Effect of a population-level performance dashboard intervention on maternal-newborn outcomes: an interrupted time series study

OBJECTIVES: To assess the effect of the Maternal Newborn Dashboard on six key clinical performance indicators in the province of Ontario, Canada. DESIGN: Interrupted time series using population-based data from the provincial birth registry covering a 3-year period before implementation of the Dashboard and 2.5 years after implementation (November 2009 through March 2015). SETTING: All hospitals in the province of Ontario providing maternal-newborn care (n=94). INTERVENTION: A hospital-based online audit and feedback programme. MAIN OUTCOME MEASURES: Rates of the six performance indicators included in the Dashboard. RESULTS: 2.5 years after implementation, the audit and feedback programme was associated with statistically significant absolute decreases in the rates of episiotomy (decrease of 1.5 per 100 women, 95% CI 0.64 to 2.39), induction for postdates in women who were less than 41 weeks at delivery (decrease of 11.7 per 100 women, 95% CI 7.4 to 16.0), repeat caesarean delivery in low-risk women performed before 39 weeks (decrease of 10.4 per 100 women, 95% CI 9.3 to 11.5) and an absolute increase in the rate of appropriately timed group B streptococcus screening (increase of 2.8 per 100, 95% CI 2.2 to 3.5). The audit and feedback programme did not significantly affect the rates of unsatisfactory newborn screening blood samples or formula supplementation at discharge. No statistically significant effects were observed for the two internal control outcomes or the four external control indicators-in fact, two external control indicators (episiotomy and postdates induction) worsened relative to before implementation. CONCLUSION: An electronic audit and feedback programme implemented in maternal-newborn hospitals was associated with clinically relevant practice improvements at the provincial level in the majority of targeted indicators.