Missed Visits Prediction Model

Working with KPWA Patient Access, we developed model for predicting visits likely to be no-show or same-day cancellation.

Missed Visits Prediction Model

Kaiser Permanente Washington’s Patient Access team sends texts to remind patients of upcoming visits, but no-shows and same-day cancellations are still common. These events make clinic operations less efficient and reduce access to care. The Patient Access team wanted to send additional text messages to prevent these events, which meant first identifying the appointments most likely to be missed. We evaluated Epic’s off-the-shelf no-show prediction model and, while it was adequate, we were able to estimate our own home-grown model that more accurately identified those appointments in our population.

Having pinpointed these appointments, we conducted a randomized quality improvement project to estimate the effectiveness of additional text message reminders for reducing no-shows and same-day cancellations in primary care and mental health. We found that the intervention reduced no-shows and same-day cancellations, so we are currently building this model into our electronic health record to support systemwide adoption.

Because many care utilization patterns have changed since COVID-19, we are taking steps to ensure models estimated pre-COVID-19 continue to perform well. We are monitoring the performance of our no-show model and also running Epic’s model in the background, to allow ongoing comparisons and optimization.

Featured publications

Ulloa-Pérez E, Blasi PR, Westbrook EO, Lozano P, Coleman KF, Coley RY. Pragmatic Randomized Study of Targeted Text Message Reminders to Reduce Missed Clinic Visits. Perm J 2022;26:21.078. E-pub 5 April 2022. doi.org/10.7812/TPP/21.078. Full text

Advanced Analytics