Evaluation of a pharmacist-led audit and feedback intervention to reduce Gentamicin prescribing errors at admission in neonatal inpatient care in Kenya: A controlled interrupted time series study

Abstract

Abstract Background In neonatal hospital care where the population with severe illness has a high mortality rate, around 14.9% of drug prescriptions have errors in Low- and Middle- Income Countries (LMICs) settings. However, there is scant research on interventions to improve medication safety to mitigate such errors. Objective Our objective was to explore a theory-informed pharmacist-led Audit and Feedback (A&F) intervention to improve routine prescribing practices with the aim of reducing gentamicin prescribing errors in neonatal care. Methods We used interrupted time series analysis modelling changes in prescribing errors for neonates ≤28 days admitted to newborn units (NBU) in 22 hospitals in Kenya between July 2021 to June 2024 and explored intervention effects in a feedback meeting at the end of the study. The study had three phases, pre-intervention period (July 2021 to June 2022), intervention period (July 2022 to June 2023), and post-intervention period (July 2023 to June 2024). The primary study was a standard single-group interrupted time-series study (ITS) design to evaluate the comparative effectiveness of enhanced A&F in reducing prescribing error trends after its introduction in 16 hospitals. Secondary analysis included comparison to prescribing error outcomes in an additional six hospitals in a contemporaneous control group that received basic A&F reports without pharmacist involvement in the NBU prescribing practices. Results Between July 2021 and June 2024, the 16 hospitals in the primary outcome analysis and the 6 additional hospitals for the secondary outcome analysis had 36,668 and 8,943 neonates with Gentamicin prescriptions at admission retrospectively. From the incidence rate ratios (IRR) of incorrect prescribing at admission, there was no step change (IRR 1.115, 95% CI: 0.920 to 1.352, p-value=0.265) or trend change (IRR 1.014, 95% CI: 0.986 to 1.042, p-value=0.344) due to the enhanced pharmacist-led A&F intervention in the 16 hospitals in the primary study. From the secondary study, change in the trend post-intervention in the 16 primary study hospitals in the primary study relative to the 6 hospitals acting as a contemporaneous control group was positive (IRR 0.933, 95% CI: 0.878 to 0.985, p-value=0.014), despite no step change due to the enhanced A&F intervention. Conclusion We found no statistically significant effect of the team-based pharmacist-led A&F intervention on reducing gentamicin medication errors in neonatal care. Prescribing errors during intervention and post-intervention periods were increasing across all hospitals in both arms of the study during and post-intervention periods. However, relative to control hospitals sites receiving routine feedback but without pharmacist involvement or pharmacist-led CMEs, the primary study sites had a positive trend in reducing Gentamicin prescription error rates at admission during and post-introduction of the pharmacist-led A&F intervention. Trial registration PACTR , PACTR202203869312307. Registered 17 th March 2022, https://pactr.samrc.ac.za/Search.aspx?TrialID=PACTR202203869312307 Why was this study done? In newborn hospital care where the population with severe illness has a high mortality rate, around 14.9% of drug prescriptions have errors in settings such as sub-Saharan Africa (SSA). However, there is scant research in SSA on actionable audit and feedback interventions over time to reduce the rates of inappropriate and potentially harmful prescribing of antibiotics. Therefore, we evaluated whether such an intervention is associated with sustained changes when it provides continuous feedback championed by pharmacists. What did the researchers do and find? We evaluated the impact of a pharmacist-led audit-and feedback intervention for in-hospital newborn care across Kenya. We found that the intervention was not associated with sustained reduction in the level or trend in incorrect antibiotic prescribing across practices, until the study was completed (after 12 months). Despite the overall increase in prescribing errors during the study period and the 12 months after the study period, a marked difference in inaccurate prescribing trend was also seen between hospital groups where the hospital pharmacist agreed to be involved with the audit and feedback intervention. What do these findings mean? The extent to which actionable audit and feedback interventions reflect the complexity of routine hospital care in SSA determine whether long-term improvements in prescribing practices can be delivered on an ongoing basis. More research is needed to understand why and how to obtain sustained reductions in antibiotic prescribing errors during hospital stay in SSA.