Abstract
BACKGROUND: Despite the high burden of pneumococcal disease, pneumococcal vaccine coverage continues to fall short of Healthy People 2020 goals. A quasi-experimental design was used to investigate the impact of pneumococcal-specific best-practice alerts (BPAs) with and without workflow redesign compared to health maintenance notifications only, on pneumococcal vaccination rates in at-risk and high-risk adults, and on series completion in immunocompetent adults aged 65+ years. METHODS: This retrospective study used electronic health record and administrative data to identify pneumococcal vaccinations using cross sectional and historical cohorts of adults age 19+ years from 2013 to 2017 who attended clinics associated with the University of Utah Health. Difference-in-differences (DD) analyses was used to assess the impact of interventions across three observation periods (Baseline, Interim, and Follow Up). Adherence to the 2-dose vaccination schedule in older adults was measured through a longitudinal analysis. RESULTS: In DD analyses, implementing both workflow redesign and the BPA raised the vaccination rate by 8 percentage points (pp) (P < 0.001) and implementing the BPA only raised the rate by 7 pp. (P < 0.001) among at-risk adults age 19-64 years, relative to implementing health maintenance notifications (i.e., usual care) only in comparison clinics. In high-risk adults age 19-64 years, the BPA with or without workflow redesign did not significantly affect vaccination rates from baseline to follow up relative to health maintenance notifications. Per DD analyses, the effect of the BPA was mixed in immunocompetent and immunocompromised adults age 65+ years. However, immunocompetent older adults attending a clinic that implemented the BPA plus health maintenance notifications and workflow redesign (all 3 interventions) had 1.94 times higher odds (Odds ratio (OR) 1.94; P = 0.0003, 95% CI 1.24, 3.01) to receive the second pneumococcal dose than patients attending a usual practice clinic (i.e., no intervention). CONCLUSIONS: A pneumococcal BPA tool that reflects current guidelines implemented with and without workflow redesign improved vaccination rates for at-risk adults age 19-64 years and increased the likelihood of adults aged 65+ to complete the recommended 2-dose series. However, in other adult patient groups, the BPA was not consistently associated with improvements in pneumococcal vaccination rates.