Characteristics of Patients Using Different Patient Portal Functions and the Impact on Primary Care Service Utilization and Appointment Adherence (Preprint)
BACKGROUND Patient portals are now widely available and increasingly adopted by patients and providers. OBJECTIVE The objective of this study was to investigate the characteristics of primary care patients using different patient portal functions and the impact of patient portal usage on patients’ primary care service utilization and appointment adherence. METHODS A retrospective observational study using a large dataset of primary care patients of University of Florida (UF) Health was conducted. A total of 4312 patient portal users and 17580 nonusers were identified during the study period July 2013 – June 2016. Patients were classified into different user subgroups based on their portal usage of messaging, laboratory, appointment, and medication functions. The association between patient demographics and portal usage were analyzed. Primary care service utilization and appointment adherence were measured by primary care office visits arrived, no-show, cancellation, and telephone encounter rates per quarter. Generalized linear models (GLM) were built to compare the time-dependent portal usage effects across different user subgroups and nonusers. RESULTS Five different user subgroups based on their dominant portal activities and usage frequency were identified. After adopting portals, patients with fewer active health problems used portal infrequently (silent users). Medicare (P = 0.04), White, and age 46-64 (P = 0.01) and 65 and older (P = 0.04) were positively associated with messaging usage. Age 19-30 (P < 0.001) was positively associated with appointment usage, and gender female (P = 0.01) was positively associated with laboratory usage. On the impact of portal usage, disease burden adjusted rate ratios (RR) between user subgroups and nonusers were obtained. The office visit and telephone encounter RRs of users to nonusers were significantly larger than 1 right after portal adoption (with P values less than 0.05), but were not significantly different 2 years after adoption, except for the silent users, whose office visit and telephone encounter rates were significantly smaller after around 1 year post portal adoption (with P values less than 0.05). Office visit cancellation rates were not changed, and no-show rates were significantly reduced or at least not changed post portal adoption. Patients who used messaging and laboratory functions more often exhibited a larger reduction in no-show among other user subgroups. CONCLUSIONS Interestingly, a high propensity to adopt patient portals does not necessarily imply more frequent usage of portals. In particular, the number of active health problems was negatively associated with portal adoption but was positively associated with portal usage. For patients with fewer active health problems, their primary care service utilization was even lower after portal adoption; with a heavy disease burden, the utilization was temporarily increased but was not significantly changed after 2 years post adoption. Overall, appointment adherence was improved after portal adoption.