OP60 Interventions to address potentially inappropriate prescribing in primary care: a systematic review

Abstract

Background Potentially inappropriate prescribing (PIP) is the use of medicines that introduce a greater risk of adverse drug-related events where a safer, as effective alternative is available. PIP is common in community dwelling older adults with an estimated prevalence of between 20 and 50% and can contribute to increased morbidity, adverse drug events and hospitalisations. This systematic review aimed to assess the effectiveness of interventions designed to reduce PIP in primary care. Methods Following the PRISMA guidelines, PubMed, Embase, Scopus and the Cochrane library databases were searched (June 2014). Search terms included inappropriate prescribing, inappropriate pharmacotherapy, and primary care. Studies were included where: the population was community dwelling older patients (≥65); the intervention targeted PIP compared to usual care/other intervention; and the outcome was change in PIP measured with explicit (criterion-based, e.g. Beers) or implicit (judgement-based) tools such as the Medicines Appropriateness Index (MAI) which measures appropriateness across ten domains (e.g., nduration, cost). Three reviewers independently assessed studies for eligibility, extracted data, and assessed methodological quality using the Cochrane risk of bias tool. As meta-analysis of results was not possible owing to heterogeneity in interventions, a narrative synthesis was carried out. Results 15 studies were included. Baseline PIP prevalence ranged from 18% to 100%. Five intervention strategies were identified: pharmacist-led advice; multi-disciplinary team (MDT) meetings; multifaceted interventions (combining two or more techniques) computerised clinical decision support systems (CDSSs); and audit and feedback. Pharmacist-led advice in four RCTs was associated with improvement in PIP (mean MAI difference of 3.1). Significant improvement in diminishing PIP was also found in one out of three MDT studies, three out of four multifaceted interventions and the only audit and feedback study. Both CDSSs studies were effective in reducing new PIP, particularly in relation to amitriptyline, but not in the discontinuation of existing PIP. The methodological quality of included studies was often poor, particularly in reporting selection and attrition bias. There were small numbers of studies within each category, considerable heterogeneity in the types of interventions grouped together and PIP assessment used. Conclusion This systematic review found that pharmacist-led interventions appear effective in decreasing PIP in primary care, with other interventions such as MDTs, multifaceted approaches and CDSSs showing promise. However, the heterogeneity of the study interventions and the variable methodological quality mean that results require further confirmation in ongoing RCTs. Furthermore, the clinical significance of reductions in MAI score is difficult to determine. Better quality evidence upon which to base approaches to improving prescribing for these vulnerable patients is needed.