A physician at a district hospital in rural Malawi is looking at an imaging study she was not fully trained to read, on a patient she cannot safely refer, because the nearest specialist who could confirm her read is a full day's travel away and may not have capacity to see the case for weeks.
She has internet access, intermittently. She has a smartphone. What she does not have is a verified channel to reach a specialist who could look at this specific study and answer her specific question in the next hour, the way a colleague down the hall would in a well-resourced hospital.
She is enrolled, as it happens, in a tele-mentorship hub through a program modeled on Project ECHO's structure. It meets weekly. The session is genuinely useful, when it runs, for the cases she can hold onto until Thursday. This case, in front of her right now, cannot hold.
So she does what clinicians in her position do every day, worldwide, in the absence of anything better: she makes the best call she can, alone, with a scan she is not confident she is reading correctly, for a patient whose outcome depends on a second opinion that simply is not reachable in time.
The tele-mentorship infrastructure that exists is real and it works, but it runs on a teaching calendar, not a consult calendar, and the patient in front of her tonight needs the second kind, not the first.
The scale of what is missing
The number that anchors this problem is large enough that it is worth sitting with before moving to mechanism.
The Lancet Commission on Global Surgery estimated that roughly 5 billion people lack access to safe, affordable surgical and anesthesia care when needed. That figure, as cited in a 2022 review in The Surgeon, traces back to the original 2015 Lancet Commission report; this article relies on that secondary citation rather than the primary document, and the chain of citation is worth naming plainly.
The same Commission's work put a companion number on the scale of unmet need: roughly 143 million additional surgical procedures are required annually to close the global surgical-access gap.
Behind both figures sits a population of local clinicians, generalists and specialists alike, who are the ones actually managing that unmet need day to day, with whatever backup they can find, which for a large share of them is effectively none.
What already exists, and what it was built to do
It would be wrong to suggest nothing has been tried here. Grant-funded tele-mentorship programs are real, they function, and they have measurably improved care in the places they operate.
Project ECHO, based at the University of New Mexico, is the best known of these. It operates a hub-and-spoke tele-mentorship model, in which specialists at an academic hub hold structured sessions with clinicians at spoke sites, working through cases and building capacity over time. This article did not locate precise, current figures for ECHO's global reach (country count, number of active hubs, clinicians trained) with sufficient confidence to report them, and does not attempt to estimate them here; ECHO's own model documentation confirms the structure without supplying an independently verified reach figure in the evidence available.
iPath and Collegium Telemedicus are smaller, more specialized teleconsultation networks (iPath focused substantially on telepathology) operating in comparable territory. Their scale relative to overall global need is, on the evidence available, modest.
What all three share, and what matters most for this article's argument, is cadence. ECHO's core mechanism is a structured, scheduled session, typically weekly, built for teaching and case-based learning over time. That is a genuinely different thing from an on-demand consult, and conflating the two is where most discussion of this space goes wrong.
The cadence mismatch is the actual gap
Project ECHO's own framing is "move knowledge, not patients," and it is the right direction. The mechanism it uses to move that knowledge, a weekly scheduled hub session, is a teaching format. It was designed to build a local clinician's durable capability over months, and it does that well.
A teaching format is not a consult format. When the physician in Malawi is looking at an ambiguous scan tonight, she does not need a curriculum. She needs an answer, from a specific qualified person, on a timeline measured in hours, not until next Thursday's session.
The missing primitive is on-demand, verified specialist reachability, layered on top of structured mentorship, not in place of it. ECHO-style programs should keep doing what they do well: building sustained local capability through a scheduled teaching relationship. What is absent is the separate, faster channel for the case that cannot wait for the next session, which is a different product with a different design requirement, even though it addresses the same underlying gap.
Why nobody has built the on-demand layer
Walk through the institutions positioned to build this, and the reason each one has not.
Grant-funded programs like ECHO are funded project by project and specialty by specialty. Each hub exists because a specific grant supports a specific disease area or region. That funding structure is well suited to sustained teaching relationships and poorly suited to a general-purpose, always-on consult infrastructure that would need to span every specialty and every region simultaneously, which no single grant cycle is designed to fund.
WHO and major global-health NGOs coordinate policy and emergency response at a systemic level. Everyday specialist curbside consults, the kind a physician in Malawi needs tonight, sit below the altitude these organizations typically operate at.
