Telemedicine Cost-Effectiveness: What the Evidence Shows

Does telemedicine save money? The most honest answer the research supports is: it depends on whose money you are counting, what the program replaces, and what you count as a cost. Patients reliably save — travel, parking, child care, and lost wages are real expenses that virtual care largely eliminates. For payers and health systems, the evidence is genuinely mixed: some programs demonstrably reduce spending, others are cost-neutral, and some increase total spending by making care easier to get. That last outcome is not automatically a failure — more care for people who needed it is the point of a health system — but it means "telehealth saves money" is a claim that always needs unpacking. This article walks through where the savings are real, where the evidence is contested, and how to read a cost-effectiveness claim without being fooled.

Readers of this site's earliest incarnation may recall that its original cost-effectiveness article leaned on University of Illinois research from telemedicine's dial-up era; the questions that work raised — who pays, who saves, and compared to what — remain the right ones decades later, as the history of telemedicine makes clear.

First, define the question

"Is telemedicine cost-effective?" is really four different questions wearing one coat:

1. Cost-minimization: Does a video visit cost less to deliver than the equivalent in-person visit?
2. Cost savings: Does introducing telehealth reduce total spending, after accounting for changed utilization?
3. Cost-effectiveness: Are the health outcomes achieved worth the money spent, relative to alternatives? Health economists formalize this as cost per unit of health gained — often cost per quality-adjusted life year (QALY), a standard metric that combines length and quality of life.
4. Value: Does the program deliver something worth paying for — access, timeliness, equity, patient time — even if the ledger does not shrink?

Much of the confusion in public debate comes from answering one of these questions and claiming to have answered another. A program can win on question one (a video follow-up is cheaper to deliver than an office visit) and lose on question two (because easy visits multiply). It can lose on question two and still win on questions three and four. Peer-reviewed systematic reviews indexed in PubMed Central have repeatedly noted that many published telehealth economic evaluations are methodologically thin — small samples, short horizons, missing cost categories — which is why blanket claims in either direction outrun the evidence.

Where the savings are real

Several categories of telehealth savings are consistently supported across the literature and are largely uncontroversial.

Patient travel and time costs

The clearest, most robust finding in telehealth economics is that patients save substantial travel and time costs. This is close to arithmetic: a visit that requires no driving, no parking, no waiting room, and no half-day off work costs the patient less. Studies of rural and specialty-care programs — where patients may otherwise travel long distances to academic centers — consistently document large per-visit savings in mileage and hours. For patients managing chronic conditions with frequent follow-ups, for caregivers who must accompany them, and for hourly workers without paid leave, these savings compound. Programs serving rural populations supported by agencies like HRSA were among the first to document this, and it remains the least disputed number in the field.

The catch: patient time and travel are usually invisible to payer-perspective analyses. A study that finds "no savings" while ignoring thousands of hours of avoided patient travel has not found that telehealth lacks value; it has found that the value lands somewhere the study did not look.

Avoided emergency department visits and transfers — in some models

Certain telehealth models substitute directly for very expensive care, and when they do, savings are large per event:

• Tele-triage and virtual urgent care can resolve low-acuity problems that would otherwise land in an emergency department, though the net effect depends heavily on whether the virtual visit truly replaces an ED visit or is an added contact.
• Telestroke and emergency telemedicine networks let rural hospitals treat patients locally with remote specialist support, avoiding some helicopter and ambulance transfers that are individually enormously costly.
• Telehealth into nursing homes and correctional facilities avoids resource-intensive transport for populations for whom every off-site visit involves staff, vehicles, and security.
• Remote patient monitoring in well-run heart failure and COPD programs aims to catch decompensation early and substitute a medication adjustment for a hospitalization. Results vary by program intensity and patient selection — this is a "some programs, not all programs" finding — but the successful examples are well documented.

The pattern across these: savings are real when telehealth substitutes for something expensive. The economics of substituting a video visit for an office visit are modest; the economics of substituting a monitored home for a hospital bed are not.

No-show reduction and clinic throughput

Missed appointments waste clinical capacity, delay care, and cost practices real revenue. Telehealth consistently reduces no-show rates in many settings — transportation, child care, and work conflicts are leading causes of missed visits, and virtual visits remove them. Behavioral health, where no-show rates run high and visits are conversation-based, shows this most clearly; it is part of why telepsychiatry is arguably telehealth's strongest overall use case. Filled slots and completed care plans are an operational savings that practices feel directly.

Earlier intervention and continuity

Care delayed is often care made more expensive. When lowering the barrier to a visit means a diabetic wound is seen this week instead of next month, or a medication side effect is caught before the patient quits the drug, the downstream savings can dwarf the visit cost. This mechanism is harder to measure than travel miles — it requires long horizons and good comparison groups — but it is the economic logic behind chronic-care telehealth and much of remote monitoring. The literature broadly supports the mechanism while remaining honest that quantifying it is difficult.

Where the evidence is genuinely mixed

Does convenience add utilization?

