Effectiveness of collaborative improvement: evidence from 27 applications in 12 less-developed and middle-income countries.

INTRODUCTION The improvement collaborative approach has been widely promoted in developed countries as an effective method to spread clinical practices, but little has been published on its effectiveness in developing country settings. Between 1998 and 2008, the United States Agency for International Development funded 54 collaboratives in 14 low- and middle-income countries, adapting the approach to resource-constrained environments. METHODS The authors analysed data on provider compliance with standards and outcomes from 27 collaboratives in 12 countries that met study inclusion criteria (at least 12 months of data available for analysis and indicators measured as percentages). The dataset, representing 1338 facility-based teams, consisted of 135 time-series charts related to maternal, newborn and child health, HIV/AIDS, family planning, malaria and tuberculosis. An average of 28 months of data was available for each chart. RESULTS Eighty-seven per cent of these charts achieved performance levels of 80% or higher, and 76% reached at least 90% performance, even though two-thirds had a baseline performance below 50%. Teams achieved average increases of 51.9 percentage points (SE = 28.0) per chart, with baseline value being the main determinant of absolute increase. Teams consistently maintained this level of performance for an average of 13 months (69% of months of observation). The average time to reach 80% performance was 9.2 months (SE 8.5), and to reach 90% performance, 14.4 months (SE = 12.0). CONCLUSION Collaborative improvement can produce significant, sustained gains in compliance with standards and outcomes in less-developed settings and merits wider application as a strategy for health systems strengthening.