Telehealth had its breakout moment out of necessity, and the obvious question afterward was whether it would fade back to a niche convenience once in-person visits were an option again. It didn’t. It settled — lower than its pandemic peak, but into a real, durable share of how people get care, not a temporary workaround.
CDC survey data puts the number at 37% of US adults who used telemedicine in the past 12 months, based on 2021 figures — and that number comes with real variation baked in. Usage was notably higher among women (42%) than men (32%), rose steadily with education — from 29% among adults without a high school diploma up to 43% among college graduates — and differed by race and ethnicity, from around a third of Hispanic, Black, and Asian adults up to roughly 39–41% among White and American Indian or Alaska Native adults. “Telehealth is available” and “telehealth is equally accessible” turn out to be different claims, and the gap between them tracks pretty closely with the same socioeconomic lines that shape access to healthcare generally.
What the evidence actually shows about quality
The more interesting question isn’t whether people use telehealth — it’s whether it works as well as showing up in person. A 2024 systematic review published in npj Digital Medicine, led by researchers affiliated with Johns Hopkins, looked at 77 studies comparing telehealth to in-person care across a range of clinical areas. The honest summary: differences in outcomes and healthcare utilization were generally small, and where they existed, they didn’t consistently favor one format over the other.
That’s a more useful finding than either “telehealth works great” or “telehealth is worse” — but it comes with a caveat worth stating plainly rather than burying: the review’s own authors noted that most of the studies they included, 61%, carried a serious or high risk of bias. The comparable-outcomes finding is real, but it’s built on evidence that’s more mixed in quality than the headline suggests, and that’s worth knowing before treating “telehealth is just as good” as settled science.
Some specifics did emerge. Patients who started with a telehealth visit were somewhat more likely to stick with prescribed medication and therapy plans, and somewhat less likely to miss a follow-up — the kind of outcome you’d expect from lower-friction access. They were also less likely to get lab work or other diagnostic testing done in a timely way, which points at exactly the kind of visit telehealth is and isn’t suited for: it can make it easier to show up, but it can’t draw blood or run a scan.
Where video and phone visits genuinely fit
The clearest, least controversial use case is anything that doesn’t require a physical examination: medication management for a stable, established condition, mental health sessions, follow-ups after an initial in-person diagnosis, discussing test results, prescription renewals. The NHS’s own guidance on video consultations puts the decision in the clinician’s hands based on the specific visit — and that’s really the honest answer to “is telehealth good for X?” It depends less on the condition in the abstract and more on whether this particular visit needs hands-on assessment.
Where it clearly doesn’t fit is anything requiring a physical exam, imaging, bloodwork, or urgent evaluation. Nobody serious is arguing telehealth should replace an ER visit or a diagnostic workup — the actual debate is narrower and more practical than the “future of medicine” framing it sometimes gets.
Talk therapy of any kind is a good example of the format working as intended: cognitive behavioral therapy for insomnia, the most evidence-backed treatment for chronic sleep problems, is increasingly delivered this way precisely because it’s a conversation-and-coaching process rather than something that requires a clinician’s hands.
The access problem underneath the convenience story
The part of the telehealth conversation that gets the least attention is that it requires two things a lot of people don’t reliably have: a broadband connection and, at the other end, a provider willing to see them remotely. A KFF Health News investigation found that roughly 3 million Americans live in mostly rural counties that lack both dependable broadband and adequate in-person healthcare access — the exact population telehealth is often pitched as helping most, sitting in the geography least able to use it.
That’s a US-specific figure, and the shape of the problem looks different elsewhere depending on how healthcare and broadband infrastructure are organized — but the underlying pattern, that remote care needs reliable infrastructure on both ends, isn’t unique to any one country.
The realistic takeaway
Telehealth isn’t a replacement for in-person medicine, and it was never really trying to be one, despite how it got talked about in 2020 and 2021. It’s a genuinely useful format for a specific, fairly large slice of routine care — the visits that are about talking, monitoring, and adjusting rather than examining — with real, unresolved gaps in who can actually reach it. That’s a less dramatic story than either “healthcare is transformed” or “it was a pandemic fad,” but it’s the one the data actually supports.
