Comorbidities, bone-sparing agent prescription history and their determinants among postmenopausal women in UK primary care settings: a retrospective database study

M. Feudjo-Tepie, S. Ferguson, A. Roddam, A. Taylor, J. Bayly, C. Critchlow

Distinguishing oral bisphosphonates from other bone-sparing therapies, this retrospective observational study, first, characterized treated osteoporosis patients in the UK, and secondly, explored factors associated with the risk of discontinuation or switching between therapies. The latter should be considered when evaluating real-world data. This retrospective observational study evaluated the characteristics of women with postmenopausal osteoporosis, including comorbidities and determinants of treatment patterns with bone-sparing agents. The UK Clinical Practice Research Datalink was used to identify postmenopausal women (aged ≥50 years) treated with a bone-sparing agent or diagnosed with osteoporosis between 1 January 1993 and 31 December 2008. Two non-mutually-exclusive subpopulations were defined: (1) patients active in the database on 31 December 2008; (2) patients treated with a bone-sparing agent since 1 January 1993. Subpopulation 1 was used to describe patient comorbidities and osteoporosis treatment history, and subpopulation 2 was used to explore the characteristics associated with bone-sparing treatment patterns use via multivariable regression for repeated multinomial responses. A total of 62,657 individuals met the inclusion criteria; subpopulation 1 comprised 38,469 women (61.4 %), of whom 21,687 received a bone-sparing agent in 2008 (99.7 % oral bisphosphonates and the remainder other agents). Those receiving other agents were more likely to have had previous treatment with bone-sparing agents, to have experienced previous fractures, and to have visited their doctor more frequently. Analyses also identified several comorbidities associated with an increased risk of discontinuation of bone-sparing agents, including heart disease, gastrointestinal disease, and renal failure. Anticonvulsant use was associated with a dramatic increase in the risk of switching. Several patient characteristics were associated with discontinuation of, or switching between, bone-sparing treatments. Patients receiving bone-sparing medication other than oral bisphosphonates were more likely to have comorbid conditions and a history of fracture and to have taken an oral bisphosphonate previously.