Rowe AK, et al. – Researchers conducted this systematic review to investigate effective strategies for improving health-care provider (HCP) performance in low- and middle-income countries (LMICs). They found that strategy effectiveness differed substantially, although some approaches were more consistently effective than others. Findings suggested that the breadth of the Health-Care Provider Performance Review (HCPPR) makes its outcomes valuable to decision-makers for informing the selection of strategies to improve HCP practices in LMICs. These outcomes also highlighted the need for analysts to use better methods to study the effectiveness of interventions.

Methods

• From the 1960s to 2016, the investigators searched 52 electronic databases for published studies and 58 document inventories for unpublished studies.
• Controlled trials and interrupted time series were considered eligible study designs.
• Only strategy vs control group comparisons were included in this analysis.
• Results of improving HCP practice outcomes expressed as percentages (eg, percentage of patients treated correctly) or as continuous measures (eg, number of medicines prescribed per patient) were presented.
• They calculated effect sizes as absolute percentage-point changes.
• For all primary outcomes, the summary measure for each comparison was the median effect size (MES).
• Strategy effectiveness was characterized with weighted medians of MES.

Results

• Among 216,477 citations screened, the researchers selected 670 reports from 337 studies of 118 strategies.
• They noted that most strategies had multiple intervention components.
• The effects were near zero for only implementing a technology-based strategy (median MES 1.0 percentage points, IQR -2.8 to 9.9) or only providing printed information for HCPs (1.4 percentage points, -4.8 to 6.2) for professional HCPs (generally, facility-based health workers).
• For percentage outcomes, training or supervision alone typically had moderate effects (10.3–15.9 percentage points).
• On the other hand, combining training and supervision had somewhat larger effects than use of either strategy alone (18.0 to 18.8 percentage points).
• Group problem-solving alone exhibited large improvements in percentage outcomes (28.0 to 37.5 percentage points), but when the strategy definition was broadened to include group problem-solving alone or other strategy components, moderate effects were more typical (12.1 percentage points).
• Findings revealed that several multifaceted strategies had large effects, but multifaceted strategies were not always more effective than simpler ones.
• The effect of training alone was small (2.4 percentage points) for lay HCPs (generally, community health workers).
• Community support plus HCP training (8.2 to 125.0 percentage points) were strategies with larger effect sizes included.
• The outcomes suggested that contextual and methodological heterogeneity made comparisons difficult, and most strategies had low quality evidence.