New Call Center and Machine Learning Research Tool Designed to Improve Healthcare Outreach

Personalized outreach has proven to be an important part of cancer care, and a consortium of researchers has developed a call center tool to promote the effectiveness of liver cancer outreach.

As part of October liver cancer awareness month efforts, a team from Iowa State University, Notre Dame, Texas A&M, Rice University and University of Texas Southwestern Medical Center worked on the project. The colleagues developed a tool designed to personalize outreach efforts as well as recommend the most suitable intervention for each patient.

The research is designed to help health care providers measure their return on health marketing investments while also educating high risk individuals about the importance of semiannual cancer screenings.

The team conducted a field experiment of 1,800 patients over a period of several months. The patients were grouped into three categories based on outreach efforts. The "usual care" group involved doctors offering patients preventative care recommendations during their regular visits. The "outreach alone" group included patients receiving their usual care as well as a letter regarding screening. Those who did not respond to the letter by scheduling an appointment also received a follow-up phone call. The "outreach with patent navigation" group had patients receiving usual care, a screening letter, an outreach call and an additional call reminder. If patients still declined to make an appointment, the researchers worked to provide motivational messaging and identify barriers.

According to the researchers, the additional call center-based outreach increased screening completion rates by up to 24 percent.

“We know how important it is for people at high risk for liver cancer to undergo regular screening,” said Ju-Yeon Lee, an associate professor of marketing at Iowa State University’s Ivy College of Business. “Research has shown regular screening reduces mortality rates for liver cancer by 37 percent and it can reduce the cost of treatment. That’s why it’s so important that outreach efforts are effective at getting patients to complete the screening.”

In addition to the call center outreach, researchers used machine learning algorithms known as causal forests to understand the differences among outreach efforts. The technology helped identify differences in how patients responded to each type of outreach.

The team concluded that increased screening from outreach alone, as well as outreach with patient navigation, was higher among female patients and those who were part of a racial or ethnic minority. It was also higher among those with a better health status, those covered by medical-assistance insurance and those who lived closer to a clinic.

According to Lee, the targeted outreach can enhance return on healthcare marketing investments by 74 to 96 percent. She added that healthcare providers can apply the approach to any type of outreach effort, not just liver cancer.

“Our research clearly demonstrates that a one-size-fits-all approach is not effective nor worth the investment,” said Lee. “We encourage health care providers to apply machine learning to develop data-driven, patient-centric outreach programs.”

Edited by Maurice Nagle