Interventions to reduce or stop inappropriate drug prescription: An overview of systematic reviews

Article type
Authors
Steutel N1, Vermeulen H2, Tabbers M3, Benninga M3, Langendam M1
1Department of Clinical Epidemiology, Bioinformatics and Biostatistics, Academic Medical Center
2IQ Healthcare, Radboud Institute for Health Sciences, Radboud University Medical Center
3Department of Pediatric Gastroenterology and Nutrition, Academic Medical Center
Abstract
Background: Previous studies showed that the existence of evidence-based guidelines does not necessarily lead to adherence by healthcare professionals. Up to 30% of patients receive inappropriate care which can be damaging and should be avoided. Thus, it is crucial to implement guidelines with ‘do-not-do’ recommendations more effectively in clinical care. Although many interventions exist for implementing recommended (pharmacological) care, it is unknown which interventions are effective in reducing unnecessary prescriptions or de-implementation.

Objectives: To perform an overview of systematic reviews (SRs) to identify which interventions reduce inappropriate drug prescribing.

Methods: Embase and MEDLINE were searched in June 2015 (update ongoing). We included SRs focused on interventions to reduce inappropriate prescribing. Risk of bias of SRs was assessed using AMSTAR. SRs were mapped into domains of the Cochrane Effective Practice and Organization of Care (EPOC) taxonomy.

Results: Our search yielded 3106 SRs – 25 SRs were included. SR risk of bias was moderate in 21 SRs and low in 4 SRs. SRs included 5-89 original studies each. Patients of all ages and in all settings were included. Intervention strategies covered all domains of the EPOC taxonomy: professional interventions were evaluated in 16/25 SRs, financial interventions in 5/25 SRs, organisational interventions in 19/25 SRs and regulatory interventions in 1 SR. For 1 professional (prescription advice based on diagnostic testing) and 2 organisational interventions (multifaceted intervention, quality monitoring) (5%) we found sufficient evidence to conclude these were effective in reducing inappropriate prescribing. For 8 interventions (14%) some evidence for their effectiveness was found, for 12 (21%) insufficient evidence and for 33 (58%) no SRs were found.

Conclusions: Our results suggest that only a small number of interventions may be effective in reducing inappropriate prescribing. For the majority of interventions mentioned in the EPOC taxonomy no aggregated evidence for their effectiveness was found. More evidence focused on interventions to reduce inappropriate prescribing is needed.