Commercial telehealth platforms are built, almost without exception, around paying, insured patients in high-income markets. Their entire product and business model, from intake to billing to liability structure, assumes a patient-facing transaction that does not exist in this context.
The result is the same pattern seen elsewhere in this series: every institution that could plausibly own this problem has a genuinely good reason, specific to its own mandate and funding model, why the on-demand layer is not the thing it builds.
An instructive parallel, inside a wealthy country
It is worth pausing on one piece of evidence that reframes how "LMIC-specific" this problem actually is.
A study of rural US primary care physicians (Bernson et al., PRiMER, 2021) found that roughly 70 percent report having no psychiatrist to readily refer to. This is not evidence about conditions in low- or middle-income countries; it is offered here only as an analogy, and a useful one. It shows that specialist-access failure is not purely a function of national income. It is fundamentally a distribution and connectivity problem, structural even inside a wealthy country with an enormous specialist workforce, when that workforce is not physically or virtually reachable from where the generalist actually stands.
That parallel matters because it suggests the fix is not simply "more specialists trained somewhere in the world." The world may already have enough specialist expertise, in aggregate, to answer a meaningfully larger share of these questions. What is missing is a verified channel connecting a specific available specialist to a specific asking generalist, at the moment the question exists, regardless of which side of which border either of them happens to be standing on.
Who could supply the missing capacity
There is a population that maps unusually well onto this specific gap: physicians in high-income countries who are retired or semi-retired, roughly 55 and older, whose clinical judgment remains sharp and whose daily schedule has more flexibility than it did during full-time practice.
This is not a new population to identify. It is the same supply-side thesis that already applies to domestic peer support and mentorship needs within wealthy health systems: a large, motivated, currently underused pool of experienced clinical judgment. Extending that same population's volunteer time to LMIC consult, under clear advisory-only guardrails, is a natural extension of an existing supply rather than a wholly new recruiting problem.
Smartphone and connectivity penetration in the regions ECHO and comparable programs already operate has also reached a point where near-real-time asynchronous consult, a photo, a case summary, a specific question, is technically feasible in most of the same places ECHO already reaches, which removes the old bandwidth argument for scheduled-only sessions.
What would actually work
On-demand, not scheduled. The core design has to depart from ECHO's weekly cadence for this specific use case: a case posted today should reach a qualified volunteer specialist within hours, not at the next calendar session.
Advisory only, with local authority explicit and unchanged. The consulting specialist offers professional judgment; the local clinician retains full clinical authority and responsibility. This has to be structural, stated plainly in how the interaction works, not buried in terms of service, because it is what makes cross-border liability exposure manageable for the volunteer specialist.
Verified on both ends. The volunteer specialist's credentials need to be verified before they can answer; the requesting clinician's affiliation and role need equivalent verification, so trust runs in both directions rather than being assumed.
De-identified case information only. No patient identifiers cross borders. The consult is about the clinical pattern and the management question, not a specific named patient's full record, consistent with how de-identified peer consultation works elsewhere in this series' proposals.
Specialty and geography matched. A general "ask any doctor" queue will not produce reliable answers for a genuinely rare or complex case. The routing has to match the specific specialty and, where relevant, the regional disease context to a volunteer who actually has that background.
Layered on top of ECHO, not competing with it. Where a structured hub relationship already exists for a given region or disease area, an on-demand layer should complement that relationship, not fragment the local clinician's attention across multiple, uncoordinated channels.
Funded through institutions that can absorb the cost, not through LMIC clinician dues. The realistic monetization path runs through foundations, global-health NGOs, and high-income-country health systems sponsoring their retiring physicians' volunteer time, not through direct fees charged to the clinicians this is meant to serve, who by definition cannot sustain high-income-country pricing.
What you can do now
If you are an LMIC clinician
Enroll in a structured program if one exists in your region or specialty, even knowing its cadence will not cover every case. Project ECHO, iPath, and Collegium Telemedicus, where reachable, build durable capability over time even though none currently solves the same-day question.
Build your own informal specialist network deliberately. Former classmates, colleagues from any training abroad, and diaspora physicians from your own country are, today, the fastest real channel available for an urgent question, precisely because no formal on-demand system yet exists to replace that informal one.
Document the cases where you had no one to ask. A specific, dated record of "I needed a specialist opinion on this and had no reachable channel" is the raw evidence that eventually justifies funding for the missing infrastructure, and right now that evidence mostly does not exist in any collected form.
