Effects of de-implementation strategies aimed at reducing low-value nursing procedures: a systematic review and meta-analysis

BACKGROUND: In the last decade, there is an increasing focus on detecting and compiling lists of low-value nursing procedures. However, less is known about effective de-implementation strategies for these procedures. Therefore, the aim of this systematic review was to summarize the evidence of effective strategies to de-implement low-value nursing procedures. METHODS: PubMed, Embase, Emcare, CINAHL, PsycINFO, Cochrane Central Register of Controlled Trials, Web of Science, and Google Scholar were searched till January 2020. Additionally, reference lists and citations of the included studies were searched. Studies were included that described de-implementation of low-value nursing procedures, i.e., procedures, test, or drug orders by nurses or nurse practitioners. PRISMA guideline was followed, and the ‘Cochrane Effective Practice and Organisation of Care’ (EPOC) taxonomy was used to categorize de-implementation strategies. A meta-analysis was performed for the volume of low-value nursing procedures in controlled studies, and Mantel–Haenszel risk ratios (95% CI) were calculated using a random effects model. RESULTS: Twenty-seven studies were included in this review. Studies used a (cluster) randomized design (n = 10), controlled before-after design (n = 5), and an uncontrolled before-after design (n = 12). Low-value nursing procedures performed by nurses and/or nurse specialists that were found in this study were restraint use (n = 20), inappropriate antibiotic prescribing (n = 3), indwelling or unnecessary urinary catheters use (n = 2), ordering unnecessary liver function tests (n = 1), and unnecessary antipsychotic prescribing (n = 1). Fourteen studies showed a significant reduction in low-value nursing procedures. Thirteen of these 14 studies included an educational component within their de-implementation strategy. Twelve controlled studies were included in the meta-analysis. Subgroup analyses for study design showed no statistically significant subgroup effect for the volume of low-value nursing procedures (p = 0.20). CONCLUSIONS: The majority of the studies with a positive significant effect used a de-implementation strategy with an educational component. Unfortunately, no conclusions can be drawn about which strategy is most effective for reducing low-value nursing care due to a high level of heterogeneity and a lack of studies. We recommend that future studies better report the effects of de-implementation strategies and perform a process evaluation to determine to which extent the strategy has been used. TRIAL REGISTRATION: The review is registered in Prospero (CRD42018105100).