UHCW NHS Trust, IBM Consulting^® and Celonis SE formed a close-knit, blended team to analyze the trust’s outpatient services through the lens of patient experience and health outcomes. The approach combined the IBM Garage model, Celonis’ AI-powered process mining, UHCW NHS Trust’s leading data analytics practices and its ongoing operation improvement activities.

The blended team analyzed over half a million pseudonymized patient journeys through the trust’s operational data, as well as in-person research and interviews with staff at the center of the process. Pseudonymized demographic data was layered over this analysis, ensuring that findings and interventions did not exacerbate health inequalities. This unique approach led to improvement opportunities and interventions in a period of weeks as opposed to months.

One such improvement focused on UHCW NHS Trust’s approach to missed patient appointments—known as did not attends (DNAs)—which are more common among those with high deprivation scores. Having identified a spike in last-minute cancellations after two SMS reminders had been sent, an IBM Garage^™ team was used to explore how to improve the likelihood of re-booking those appointment slots. They subsequently found that by adapting the timing of these text reminders—sending an earlier notice to patients 14 days in advance of their appointments and a follow-up four days before—the trust could increase timely cancellations in the cohort eligible for two text messages, allowing the potential for those appointment slots to be re-used.

As part of the project, IBM and UHCW NHS Trust also piloted IBM^® watsonx.ai^™ technology to train, tune and deploy machine learning models to support hospital staff in scheduling and validating patients on the elective backlog. And the solution uses generative AI (gen AI) capabilities to read clinical letters and aid in verifying patients’ waiting list statuses.

As process improvements were deployed, UHCW NHS Trust began to realize near immediate results. For example, the adjustment to its SMS reminders reduced DNA rates from 10% to 4% among patients in the subset of outpatients who received two SMS reminders.

With patients now cancelling or rescheduling their appointments earlier, the trust now has the option to re-purpose these slots to boost clinic activity by up to 6%. This efficiency, in turn, helped support UHCW NHS Trust to see around 700 extra patients each week as part of its overall elective recovery interventions. Current projections conducted by the trust estimate that this extra activity could contribute to a 10%–15% reduction in the overall backlog.

Further, the IBM watsonx.ai pilot has demonstrated greater accuracy and speed than human reviewers. The solution was able to review outpatient clinic letters in just 18 hours, compared to the four years required for a human equivalent. And this pilot offers scope for UHCW NHS Trust to further improve patient classification, shortening its waiting lists as those who no longer required treatment can be removed.

Part of the National Health Service (NHS), UHCW NHS Trust serves as one of the largest acute teaching hospitals in the country. The organization provides a broad range of specialized and emergency medical services to the local community and spans two campuses—the University Hospital in Coventry, UK and the Hospital of St. Cross in Rugby, UK.

An IBM strategic partner founded a little over a decade ago, Celonis offers AI-driven software-as-a-service (SaaS) services, specializing in process mining solutions. The company is simultaneously headquartered in Munich, Germany and New York City with additional offices spread throughout the United States, Europe, India and Japan.