according to authors of a new study published in the Journal of the American Medical Association.
Results from the clustered, randomized clinical trial indicate that systems-based interventions, such as automating reminders in electronic health records (EHRs), outreach in the form of phone calls or letters, and assistance with barriers to health care, such as housing insecurity, can increase the number of patients who complete appropriate diagnostic follow-up after an abnormal result.
Patients who received an EHR reminder, outreach call or letter, and additional calls to screen for and assist with nine barriers to health care – housing insecurity, food insecurity, paying for basic utilities, family caregiving, legal issues, transportation, financial compensation for treatment, education, and employment – completed follow-up within 120 days of study enrollment at a rate of 31.4%. The follow-up rate was 31% for those who received only an EHR reminder and outreach, 22.7% for those who received only an EHR reminder, and 22.9% for those who received usual care.
“The benefits of cancer screening won’t be fully realized without systems to ensure timely follow-up of abnormal results,” said Anna Tosteson, ScD, director of the Dartmouth Institute for Health Policy and Clinical Practice in Lebanon, N.H., a coauthor of the study.
Current payment incentives and quality-of-care indicators focus on getting people in for screening but should also address completion of screening – meaning timely and appropriate follow-up of results that could be indicative of cancer, Dr. Tosteson said.
“There’s a disconnect if you have screening rates that are high but once people have an abnormal result, which is potentially one step closer to a cancer diagnosis, there are no systems in place to help clinicians track them,” said study coauthor Jennifer Haas, MD, director of the Center for Primary Care Research at Massachusetts General Hospital in Boston.
In a 2016 study, researchers found that follow-up rates after abnormal cancer screenings varied widely. While 95.6% of patients with abnormal breast cancer screenings underwent timely follow-up testing, only 68.1% of patients with colorectal abnormalities and 44.8% of patients with cervical abnormalities did so.
Researchers for the new study used guideline recommendations and specialist input to create automated EHR algorithms that determined a follow-up period and diagnostic test.
They put the algorithm into practice with 11,980 patients who were part of 44 primary care practices within three health networks between August 2020 and December 2021. All patients had received abnormal test results for colorectal, breast, cervical, or lung cancer in varying risk categories.
All patients received usual care from their providers, which consisted of a “hodgepodge of whatever their clinic usually does,” Dr. Haas said. Without standards and systems in place for follow-up, the burden of testing and tracking patients with abnormal results typically falls on the primary care provider.
The researchers intervened only when patients were overdue for completion of follow-up. They then staggered the interventions sequentially.
All study participants received an automated, algorithm-triggered EHR reminder for follow-up in their patient portal along with routine health maintenance reminders. To view the reminder, patients had to log into their portal. Participants in the outreach and outreach and navigation groups also received a phone call, an EHR message, or a physical letter 2 weeks after receiving an EHR notification if they hadn’t completed follow-up. Research assistants performed the outreach after having been prompted by the algorithm.
After another 4 weeks, those in the EHR, outreach, and navigation group received a call from a patient navigator who helped them address nine barriers to health care, chiefly by providing them with referrals to free resources.
Among patients who received navigation, outcomes were not significantly better than among those who received EHR and outreach, indicating social determinants of health did not significantly affect the population studied or that the modest approach to navigation and the resources provided were insufficient, Dr. Haas said.
The complexity of an automated platform that encompasses many types of cancers, test results, and other data elements could prove difficult to apply in settings with less infrastructure, said Steven Atlas, MD, MPH, director of the Practice-Based Research and Quality Improvement Network in the division of general internal medicine at Mass General.
“I think there’s a role for the federal government to take on these initiatives,” Dr. Atlas said. Government intervention could help create “national IT systems to create standards for creating code for what an abnormal result is and how it should be followed,” he said.
While interventions improved patient follow-up, the overall rates were still low.
“What concerns me is that despite the various interventions implemented to encourage and support patients to return for follow-up testing, over 60% of patients still did not return for the recommended testing,” said Joann G. Elmore, MD, MPH, professor of medicine at the University of California, Los Angeles. Dr. Elmore was not involved with the study.
The research took place during the COVID-19 pandemic, which may have reduced follow-up, the study authors wrote. Still, given that previous research has shown that follow-up tends to be low, the rates highlight “the need to understand factors associated with not completing follow-up that go beyond reminder effort,” they wrote. These include a need for patient education about the meaning of test results and what follow-up procedures involve.
The study was supported by the National Cancer Institute and the American Cancer Society. The authors have disclosed no relevant financial relationships.
A version of this article appeared on Medscape.com.