Telehealth has been a boon for modern-day patients, allowing people who might have difficulty accessing in-person appointments to continue seeing their physicians. But how many patients actually follow through on their physician’s recommendations afterward?
A new study suggests that
Investigators retrospectively examined test and referral orders for more than 4,000 patients to see how many complied with recommendations to have a colonoscopy, consult a dermatologist for a suspicious skin lesion, or undergo a cardiac stress test.
Completion of a recommended test or specialty referral was termed “diagnostic loop closure.” In particular, the researchers wanted to compare loop closure after telehealth versus in-person visits.
Rates of loop closure were low across all visit modalities but were lower for tests and referrals ordered during telehealth visits, compared with in-person visits – especially for colonoscopies.
“The take-home message for practicing clinicians is that they should be especially aware of follow-up for tests or referrals ordered during telehealth visits,” said corresponding author Maëlys Amat, MD, MBA, a primary care physician at Healthcare Associates, Beth Israel Deaconess Medical Center, Boston.
The study was published online on in JAMA Network Open.
‘Unintended side effects’
“Diagnostic errors present a huge safety concern, impacting many patient lives and costing the health care system billions of dollars, said Dr. Amat, who is also an instructor at Harvard Medical School.
“Telehealth utilization increased rapidly during the COVID pandemic, and although there are clear benefits to utilizing telehealth, our team sought to investigate unintended side effects of this technology and highlight opportunities for improvement,” she said.
To investigate the question, the researchers reviewed medical records of 4,113 patients, with a mean age of 59 years, at two Boston-based primary care sites: an urban hospital–based primary care practice and an affiliated community health center.
Orders for tests or referrals in both centers were placed electronically through the medical record. During an in-person visit, the patient was handed a form with a phone number to call to schedule the test or referral. Patients with limited English proficiency or complex needs may have received help with the scheduling the referral during check-out.
For telehealth visits, the clinician gave the patient the phone number to call to schedule the test or referral during the visit itself. In all scenarios, patients did not receive communication after the visit reminding them about the referral or test.
A loop was considered “closed” if the orders were completed within 365 days, 90 days, or 45 days for colonoscopy, dermatology visits, or cardiac stress testing, respectively.
Of the tests, 52.4% were ordered during an in-person visit, 27.8% were ordered during a telehealth visit, and 19.7% were ordered without a visit.
Tracking systems, virtual checkout
Fewer than half of the orders (42.6%) placed during a telehealth visit were completed within the designated time frame, compared with 58.4% of the orders placed during an in-person visit and 57.4% placed without a visit.
Patients who had telehealth visits were roughly half as likely as those who had in-person visits to close the loop on high-risk tests and referrals, even in an analysis that adjusted for test type, patient demographic characteristics, comorbidities, clinical site, clinician type, and patient engagement (odds ratio, 0.55; 95% confidence interval, 0.47-0.64).
Only 39.8% of colonoscopy referrals ordered during a telehealth visit were completed during the 365-day time period, compared with 56.9% ordered during an in-person visit and 56.7% ordered without a visit.
Follow-through with dermatology referrals within 90 days was roughly the same across all types of visits (63.1% for telehealth, 61.5% for in-person, and 62.9% for no visit). No significant differences were found between telehealth and in-person visits or orders placed without a visit.
Although patients seen via telehealth were less likely than those seen in person to follow through on cardiac stress tests within the 45-day window (59.1% vs. 63.2%), this difference didn’t reach statistical significance.
“Ideally, clinicians would implement automatic tracking systems to help ensure that an ordered test or referral is completed,” Dr. Amat commented. “However, if these systems aren’t yet in place, we strongly encourage clinicians to create their own work flows for tracking tests to completion.”
Additionally, “clinicians should consider implementing a virtual checkout system, similar to what is done during in-person visits, to help patients better understand recommended next steps,” she continued.
Other potentially helpful ways to improve loop closure include automatic tracking for outstanding tests, interventions such as telephone outreach to patients, automated text and email reminders, and the use of referral managers – especially in remote, rural areas or for “disadvantaged patients with limited health care access and literacy.”