Poor adherence to osteoporosis prevention drugs is a major problem among patients, with at least 50% not following the prescribed regimen and early discontinuation rates ranging from 50–78%, note the authors of a recent study. To examine this issue, researchers evaluated the real-world adherence (persistence) to osteoporosis prevention drugs in a total population of 5.8 million patients treated in primary care practices in Spain.1
Researchers compared the use of pharmacologic anti-osteoporosis agents with weekly alendronate, a first-line treatment. Other agents used in this population-based, retrospective, de-identified cohort included oral bisphosphonates, such as ibandronate and risedronate; strontium ranelate, which is not available in the U.S.; selective estrogen receptor modulators bazedoxifene and raloxifene; teriparatide; or denosumab. Eligible participants were women at least 50 years old who had initiated and purchased at least 56 daily doses of any of the agents between Jan. 1, 2012, and Dec. 31, 2012, and had medical data available for at least 12 months prior to that 2012 dispensing date (i.e., index date).
Each patient was studied from at least 12 months prior to the index date to the end of follow-up, which was defined as the end of data availability, treatment switching and/or treatment cessation or death. To account for potential confounders, multivariable Cox regression models were used with a list of pre-specified confounders, including age, gender, body mass index, smoking, alcohol consumption, comorbidity index, previous number of fractures (0, 1, 2 or ≥3), previous use of anti-osteoporosis medication(s), socioeconomic status and use of oral corticosteroids. Outcomes were two- and one-year persistence, with a permissible gap of up to 90 days.
A total of 19,253 women were included in the evaluation. The unadjusted two-year persistence ranged from 10.3% for strontium ranelate to 45.4% for denosumab. The lowest one-year persistence rate was 36%, for strontium ranelate, and the highest was 66%, for denosumab. The unadjusted two-year persistence ranged from 10.3% to 45.4%. Two-year persistence for all agents was 10.3% for strontium ranelate, 17.2% for risedronate, 18.9% for teriparatide, 19.2% for ibandronate, 25.4% for raloxifene, 26.8% for bazedoxifene, 28.9% for weekly alendronate and 45.4% for denosumab.
Although persistence rates were low, denosumab and teriparatide users had better persistence at the end of the first year, 65.8% and 57.6%, respectively, compared with weekly alendronate users at 47.8%. At the end of the second year, only denosumab users had better a persistence rate (45.4%) compared with weekly alendronate users, at 28.9%. A confounding variable that may have contributed to these results is disease severity—lower bone mineral density or fragility fractures—among users of denosumab and teriparatide, because these are not usually first-line agents and are also not available in an oral dosage. Confounding by indication may also have been present. Of the prespecified confounders, body mass index, socio-economic status, smoking and alcohol consumption all had missing information.