The impact of physician use of electronic medical records on colorectal cancer screening.

Abstract

288 Background: The use of electronic medical records (EMR) by primary care physicians (PCP) is promoted as a method to enhance screening rates, however the impact of these systems on screening participation is uncertain. The purpose of this study was to determine if the use of EMR among a population-based sample of Ontario PCPs affects the colorectal cancer (CRC) screening participation of their patients. Methods: A survey was sent to 1,463 randomly selected PCPs in Ontario regarding adoption of EMR and cancer screening management enabled by EMR. Patients who were treated by PCP survey respondents were identified from administrative health care databases and were included if they were eligible for CRC screening (age 50-74) during the study window (April 1, 2012 – March 31, 2013). We identified patients who were up to date with CRC screening (FOBT within 2 years, flexible sigmoidoscopy within 5 years or colonoscopy within 10 years) as of March 31, 2013. Logistic regression models were created to examine the association between EMR usage and screening participation, adjusted for physician practice model, physician sex, years in practice, patient age patient sex, patient SES, and patient comorbidities using propensity scores. Results: A total of 855 PCPs responded to the survey (58.4% response rate) and 845 could be linked to administrative databases. The patient cohort included 330,441 eligible patients linked to a PCP included in the study. The proportion of patients up to date with CRC screening as of March 31, 2013 was 61.6% (203,720/330,441). After propensity score adjustment, patients were more likely to be up to date with CRC screening if their physician used an EMR (OR 1.16, 95% CI 1.04 – 1.29). Conclusions: PCP use of EMR was associated with a small, but measurable increase the likelihood that their patients were up to date with CRC screening. Enhanced use of the EMR might improve the uptake of CRC screening on a population level.