Evaluation of a Multilevel Program to Improve Clinician Adherence to Management Guidelines for Acute Ischemic Stroke

IMPORTANCE: Promotion of clinician adherence to stroke guidelines can improve stroke outcomes. OBJECTIVE: To investigate the outcomes of a multilevel system program on clinician adherence to guidelines for treatment of patients with acute ischemic stroke (AIS). DESIGN, SETTING, AND PARTICIPANTS: This quality improvement study used a prospective interrupted time series (ITS) and difference-in-difference (DID) design, from August 1, 2018, to January 31, 2020, divided into preprogram term and short and long postprogram terms; each term had 6 months. Data were collected during hospitalization and at discharge with an automated medical record data capture system in 58 public hospitals in Zhejiang province, China. Data were analyzed from August 2018 to January 2020. EXPOSURES: The multilevel system program included a modularized standard template for medical records, centrally supported continuing education, continuous monitoring and feedback, and collaborative workshops. MAIN OUTCOMES AND MEASURES: The primary outcome was adherence to 12 key performance indicators (KPIs), expressed as (1) percentage of patient-applicable KPIs achieved in each participant and (2) percentage of participants among whom all applicable KPIs were achieved (dichotomous all-or-none measure). The secondary outcome was severe disability or death (modified Rankin Scale 5-6) at discharge. RESULTS: Among 45 091 patients (mean [SD] age, 69 [12] years; 18 347 female [40.7%]), 28 721 from 30 hospitals received the program and 16 370 from 28 hospitals continued routine care. In adjusted DID analysis, the program was associated with an increase in the absolute percentage of KPIs achieved per patient (6.46%; 95% CI, 5.49% to 7.43%), absolute rate of all-or-none success (8.29%; 95% CI, 6.99% to 9.60%), and decreased rate of severe disability or death at discharge (−1.68%; 95% CI, −2.99% to −0.38%). The ITS result showed the program was associated with an increase in KPIs achieved per patient per week (slope change in short-term period, 0.36%; 95% CI, 0.20% to 0.52%; level change in long-term period, (9.64%; 95% CI, 4.58% to 14.69%) and in all-or-none success (slope change in short-term period 0.34%; 95% CI, 0.23% to 0.46%; level change in long-term period 5.89%; 95% CI, 0.19% to 11.59%). CONCLUSIONS AND RELEVANCE: The centrally supported program was associated with increases in clinician adherence to guidelines and reduced the proportion of severely disabled or deceased patients with AIS at discharge, providing support for its wider implementation.