Why cancer outcomes could depend on AI care coordination

The healthcare industry is just beginning to uncover the impact of delayed cancer screenings and care on oncology outcomes. It’s a challenge that accelerated the use of artificial intelligence (AI) worldwide in monitoring cancer progression or recurrence—and it demonstrates AI’s strong potential for ensuring patients receive the right treatment at the right time.

Two out of three patients aren’t getting recommended cancer screenings during the pandemic, according to a Prevent Cancer Foundation analysis. Further, 32% aren’t sure which screenings they should be getting. Among the top missed appointments, according to the foundation: mammograms, cervical cancer screenings and skin checks.

As a result, some researchers are already seeing patients diagnosed with later-stage cancers, in part due to delayed screenings. These gaps in care puts intense pressure on physicians to devise creative ways to encourage individuals to get caught up on recommended screenings. They also heighten the need for physicians to lean on data and tech-enabled support to identify individuals at risk and coordinate the right care at the right time to save lives.

One emerging AI solution prompts medical assistants to ask questions that provide the information physicians need to make better treatment decisions—faster. For example, in California, researchers developed an AI algorithm to identify the types of treatments best suited for treating specific subtypes of cancers. Meanwhile, in Texas, one medical center uses AI to predict which cancer patients are likely to visit the emergency department (ED) in the next month and recommend ways that physicians and clinicians can help patients avoid a costly ED visit.

How can physician practices leverage AI in cost-effective ways to improve care coordination and outcomes for cancer patients or those at risk for cancer while enhancing efficiency and revenue? Here are three approaches to consider.

No. 1: Use AI to identify critical gaps in patient data—and prompt staff for follow-up. AI-fueled software can proactively identify the types of data that are needed—and when—based on each patient's unique profile and conditions. This includes not just delayed cancer screenings, but also missing lab tests or imaging scans, unfilled prescriptions, lack of information from specialty referrals or unscheduled follow-up appointments for care. From there, medical assistants automatically receive prompts to reach out to patients to determine why care has been postponed, engage them in the appropriate next steps in care, track down missing test results and connect with other specialty providers to assess whether patients were able to make an appointment and, if so, obtain feedback.

During office visits, AI also can direct medical assistants to ask specific questions that provide the information physicians need to make better treatment decisions faster. Getting to the why is critical. For example, for patients who are undergoing hormone therapy following a diagnosis of prostate cancer, questions might include, “I noticed you didn’t receive your last hormone injection. Tell me more. Did you have any concerns about it?” This type of questioning enables medical assistants to uncover barriers to care—such as apprehension regarding the side effects of treatment—and take action (in this instance, by exploring options for mitigating side effects or considering other treatment options). At one California urology practice, the use of AI-powered care coordination to uncover barriers to recommended care or screenings helps clinicians detect prostate cancer 20% sooner. It’s an approach that improves outcomes while boosting efficiency and revenue.

No. 2: Proactively eliminate barriers to cancer care. Clinical automation, powered by AI, can assist patient navigators—health professionals who help facilitate access to care, including for individuals with complex care needs—in exploring creative ways to support earlier detection of disease or better outcomes.

Consider that social determinants of health, such as economic insecurity or lack of childcare or transportation, often disrupt cancer care in disadvantaged populations. With the help of AI-guided conversations and clinical workflows, patient navigators become empowered to dig deeper. They identify the factors that complicate access to care and leverage practice resources, community connections and more to ensure patients get the services they need.

This tactic is well-suited for urology, gastroenterology, obstetrics, dermatology and more. For example, one urology practice in the Southwest identified 223 new treatments for prostate cancer patients in nine months based on responses to condition-specific, AI-guided conversations. By meeting patients where they are, physician practices can more effectively reduce risk and improve health.

No. 3:Use AI to identify new and better ways to treat cancer. By applying AI analyses to patient care, physicians gain greater insight around the types of medications that are best suited for specific populations. They also are alerted to alternative therapies for patients who have been newly diagnosed with other types of chronic illness, such as heart disease. Such modifications could improve cancer symptoms and decrease the side effects of cancer treatment. That’s a finding that proved true for the Southwest urology practice, where AI analyses helped physicians remove or reduce side effects and improve symptoms in 196 prostate cancer patients within a nine-month period.

Today, 62% of physicians say patients’ lack of compliance with care recommendations affects quality performance. Relying on AI to automatically monitor adherence and stay alert to changes in medical history could uncover the need for early intervention, preventing problems before they start.

Eliminating Cracks in Cancer Care

Advanced tools that enhance patient data collection and engagement can reduce gaps in care for patients with cancer or those at risk while minimizing the associated administrative burden on providers. It’s a more modern technique for care management and collaboration that leads to faster identification of disease progression and treatment options. As physician practices seek to optimize care resources to create a better experience for physicians and patients, incorporating AI-based software into specialty care coordination is a tactic that closes essential gaps in data, increases efficiency and saves lives.

Shirley H. Lee, CRNP-FNP, MPH, is vice president of clinical strategy for Preveta, a care coordination platform for specialty care across medical practices and hospital systems. She is a graduate of the Johns Hopkins School of Nursing and Johns Hopkins Bloomberg School of Public Health.

This article was initially published by our sister publication Medical Economics.