Analysis of telehealth’s impact shows ‘modestly lower’ use of certain low-value tests

Key takeaways:

• High telehealth adoption was tied to lower use of seven low-value tests, like cervical cancer screening.
• High telehealth adoption also differentially decreased spending on visits per Medicare beneficiary.

Health care systems that adopted a higher volume of telehealth services less frequently used several low-value screening and diagnostic tests, according to results of a cohort study published in JAMA Internal Medicine.

These “modestly lower” reductions, as the researchers wrote, came despite an increase in total visits, “giving them more theoretical opportunities to get a low-value test,” the study’s lead author Ishani Ganguli, MD, MPH, an associate professor of medicine at Harvard University, noted.

The findings are consistent with those of a prior study, which showed that medium and high frequency of telehealth use reduced some low-value care like thyroid testing.

“In theory, the widespread adoption of telemedicine following the [COVID-19] pandemic could influence low-value testing — for example, telemedicine may reduce low-value testing in that when patients are not in the office, clinicians may be less likely to order these tests or patients may be less likely to complete them,” Ganguli told Healio.

“There was no national evidence on how telemedicine may influence low-value testing and spending,” Ganguli added. “We were especially interested in this question because Congress has only extended public health emergency provisions for Medicare to cover telemedicine through [March 31] and there’s active policy debate — hinging on how telemedicine impacts care quality and spending — about whether and how telemedicine coverage should continue.”

In the cohort analysis, the researchers used fee-for-service Medicare claims data from 2019 to 2022 to assess the possible links between telehealth adoption, total visits and the usage of — and spending on — 20 low-value screening, chronic preoperative, chronic condition management or acute diagnostic tests.

The study sample included 1,382,033 Medicare beneficiaries attributed to health systems with high telehealth adoption and 999,051 beneficiaries who used health systems with low telehealth adoption.

The researchers found that during the study period, those in high-telehealth systems had a small differential rise in visits [difference-in-differences (DiD) visits per beneficiary = 0.12; 95% CI, 0.03-0.21] and differential decreases in the use of seven of the 20 low-value tests:

• cervical cancer screening (DiD = 0.45 percentage points; 95% CI, 0.72 to 0.17);
• screening electrocardiograms (DiD = 1.3 percentage points; 95% CI, 1.96 to 0.65);
• screening metabolic panels (DiD = 1.84 percentage points; 95% CI, 2.87 to 0.8);
• preoperative complete blood cell counts (DiD = 0.64 percentage points; 95% CI, 1.06 to 0.22);
• preoperative metabolic panels (DiD = 1.35 percentage points; 1.91 to 0.8);
• total or free triiodothyronine level testing for hypothyroidism (DiD = 0.9 percentage points; 95% CI, 1.38 to 0.41); and
• imaging for uncomplicated low back pain (DiD = 1.66 percentage points; 95% CI, 2.35 to 0.98).

Those in high-telemedicine systems also experienced statistically significant differential decreases in spending on visits per beneficiary ($47.87; 95% CI, $86.85 to $8.88) and on two of the low-value tests: cervical cancer screening ($0.56; 95% CI, $0.89 to $0.23) and preoperative blood cell counts ($0.15; 95% CI, $0.24 to $0.06).

The researchers acknowledged that the results may not be generally applicable to patients who receive their primary care outside of health systems or those enrolled in Medicare Advantage or other plans.

When asked about the direction of future research in this area, Ganguli told Healio, “We want to understand what explains these findings — are doctors less likely to order low-value tests during virtual visits, are patients less likely to complete them or a little bit of both?

“We know that offering virtual visits along with in-person visits can be great for patient and clinician convenience,” Ganguli added. “This study shows that virtual visits may have some additional unexpected benefits — by making clinicians or patients think twice about getting tests that aren’t really going to help them.”

For more information:

Ishani Ganguli, MD, MPH, can be reached at iganguli@bwh.harvard.edu.