Effectiveness of de-implementation of low-value healthcare practices: an overview of systematic reviews

Abstract

Background: Reducing low-value care (LVC) is crucial to improve the quality of patient care while increasing the efficient use of scarce healthcare resources. Recently, strategies to de-implement LVC have been mapped against the Expert Recommendation for Implementing Change (ERIC) compilation of strategies. However, such strategies' effectiveness across different healthcare practices has not been addressed. This overview of systematic reviews aimed to investigate the effectiveness of de-implementation initiatives and specific ERIC strategy clusters.

Methods: We searched MEDLINE (Ovid), Epistemonikos.org and Scopus (Elsevier) from 1 January 2010 to 17 April 2023 and used additional search strategies to identify relevant systematic reviews (SRs). Two reviewers independently screened abstracts and full texts against a priori-defined criteria, assessed the SR quality and extracted pre-specified data. We created harvest plots to display the results.

Results: Of 46 included SRs, 27 focused on drug treatments, such as antibiotics or opioids, twelve on laboratory tests or diagnostic imaging and seven on other healthcare practices. In categorising de-implementation strategies, SR authors applied different techniques: creating self-developed strategies (n = 12), focussing on specific de-implementation strategies (n = 14) and using published taxonomies (n = 12). Overall, 15 SRs provided evidence for the effectiveness of de-implementation interventions to reduce antibiotic and opioid utilisation. Reduced utilisation, albeit inconsistently significant, was documented in the use of antipsychotics and benzodiazepines, as well as in laboratory tests and diagnostic imaging. Strategies within the adapt and tailor to context, develop stakeholder interrelationships, and change infrastructure and workflow ERIC clusters led to a consistent reduction in LVC practices.

Conclusion: De-implementation initiatives were effective in reducing medication usage, and inconsistent significant reductions were observed for LVC laboratory tests and imaging. Notably, de-implementation clusters such as change infrastructure and workflow and develop stakeholder interrelationships emerged as the most encouraging avenues. Additionally, we provided suggestions to enhance SR quality, emphasising adherence to guidelines for synthesising complex interventions, prioritising appropriateness of care outcomes, documenting the development process of de-implementation initiatives and ensuring consistent reporting of applied de-implementation strategies.

Registration: OSF Open Science Framework 5ruzw.

Keywords: de-implementation; Antibiotic; Effectiveness; Laboratory tests; Low-value care; Overview of reviews.

Fig. 1

PRISMA Flow diagram

Fig. 2

Confidence assessments of the included SRs (see also reference [73]  to explore the online figure)

Fig. 3

ERIC strategy clusters addressed in SRs (n = 46)

Fig. 4

Harvest plot for LVC utilisation and appropriateness (n_SR=46) Explanation figure: This harvest plot represents the ‘low-value care use (utilisation)’ and ‘appropriateness of care use’ outcomes. Each included SR is represented as a bar in the matrix at least once and twice if the SR reported on both outcomes Bar position: (see Table 2): Row ‘positive’ (< 75% or MA showing an effect), ‘inconsistent positive’ (< 50–75%), or ‘no change’ (≥ 50% or MA showing no effect) Bar colour: Healthcare practice (see legend) Bar height: Confidence in the results according to the AMSTAR 2 assessment Number above the bar: Number of included primary studies in the SR Lowercase “u” under a bar: Relevant information for assessing the primary studies’ statistical significance was not reported in the SRs

Fig. 5

Harvest plots for ERIC strategy clusters (n_SR=28) Explanation figure: This harvest plots represents the ‘low-value care use (utilisation)’ outcome Bar position: (see Table 2): Row ‘positive’ (< 75% or MA showing an effect), ‘inconsistent positive’ (< 50–75%) or ‘no change’ (≥ 50% or MA showing no effect) Bar colour: Healthcare practice (see legend) Bar height: Confidence in the results according to the AMSTAR 2 assessment Number above the bar: Number of included primary studies in the SR Lowercase ‘u’ under a bar: Relevant information for assessing the primary studies’ statistical significance was not reported in the SRs

Fig. 6

Harvest plot for comparison of single versus multifaceted strategies (n_SR=8) Explanation figure:  This harvest plot represents the ‘low-value care use (utilisation)’ outcome Bar position:  (see Table 2): Row ‘positive’ (< 75% or MA showing an effect), ‘inconsistent positive’ (< 50–75%) or ‘no change’ (≥ 50% or MA showing no effect) Bar colour: Healthcare practice (see legend) Bar height: Confidence in the results according to the AMSTAR 2 assessment Number above the bar: Number of included primary studies in the SR Lowercase ‘u’ under a bar: Relevant information for assessing the primary studies’ statistical significance was not reported in the SRs

Fig. 7

Effectiveness of discrete versus multifaceted strategies (n_SR=8)