How AI/ML Helped Hartford HealthCare Optimize Infusion Scheduling

By Helen Gao, Senior Product Manager, LeanTaaS and Obehi Ukpebor, Head of Customer Success, Infusion, LeanTaaS

Hartford Health Care (HHC) is Connecticut’s only truly integrated healthcare system, with more than 33,000 employees, $4.3 billion in operating revenue, and a medical staff of 4,000 providers. The health system is also home to the HHC Cancer Institute, which plays a major role in system-wide innovation and research efforts. The Institute leadership constantly evaluates what can be done to improve patients and family caregiver’s quality of life as well as creating healthier communities.

As one of the leading comprehensive cancer care providers in the state, the Cancer Institute treats nearly 6,000 new patients each year, or one-third of all cancer patients in Connecticut. This volume of patients has also created one of the biggest challenges facing the Institutes’ infusion center of adequately scheduling appointments for patients receiving chemotherapy. The historical model was a manual process, largely overseen by nurses attempting to work around the schedules of providers. This approach would often cause many logistical problems that could resemble a challenging game of Tetris.

Hartford HealthCare needed a solution that would eliminate guesswork and anecdotal information from the decision-making process. Specifically, it needed a data-driven approach involving predictive and prescriptive analytics to help both plan for tomorrow and manage immediate needs for today.

Identification of this issue led Institute leadership to LeanTaaS for the company’s iQueue for Infusion Centers. iQueue leverages artificial intelligence and machine learning to optimize scheduling templates, level-load the daily schedule across nursing staff and flag future problem days for preventive action. It also identifies which appointments should be rescheduled to improve the experience for patients and staff alike.

A Straightforward and Seamless Implementation

While implementing new technology and training staff in its use can often take health systems a significant amount of time and legwork, Hartford HealthCare was able to fully launch iQueue in just one month with minimal IT involvement—including the ability to review templates ahead of the go-live.

The entire process included several overarching phases. One was piloting the iQueue platform in just one of the Cancer Institute’s locations, then slowly expanding to additional facilities, until the technology was available system-wide. The first step began with a request for data extraction from the health system’s database to establish a real-time feed with the Cancer Institute. Despite iQueue only being tested among one center to start, LeanTaaS requested data across the entire health system to create a scalable process to ensure downstream success early.

Following the data extraction, LeanTaaS also provided both onsite and virtual training to the frontline staff who would be using iQueue most frequently, including schedulers, charge nurses, nurse managers, and center executives. This became a crucial piece of the implementation given iQueue’s ability to be used within the EHR and the platform’s automatic generation of an optimized template for various use cases.

These optimized templates can help guide decision making around a higher volume of patients during holiday weeks, staffing shortages, or peaks of same-day cancellations. While centers are able to request that LeanTaaS update templates to reflect changes in patterns or same-day add-ons, iQueue was able to be customized to the Cancer Institute to specifically predict and account for these patterns.

Since iQueue was first implemented across the system, the platform has required minimal IT involvement for ongoing maintenance, updates, or support following its deployment.

The Result: Fully Utilizing Assets to Drive Revenue Growth

Since iQueue’s implementation, the technology has dramatically changed the way the Cancer Institute schedules its patients. It has allowed the centers to improve capacity so that providers can see patients throughout the entire day in a more seamless, patient-centered way.

Across the board, the Cancer Institute has experienced reduced patient wait times, improved asset utilization without having to add nurses or chairs. There has been improved patient access to treatment throughout the day, and a positive impact on nursing staff’s workday. Specific results have included:

1) Significantly reduced patient wait times

• Infusion wait time has decreased by 27.3% (35% for unlinked).
• Drug wait time has decreased by 5.1% (13.5% for unlinked).

2) An increase in optimized asset utilization

• The average daily scheduled volumes have increased by 9.1%.
• The average daily completed volumes have increased by 9.6%.
• The effective turns per chair has increased by 22.1%, from 1.73 to 2.11.
• Maximum chair occupancy has decreased by 10.0%.

3) Improved patient access to care

• The average daily scheduled patient hours have increased by 9.0%.
• The average number of scheduling lead days has decreased by 14.3% from 14 days to 12 days.

As health systems like Hartford HealthCare continue to face pressure to accommodate growing patient volume and demand, the manual processes that have historically been used are quickly becoming unsustainable. By harnessing the power of predictive and prescriptive analytics, infusion centers can more efficiently use existing resources to handle more appointments, and provide better access to new and existing patients.

To learn more about how hospitals can maximize an infusion center’s effective capacity, we invite you to view similar case studies from other leading health systems across the country.