The central contested question in telehealth economics: when visits get easier, do people have instead-of visits or in-addition-to visits? Analyses of direct-to-consumer virtual urgent care have found that a meaningful share of virtual visits represent new utilization — contacts that would not have happened otherwise — and some virtual visits are followed by an in-person visit for the same problem anyway, doubling the cost. Other analyses, particularly of telehealth integrated into an existing primary care relationship, find much higher substitution rates.

Both findings can be true, because "telehealth" is not one thing. A standalone app marketing convenience to healthy adults and a health system offering video follow-ups to its own chronic-care patients have different economics. And even where utilization rises, judging that as waste requires knowing whether the added care was low-value or was unmet need finally being met — a question that is partly empirical and partly about what we think a health system is for.

Duplication and the fragmentation problem

A virtual visit that ends with "you'll need to come in so we can examine that" has added a step, not removed one. How often this happens varies by specialty and by how well the program routes patients to the right modality in the first place. Poorly targeted telehealth — video visits for complaints that predictably require hands-on evaluation — generates duplicate costs. Well-targeted telehealth front-doors the system efficiently. Program design, not the technology, determines which one you get.

Implementation and fixed costs

Cost analyses often start the clock after the program is running. Platforms, integration with electronic records, licensing across states, staff training, and workflow redesign are real costs, and small practices feel them disproportionately — a topic covered from the operational side in how to start a telemedicine program. Mature programs amortize these costs over high visit volumes; pilots often never do, which is one reason pilot studies and steady-state programs report such different economics.

Whose perspective? The question that decides the answer

Health economists insist on stating the perspective of an analysis because it changes the conclusion. Telehealth is a case study in why.

Two implications follow. First, a payer can rationally worry about telehealth increasing claims while patients and society are clearly better off — both can be looking at accurate numbers. Second, reimbursement policy sits at the center of the fight: whether payers reimburse virtual visits at parity with in-person visits changes the economics for every practice, which is why telehealth reimbursement policy — and the state-by-state parity landscape tracked by the Center for Connected Health Policy — is inseparable from the cost-effectiveness debate. Medicare's telehealth coverage rules have shifted repeatedly in recent years and remain, as of early 2026, subject to congressional action in short increments; readers should verify current coverage status directly with CMS rather than relying on any article's snapshot.

Access versus cost: the framing that matters

Much telehealth commentary implicitly assumes the goal is spending less. But health systems also exist to produce health, and the strongest case for telehealth in much of the literature is not "cheaper care" but "care that otherwise would not happen": the rural patient hundreds of miles from a specialist, the person whose mobility, work schedule, or disability makes office visits prohibitive, the region with no local psychiatrist at all.

From a strict payer ledger, new care for previously unserved people is a cost increase. From a cost-effectiveness standpoint, it can still be an excellent buy — the relevant question becomes whether the health gained justifies the spending, the same question we ask of any drug or procedure. Framing telehealth purely as a cost-cutting tool sets it up to fail a test that most of medicine is never asked to pass. The fairer test is value: what did we get, for whom, at what cost, compared to the realistic alternative — which for many patients is not an in-person visit but no visit.

How to evaluate a cost-effectiveness claim

Whether the claim comes from a vendor deck, a press release, or a journal abstract, the same checklist applies:

• Whose perspective? Payer, patient, system, or society. A claim that does not say is a claim that has not been thought through.
• Compared to what? In-person care, no care, or nothing at all? The comparator quietly does most of the work in any economic analysis.
• What happened to total utilization? Per-visit savings mean little if visit counts changed. Look for whether the analysis measured substitution versus addition.
• Were all costs counted? Implementation, technology, staff time, patient equipment, and downstream in-person follow-ups — or just the visit fee?
• What was the time horizon? Prevention-based savings need years to appear; pilot-length studies structurally cannot find them, and short studies structurally miss late costs too.
• How strong is the design? Randomized and quasi-experimental designs with concurrent comparison groups beat before-and-after comparisons, which are confounded by everything else that changed.
• Who paid for the analysis? Industry-funded studies are not automatically wrong, but the incentive to publish favorable results is real, and independent replication is worth more.
• Is the outcome side measured at all? Costs without outcomes is half an analysis. Cheaper care that is worse care is not a bargain; costlier care that is much better care may be.

Systematic reviews in the peer-reviewed literature — searchable through PubMed Central — repeatedly reach a version of the same verdict: telehealth's economic performance is heterogeneous, evidence quality is uneven, and the most defensible conclusions are specialty-specific and model-specific rather than global. That is not a disappointing conclusion; it is a map. It tells health systems to copy the models with demonstrated substitution economics, to design against duplication, and to defend access-expanding programs on value rather than pretending they are cost-cutters.

The bottom line

Telemedicine reliably saves patients money and time; that much is settled. Whether it saves systems money depends on design: programs that substitute for expensive care — transfers, admissions, EDs, no-shows — show real savings, while programs that mostly add convenient visits tend toward cost-neutral or cost-increasing, which may still be worth it when the added visits represent met need. Anyone claiming telehealth always saves money, or never does, is selling something simpler than the evidence. The literature's actual lesson is that telehealth is not a discount — it is a redesign, and redesigns are only as economical as their designers make them. How usable that design is for the patients who need it most is its own evidence question, taken up in telehealth usability.