Longitudinal patterns of potentially inappropriate prescribing in early old-aged people

Abstract

Purpose: It is contentious whether potentially inappropriate prescribing (PIP) is predominantly a phenomenon of late life or whether it has its origins in early old age. This study examined the pattern of PIP in an early old-aged population over 5 years. Methods: Secondary data analysis of a population-based primary care cohort, of patients aged 60–74 years. Medication data were extracted from electronic patient records in addition to information on comorbidities and demographics. Explicit START criteria (PPOs) and STOPP criteria (PIMs) were used to identify PIP. Generalised estimating equations were used to describe trends in PIP over time and adjusted for age, gender and number of medicines. Results: A total of 978 participants (47.8%) aged 60–74 years were included from the cohort. At baseline, PPOs were detected in 31.2% of patients and PIMs were identified in 35.6% at baseline. Prevalence of PPOs and PIMs increased significantly over time (OR 1.08, 95% CI 1.07; 1.09 and OR 1.04, 95% CI 1.0; 1.06, respectively). A higher number of medicines and new diagnoses were associated with the increasing trend in both PPO and PIM prevalence observed over time, independent of PPOs and PIMs triggered by drug combinations. Conclusions: Potentially inappropriate prescribing is highly prevalent among early old-aged people in primary care and increases as they progress to more advanced old age, suggesting that routine application of STOPP/START criteria in this population would significantly improve medication appropriateness.