As we stagger through the first (and possibly second) wave of the COVID-19 pandemic in America, there have been proclamations of transforming healthcare delivery with technology alternating with arguments that healthcare will quickly return to normal with in-office encounters and its associated hassles (i.e., co-payments, waiting rooms, exposure to other patients who might be infected). I have been skeptical of telemedicine’s potential to move beyond urgent care for primarily self-limited conditions. I am optimistic that telemedicine (and/or remote patient monitoring) could augment physical offices managing patients with chronic disease, but we as an industry have not generated the evidence to support that shift for employers and payers.
Last week, I came across a study lending credence to the notion that the rapid adoption of telemedicine may be transient. Ateev Mehrotra and a group of researchers published telemedicine utilization through the pandemic utilizing a dataset from Phreesia, including over 50,000 providers across all 50 states. The group states the included providers manage over one million visits each week. Here are a few findings that I found interesting:
- Although visits to ambulatory providers dropped 58% in late March, visits rebounded to pre-pandemic levels in September. This rebound was observed among commercial, Medicaid and Medicare populations.
- Patients under the age of five are still not visiting physicians as often as older patients.
- Compared to the first week of March, visits during the first week of October were
- at least 10% higher for adult primary care and dermatology
- at least 10% lower for pulmonology, otolaryngology, cardiology, allergy/immunology and neurology
- The percentage of all visits that were conducted by telemedicine started at zero percent through mid-March, rose to 14% in early April, and then dropped to six percent in early October (the latest set of data recorded).
- Groups of six or more providers saw higher rebounds (14%) than practices with five or fewer providers (-6%). Larger organizations re more likely to use telemedicine across all time points measured starting in late March (the latest measurement in early October found larger practices using telemedicine 9.4% compared to 4.3% for smaller practices). One-third of practices, regardless of size, never used telemedicine at all over the study period.
The findings suggest that telemedicine was adopted in response to the public health crisis without regard to how the technology might have fundamentally changed how healthcare might be delivered after the pandemic. The pandemic may accelerate the demise of small practices, driving patients to larger groups of physicians. These groups (increasing owned by large health systems or, in the case for some specialties, private equity) will negotiate with a shrinking number of payers where all parties may have limited incentives to increase value for patients.
As Anderson et al. described back in 2003, we Americans spent more, as a percentage of gross domestic product (GDP), than other countries on healthcare, but our rates of health service were not as high as some other Organization of Economic Cooperation and Development (OECD) countries (all data from 2000):
| Metric | Germany | Japan | Switzerland | United States | OECD median |
|---|---|---|---|---|---|
| Total health spending, % of GDP | 10.6% | 7.8% | 10.7% | 13.0% | 8.0% |
| Nurses per 1000 population | 9.3 | 7.8 | — | 8.3 | 7.6 |
| Physicians per 1000 population | 3.6 | 1.9 | 3.5 | 2.8 | 3.1 |
| Physician visits per capita | — | — | — | 5.8 | 5.9 |
| Acute care hospital days per capita | 1.9 | — | 1.3 | 0.7 | 1.0 |
| MRI units per million | 6.2 | 23.2 | 13.0 | 8.1 | 4.7 |
| CT scanners per million | 17.1 | 84.4 | 18.5 | 13.6 | 12.2 |
| Patients undergoing dialysis per 100,000 population | 64.0 | 162.4 | — | 86.5 | 39.8 |
As a tribute to Uwe Reinhardt, the surviving authors published a follow-up article in 2019. Here are their updated results (data from 2015 or 2016):
| Metric | Germany | Japan | Switzerland | United States | OECD median |
|---|---|---|---|---|---|
| Total health spending, % of GDP (2016) | 11.3% | 10.9% | 12.4% | 17.2%* | 8.9% |
| Percent of the population over 65 (2016) | 21.1% | 27.3% | 18.0% | 17.9% | 18.2% |
| Nurses per 1000 population (2015) | 13.3 | 11.0 | 18.0 | 7.9 | 9.9 |
| Physicians per 1000 population (2015) | 4.1 | 2.4 | 4.2 | 2.6 | 3.2 |
| Physician visits per capita (2015) | 10.0 | 16.0 | — | 6.5 | 6.5 |
| Generalist physicians, % of total physicians (2015) | 41.2% | — | 27.5% | 11.9% | 27.9% |
| MRI units per million (2015) | 33.6 | 51.7 | — | 39.0 | 12.6 |
| CT scanners per million (2015) | 35.1 | 107.2 | 36.2 | 41.0 | 17.8 |
*On October 16th, Altarum published its August 2020 Health Sector Economic Indicators Spending Brief. The report estimated that America’s health-spending share of GDP was 18.2%.
