified bisphosphonate-treated female veterans ages ≥50 in the VA healthcare system in 2003-11 who lacked an alternative indication for bisphosphonate treatment. We classified patients as “high risk” if they had a prior fracture at the hip, wrist, spine, or proximal humerus; a baseline femoral neck bone mineral density (BMD) T-score ≤-2.5; or age ≥75 and lumbar spine, total hip, or 1/3 radius BMD T-score ≤-2.5. We used generalized estimating equation regression models with a gamma distribution to estimate the percentage difference in quarterly cost associated with bisphosphonate switch behaviors, classified quarterly as non-switching, switching, discontinuing (defined as a 90-day gap from the end of the prior supply), or restarting the initial bisphosphonate after a prior discontinuation or switch. Models were stratified on risk level and adjusted for demographic characteristics, comorbid conditions, other drug exposures in the prior year, and costs in the prior year; separate models were run for high- and non-high-risk patients to assess any differences due to disease severity.

Results: 36,280 patients met all eligibility criteria, including 5,567 high-risk and 30,713 non-high-risk patients. The mean (SD) age of the cohort was 65.7 (12.5). Race was known in 87.1%; of those 85.7% were Caucasian and 10.2% were Black. Over an average follow-up time of 4.3 years, 6.9% switched, 83.8% discontinued, and 38.3% reinitiated at least once. Compared to patients who continued with their initial bisphosphonate (N=5,691), patients with change behaviors had significantly higher total healthcare costs: switchers, discontinuers, and reinitiators had 15.1%, 5.0%, and 17.2% higher costs, respectively (all p values <0.001). Findings were similar in the high- and non-high-risk strata.

Conclusion: VA patients who switched or discontinued had significantly higher quarterly total healthcare costs compared to those who continued on their initial treatment, even if they later reinitiated their original bisphosphonate. While it is not certain if higher costs are directly attributable to the switch behavior, efforts to reduce bisphosphonate change behaviors may decrease healthcare costs in this population.

---

« Back to 2013 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/association-between-bisphosphonate-switching-behavior-and-cost-outcomes-in-postmenopausal-united-states-veterans/