Optimizing Intervention Sequences to Reduce Physician Blood Urea Nitrogen Test Overuse: A Multi-Hospital Analysis.

Background: Overuse of hospital laboratory testing has been identified as a priority for quality improvement (QI). A multifaceted initiative to reduce blood urea nitrogen (BUN) ordering was implemented in hospitals across one Canadian province, preceded by either a system-focused (SF) [electronic medical record (EMR)], person-focused (PF) [performance audit and education], or no intervention.

Objective: The purpose of this study was to demonstrate the impact of sequencing and combining interventions on Medicine physician BUN test ordering practice beyond a single hospital context.

Design: An interrupted time series with segmented regression analysis was completed. The total monthly BUN count for six hospital Medicine programs located in three different health zones in Alberta for a period of 6 and 7 years were grouped into EMR hospitals (n = 3) and non-EMR hospitals (n = 3) post-QI initiative participation.

Main Measures: Monthly BUN test order count.

Key Results: Monthly BUN test ordering for each hospital medicine program resulted in a cumulative reduction of 51 to 95%, respectively. The highest reduction (95%, slope p < 0.001) occurred with the intervention sequence of PF followed by SF, with EMR implementation. A similar reduction (93%, slope p = 0.095) was observed when PF and SF were implemented concurrently, followed by an additional PF intervention. Hospitals with EMR implementation showed less data variability month-to-month compared to non-EMR hospitals. Lower reductions occurred with PF followed by another PF intervention (57%, slope p = 0.33) and a single PF intervention without follow-up (51%, slope p = 0.62).

Conclusions: Reviewing total monthly BUN ordering over several years revealed that no intervention sequence or combination was similar; however, all (urban and rural) hospitals had continued reductions. An intervention applying the sequence of PF, SF, with EMR implementation while incorporating other influential factors is essential for sustained behavioral change. Effective implementation may require consideration of hospital workflows, practitioner norms, costs, and policy changes for broader adaptability.