Abstract
BACKGROUND: More than 1 in 7 patients with human immunodeficiency virus (HIV) infection in the United States are unaware of their serostatus despite recommendations of US agencies that all adults through age 65 be screened for HIV at least once. To facilitate universal screening, an electronic medical record (EMR) reminder was created for our primary care practice. Screening rates before and after implementation were assessed to determine the impact of the reminder on screening rates. METHODS: A retrospective cohort analysis was performed for patients age 18-65 with visits between January 1, 2012-October 30, 2014. EMR databases were examined for HIV testing and selected patient characteristics. We evaluated the probability of HIV screening in unscreened patients before and after the reminder and used a multivariable generalized linear model to test the association between likelihood of HIV testing and specific patient characteristics. RESULTS: Prior to the reminder, the probability of receiving an HIV test for previously unscreened patients was 15.3%. This increased to 30.7% after the reminder (RR 2.02, CI 1.95-2.09, p < 0.0001). The impact was most significant in patients age 45-65. White race, English as primary language, and higher median household income were associated with lower likelihoods of screening both before and after implementation (RR 0.68, CI 0.65-0.72; RR 0.74, CI 0.67-0.82; RR 0.84, CI 0.80-0.88, respectively). CONCLUSIONS: The EMR reminder increased rates of HIV screening twofold in our practice. It was most effective in increasing screening rates in older patients. Patients who were white, English-speaking, and had higher incomes were less likely to be screened for HIV both before and after the reminder.