Industries / Health / Healthcare / Urgent care
Links to telehealth
This is the work of joining video and phone care to walk-in clinics: virtual urgent care services, and the routes between a video visit and an in-person one.
5
Milestones achieved, from 2020 to 2026.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
2020
Virtual urgent care multiplies
At NYU Langone Health in New York, video visits in urgent care rose from 102.4 a day to 801.6 a day between 2 March and 14 April 2020, a 683% increase, after the system added virtual urgent care staff for COVID-19. After the expansion, 56.2% of virtual urgent visits were related to COVID-19, and patients aged 20 to 44 used the service most.
Credit
NYU Langone Health; D. M. Mann, O. Nov and colleagues
2022
Amazon Care shuts down
Amazon Care began in 2019 as a pilot for Amazon's Seattle employees. It combined video visits with home visits for primary and urgent care, had no clinics of its own, and was sold to other employers from 2021; in February 2022 Amazon said six companies, Hilton among them, used it. In August 2022 Neil Lindsay of Amazon Health Services told staff it would close on 31 December 2022 because it was "not a complete enough offering for the large enterprise customers we have been targeting".
LessonBy Amazon's own account, video and home visits without clinics did not meet the needs of the large employers it was built to serve.
Credit
Amazon
2022
Sweden weighs app visits
From 2016 private companies offered video doctor visits to anyone in Sweden through contracts with health centers in Jönköping and later Sörmland; each patient's home region paid for the visit. Sweden's Agency for Health and Care Services Analysis (Vårdanalys) counted nearly 1.2 million such visits in 2019 and twice as many in 2020, for which regions paid just over SEK 500 million and SEK 900 million. In 2020 four companies, Kry, Doktor.se, Min doktor and Doktor24, had 84% of these visits. The agency found that digital visits partly replaced visits to health centers but also added new ones, were used most by children, young adults, city dwellers and people without chronic illness, and raised regions' costs.
LessonA per-visit fee open to everyone drew the healthiest users and added visits instead of only moving them.
Credit
Vårdanalys
2023
Virtual walk-ins, then the ED
Using Ontario's health administrative data, researchers compared 132,168 patients of 13 virtual walk-in clinics, direct-to-consumer online clinics, from April to December 2020 with all Ontarians who had a virtual visit with a family doctor. The virtual walk-in patients were healthier, yet 8.3% visited an emergency department within 30 days, against 4.1%, and 0.2% later saw the same doctor in person, against 11%. A second study, of 1.9 million matched patients in 2021 and 2022, found emergency visits within 7 days in 3.3% of patients after a virtual visit with an outside doctor and 2.0% after one with their own.
LessonVideo care cut off from a patient's usual doctor was followed by more emergency visits, not fewer.
Credit
L. Lapointe-Shaw, T. Kiran and colleagues
2026
A state's virtual ED, measured
The Victorian Virtual Emergency Department (VVED), run by Northern Health with Victorian Government support, offers free video calls with emergency doctors and nurses, 24 hours a day, to anyone in Victoria, Australia. A 2026 study of linked statewide hospital data covered 100,422 presentations from July 2022 to June 2023: 75.5% were managed at home and 15.3% were referred to a physical emergency department. Of those managed at home, 8.2% went to an emergency department within 48 hours and 11.4% within 7 days.
Credit
Northern Health; L. M. Sher and colleagues, in a project of La Trobe University, Northern Health and the Victorian Department of Health
Watch for
Waiting room to video room
At a metropolitan emergency department in Victoria, non-urgent patients in the waiting room were offered a VVED video consultation in a soundproof room next door. Of 339 who took part, 258 (76%) did not need the physical emergency department; 30 (9%) came back to it within 7 days, and case reviews found three adverse events that caused temporary harm.
Achieved whenother emergency departments adopt the pathway and an outside evaluation with a comparison group reports waits and return visits.
Claimed by
researchers from Northern Health, which runs VVED, and La Trobe University, July 2026
Open
01
Home exam kits on manikins
A bench test of the exam kits patients use at home in video visits, such as TytoCare's otoscope, cheap consumer digital otoscopes and a phone camera with an ear attachment, on an otoscopy training manikin whose cases are known. Volunteers who are not clinicians capture the images; clinicians rate them without knowing the device. Because the manikin's case is fixed in advance, the test scores what a remote clinician can see without involving any patient.
The briefCredit · unclaimed02
Video service limits, coded
A dated, machine-readable table of what virtual urgent care services say they treat by video, what they say needs an in-person visit, and their age limits, coded to the International Classification of Primary Care (ICPC-2), for at least 40 services in at least 5 countries and refreshed every 3 months.
The briefCredit · unclaimed
Open 01 · the brief
Home exam kits on manikins
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A bench test of the exam kits patients use at home in video visits, such as TytoCare's otoscope, cheap consumer digital otoscopes and a phone camera with an ear attachment, on an otoscopy training manikin whose cases are known. Volunteers who are not clinicians capture the images; clinicians rate them without knowing the device. Because the manikin's case is fixed in advance, the test scores what a remote clinician can see without involving any patient.
Why it is still open
The published comparisons used real patients and small numbers: 23 adults at an academic otolaryngology clinic in 2020, where 95% of self-taken digital otoscope images were acceptable to clinicians, and 50 children at the University of Virginia in a study published as a TytoCare case study. A search found no test on training manikins, where the right answer is known.
Built with what exists
Kyoto Kagaku's EAR Examination Simulator II, which holds 9 cases including a normal ear, comes in light, medium and dark skin tones and takes a real otoscope; TytoCare and consumer digital otoscopes; a web form for blinded rating.
First step
Buy one simulator and two consumer otoscopes, and run 5 volunteers through the 9 cases.
Done looks like
At least 4 devices, each used by 20 volunteers on all 9 cases; for each device, the share of captures that two blinded clinicians rate adequate, the share where they name the manikin's case correctly, and the median capture time, with all images published.
Open 02 · the brief
Video service limits, coded
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A dated, machine-readable table of what virtual urgent care services say they treat by video, what they say needs an in-person visit, and their age limits, coded to the International Classification of Primary Care (ICPC-2), for at least 40 services in at least 5 countries and refreshed every 3 months.
Why it is still open
The nearest inventory, a 2020 review, listed 21 US websites that sell prescriptions after an online visit, with their conditions in an uncoded table that has not been updated. Where services draw the line between video and in-person care matters: in Ontario, patients of virtual walk-in clinics were twice as likely as other virtual patients to visit an emergency department within 30 days.
Built with what exists
Saved copies of each service's public pages; Playwright or Browser Use 1.0 to collect them (see the Note "Deleting 80% to start over"); an AI model such as Claude to draft the codes; two human coders to check them; ICPC-2, the WONCA classification of reasons for primary care visits.
First step
Pick 10 services in two countries, save their condition pages, and code them with two coders.
Done looks like
At least 40 services in 5 countries coded, the agreement between the two coders reported as Cohen's kappa, and a published list of changes every quarter.