摘要：To measure the effect of an electronic health record (EHR) alert on chronic hepatitis B (CHB) screening among at-risk Asian and Pacific Islanders (API). API patients who had not yet completed hepatitis B surface antigen (HBsAg) testing were identified by a novel EHR-based population health tool. At-risk API patients in Cohort 1 (primarily privately insured) and Cohort 2 (includes Medicare and/or Medicaid) were randomized to alert activation in their electronic medical charts or not. In total, 8299 API were found to be deficient in HBsAg completion at baseline within our health system. In Cohort 1, 1542 patients and 1568 patients were randomized to the alert and control respectively. In Cohort 2, 2599 patients and 2590 patients were randomized to the alert and control respectively. For both cohorts combined, 389 HBsAg tests were completed in the alert group compared to 177 HBsAg tests in the control group (p < 0.0001; OR = 2.3; 95% CI 1.94–2.80), but there was no increased detection of HBsAg positivity from the alert (15 versus 13 respectively, p = 0.09; OR = 0.5; 95% CI 0.24–1.09). Our results demonstrate that personalized, automated electronic alerts increase screening for CHB, but more comprehensive measures are needed to detect HBsAg positive patients. NIH Trial Registry Number: NCT04240678.
其他摘要：Abstract To measure the effect of an electronic health record (EHR) alert on chronic hepatitis B (CHB) screening among at-risk Asian and Pacific Islanders (API). API patients who had not yet completed hepatitis B surface antigen (HBsAg) testing were identified by a novel EHR-based population health tool. At-risk API patients in Cohort 1 (primarily privately insured) and Cohort 2 (includes Medicare and/or Medicaid) were randomized to alert activation in their electronic medical charts or not. In total, 8299 API were found to be deficient in HBsAg completion at baseline within our health system. In Cohort 1, 1542 patients and 1568 patients were randomized to the alert and control respectively. In Cohort 2, 2599 patients and 2590 patients were randomized to the alert and control respectively. For both cohorts combined, 389 HBsAg tests were completed in the alert group compared to 177 HBsAg tests in the control group ( p < 0.0001; OR = 2.3; 95% CI 1.94–2.80), but there was no increased detection of HBsAg positivity from the alert (15 versus 13 respectively, p = 0.09; OR = 0.5; 95% CI 0.24–1.09). Our results demonstrate that personalized, automated electronic alerts increase screening for CHB, but more comprehensive measures are needed to detect HBsAg positive patients. NIH Trial Registry Number : NCT04240678.
