(P-427) Use of and Adherence to Bisphosphonates in Elderly Patients with Complex Health Needs: An Analysis of UK Primary Care Records

Background: As the UK population ages, more preventative treatments for common and chronic diseases are prescribed. However, the utilisation of these medications in old people with complex health needs is not well-understood.

Objectives: We aimed to assess the population and patient-level utilisation of oral bisphosphonates (BP) in older people with complex health needs using three different definitions.

Methods: All participants aged >65 on 01/01/2010, registered for >=1 year were identified from a UK primary care database (CPRD GOLD) with linkage to Hospital Episode Statistics and Office for National Statistics death data. Patients were followed up until the earliest of patient leaving the practice, practice leaving the database, data extraction and death. Three cohorts were identified using different indicators of complex health needs based on records in the previous year: 1) hospitalisation: >=1 unplanned hospitalisation, 2) frailty: electronic frailty index (eFI) ≥3, and 3) polypharmacy: number of medications prescribed ≥10. In addition, an overlap group was created comprising patients who met the criteria of all three cohorts.
For each cohort, we assessed the population-level incidence of BP use and patient-level compliance of new BP users in the first year after start date. Treatment episodes were defined as continuous drug use, allowing for a maximum 90-day gap between prescriptions and adding a 90 days washout period at the end of each episode. To assess adherence, duration (median, IQR) and medication possession ratio (MPR) - calculated as defined daily dose/duration - in BP users were estimated.

Results: There were 90597, 110225 and 116076 patients included in the hospitalisation, frailty and polypharmacy cohort, respectively, with 28259 patients belonging to all three cohorts. Of those, 2294, 2527, 2576 and 805 initiated BP in the first year, respectively. Median duration was 420 days (IQR 172, 1113) in the hospitalisation cohort, 413 days (167, 1012) in the frailty cohort, 406 days (167, 1045) in the polypharmacy cohort, and 371 days (152, 962) in the overlap group. Median MPR was 71.9% (38.4%, 90.9%), 70.4% (38.4%, 89.3%), 70.4% (38.4%, 90.0%) and 68.1% (38.4%, 89.9%) for the three cohorts and the overlap group, respectively.

Conclusions: BP are commonly used amongst frail and elderly people in the UK. Treatment duration is however low and well shorter than the necessary 3 or more years to achieve fracture protection. Further research into the safety and effectiveness of BP in these conditions is needed.