Johansen compared utilization and expenditures across two different time periods (1996-1997 and 2011-2012) using the Medicare Expenditure Panel Survey (MEPS). Total expenditures increased 47.2% ($362/individual/month) over the two intervals. He found no difference in total visits, outpatient visits, physician visits, specialty physician visits, emergency department visits or inpatient hospitalizations, but expenditures increased across all categories. Primary care physician visits decreased across the two intervals.
If Americans are not using health services more than people in other countries, it seems unlikely that offering a new service (i.e., telemedicine, other virtual care services) without adjusting the overall value of the services delivered will make a difference in healthcare value. Unless telemedicine replaces office visits and other existing encounter types, payers will view telemedicine as a service like urgent care or convenience clinics and develop insurance products that cater to the needs of employers and members who want that service, further raising the price of healthcare and lowering its value.
There are new market entrants (some relatively new to healthcare [Amazon, Walmart, telehealth start-ups] and others who are repurposing their offerings to better meet market demand [brick-and-mortar pharmacies, large health systems with capital reserves and a brand name to attract interested patients]) who might inject some new dynamics into how healthcare is delivered. Historically, patients have been reluctant to transfer even a portion of their care to someone who is not connected to their primary provider (family medicine, oncology, nephrology), making any meaningful change in healthcare delivery more challenging. Since most provider groups are connected to specialists and hospitals, the providers’ employers are reluctant to allow patients to migrate outside the organization for even portions of their care.
Using Virtual Care for Chronic Disease Management
If a provider and patient are confident about a diagnosis with a clearly defined treatment course, virtual care could help manage that diagnosis at a lower price point with more frequent evaluations using a more consistent process (i.e., higher value) than intermittent face-to-face encounters. Most non-communicable chronic diseases could be managed this way. Hypertension may have the most data supporting a healthcare team virtual approach, but diabetes, heart failure, depression and substance abuse could also be managed this way.
Using Telemedicine to Influence the Decision to Have a Procedure Performed
Directing individuals to high-value specialists or high-value diagnostic facilities could help patients save on their healthcare expenses. More importantly, patients who may not be as well-versed in the medical literature as a clinician may benefit from a frank discussion about risks and benefits of a procedure along with a discussion about all available treatment options. The telemedicine-enabled consultant may have less of an incentive to push for a procedure than someone who has a stake in an ambulatory surgical center nearby. As the odds of a procedure successfully addressing the patient’s symptoms drop (e.g., surgical interventions on the lower back may only be 50% at resolving patients’ back pain) and the cost of the procedure or the procedure’s complications rises (back surgeries are expensive), the value of a telemedicine consult for all patients with those symptoms prior to making a decision to undergo surgery increases.
Telemedicine to Move Some Services to Higher-Value Locations
Another way telemedicine and virtual care visits could increase healthcare value would be to shift services that have traditionally been performed in high-cost venues to lower cost ones. CMS recently agreed to cover cardiac rehabilitation (with or without exercise) and pulmonary rehabilitation with exercise using telemedicine codes during the pandemic period. Setting the telehealth payment rate at or near the in-person rate will appease providers who have invested in telemedicine technology, but it may not address the payer’s short-term interest as well as the patient’s and employer’s long-term interest in modulating healthcare costs.
Using telemedicine in this way only makes sense if patients view an operative procedure as an opportunity to look for the highest-value provider, regardless of institutional affiliation. Once the decision to have a procedure is made, the surgeon usually determines the site of service. Without a method to push for alternatives, patients are unlikely to argue for a higher-value procedure location. One risk of promoting telemedicine as it is currently deployed is that telemedicine could further consolidate power within large medical systems who own the providers, the surgical centers and the telemedicine platforms to deliver care across even wider geographies. Medical center-owned proceduralists could then direct patients from far away to specific surgical suites, further marginalizing smaller provider groups. Employers and payers need to refine methods to intervene at the moment of referral or consideration of a procedure to help patients consider other options beyond the specialists linked to their primary care provider’s employer.
Summary
Telemedicine could be used to transform chronic disease management, increase the provision of second opinions virtually and promote testing or treatment for discrete services (e.g., MRI testing, colonoscopy, cardiac bypass surgery) to increase healthcare value. For chronic disease management, payers and employers could develop parallel care services and/or develop value-based agreements with primary care providers to facilitate the evolution of these new care processes. Promoting second opinions using telemedicine might help avoid some unnecessary procedures, but only if the virtual evaluations can provide alternative interventions that address the symptoms triggering the request for the procedure. Finally, telemedicine could help migrate some services out of hospitals and outpatient treatment locations into patients’ homes. All three of these approaches are likely to be commandeered by high-profile medical centers with the capital and reputation to provide a superior online experience with some patient reported outcomes to validate their utility compared to other telemedicine providers. Payers and employers may need to consider refining platform strategies that might connect high-value providers with telemedicine capabilities to patients who need specific services via telemedicine to extract any value from telemedicine and virtual care.