Bone mineral density testing, and bisphosphonate and oestrogen prescribing associated with depot medroxyprogesterone acetate utilization-The impact oftheboxed warning.

Abstract

Depot medroxyprogesterone acetate (DMPA) prevents follicular maturation and ovulation, resulting in oestrogen deficiency. Oestrogen deficiency is particularly concerning as it is associated with decreases in bone mineral density (BMD) or bone mass. Widespread use of concurrent oestrogen or bisphosphonates with DMPA has been observed in a previous single-centre US study. The authors also reported that more than half of obstetrician-gynaecologists ordered BMD testing solely due to DMPA. The small sample size of the survey, potential for volunteer bias and recall bias limit the utility of this study. Given the study limitations, we sought to examine the trends of concurrent bone mineral density (BMD) testing, and bisphosphonate (BPA) and oral oestrogen-only (EST) prescribing during depot medroxyprogesterone (DMPA) use in a large administrative database. From IMS Health LifeLink™ database (2001-2013), we identified DMPA and combined hormonal contraceptive (CHC) users as women with no contraceptive claim 6months before and after the defined episode. We examined concurrent use as the proportion of BPA or EST claims and concurrent therapy days during contraception use. We also determined the proportion of users who had a concurrent BMD test by calendar year. We identified 203477 DMPA episodes associated with 600 BPA and 7060 EST claims and 1297806 CHC episodes associated with 2792 BPA and 145600 EST claims. Most concurrent BPA use overlapped with 10% or less of the DMPA episode, whereas most concurrent BPA use overlapped with >90%-100% of CHC episode. No difference was noted with concurrent EST use. Concurrent BMD testing was higher among DMPA users (0.68%) compared to CHC users (0.11%). Across calendar year, BMD testing peaked in 2004 and declined steadily to the baseline. Most DMPA concurrent BPA (69.3%) users had a BMD test. Our findings indicate the absence of BST or EST use solely due to DMPA initiation and present alternative explanations. Published research and publicity leading to the final implementation of the boxed warning in 2004 may explain observed trends for BMD testing. The majority of patients who had a BMD test ordered were also receiving bisphosphonates, suggesting high-risk fracture status or appropriate monitoring.