If you are a retired or semi-retired specialist
Consider structured volunteer consult time seriously. The population of experienced, high-income-country specialists with schedule flexibility is exactly the supply side this gap needs, and the guardrails required (advisory only, de-identified, local authority retained) are the same ones that already govern responsible peer consultation elsewhere in medicine.
Start with an existing structured program if you want a proven on-ramp. Project ECHO and comparable hub-based programs are a lower-friction way to begin volunteering specialist time than waiting for an on-demand platform that does not yet exist at scale.
If you fund or lead global health programs
Ask explicitly whether your grantees' model is scheduled or on-demand, and fund the gap deliberately if it is only the former. Most existing tele-mentorship investment optimizes for durable capability building, which is valuable and insufficient on its own for the case that cannot wait a week.
Treat retired-physician volunteer supply as an underused asset, not a hypothetical one. The same population already engaged in domestic peer-support and mentorship efforts is a realistic, largely untapped source of consult capacity for exactly this problem.
Frequently asked questions
How can doctors in developing countries get a specialist's opinion? Currently, mainly through grant-funded, scheduled tele-mentorship programs like Project ECHO, iPath, or Collegium Telemedicus, where those programs reach their region and specialty, or through informal personal networks such as former classmates or diaspora physicians. No general-purpose, on-demand, verified specialist consult channel currently covers the full scope of need.
What is Project ECHO and how does it work? Project ECHO, based at the University of New Mexico, uses a hub-and-spoke tele-mentorship model in which specialists at an academic hub hold structured, typically weekly sessions with clinicians at spoke sites to build local capability through case-based teaching over time. It is a scheduled teaching format, not an on-demand consult service.
How many people worldwide lack access to surgical care? Roughly 5 billion people lack access to safe, affordable surgical and anesthesia care when needed, and an estimated 143 million additional surgical procedures are needed annually to close the gap, according to the Lancet Commission on Global Surgery (2015), as cited in a 2022 review in The Surgeon (PMID 34930698).
Can retired physicians volunteer remotely for global health? Yes, and it is an underused resource. Retired and semi-retired physicians in high-income countries represent a large, experienced volunteer pool with schedule flexibility; the same supply-side population already informally addresses domestic peer-support and mentorship gaps, and extending it to structured, advisory-only, verified LMIC consult work follows the same logic.
Is the specialist-access gap only a problem in low- and middle-income countries? No. A 2021 study found roughly 70 percent of rural US primary care physicians report having no psychiatrist to readily refer to, illustrating that the underlying mechanism, distance from a reachable specialist rather than global specialist scarcity alone, scales into wealthy countries as well (Bernson et al., PRiMER, 2021). This is offered as an analogy, not as evidence about LMIC conditions specifically.
Why doesn't Project ECHO already solve on-demand specialist access? Because its core mechanism is a structured, scheduled session, typically weekly, designed to build durable local clinical capability over time rather than to answer an urgent, individual case on the same day it arises. A clinician with an urgent question today cannot wait for the next scheduled hub session, and ECHO's model was not built to fill that specific gap.
The bottom line
The physician in rural Malawi is not failing her patient for lack of skill, and she is not without any support at all. If she is fortunate, she is enrolled in a real, functioning tele-mentorship hub that will genuinely improve her practice over the coming months.
None of that helps her tonight, in front of a scan she is not confident reading, with a patient she cannot safely refer and a specialist consult that will not arrive before Thursday.
Roughly 5 billion people live within the reach of exactly this gap, per the Lancet Commission's widely cited estimate. The infrastructure that exists to address it, Project ECHO foremost among the programs with real evidence behind them, works precisely as designed, and what it was designed to do is teach, on a schedule, not answer, on demand.
The fix is not obviously more ECHO hubs, though more would help. It is a second, faster layer: verified, on-demand, advisory-only specialist reachability, drawing on a large and currently underused pool of experienced, semi-retired specialists in wealthy countries, matched by specialty and geography to the clinician who has the question right now.
Until that layer exists, the physician in the district hospital keeps making the best call she can, alone, on a timeline the calendar sets for her rather than the patient in front of her.
Part of a series on the missing professional infrastructure of healthcare. Previously: No Portable Passport
Evidence note: the 5 billion figure for unmet surgical and anesthesia access and the 143 million additional annual procedures figure both trace to the Lancet Commission on Global Surgery (2015), as cited secondhand in Corbally, The Surgeon, 2022 (PMID 34930698); this article relies on that secondary citation and did not independently verify the primary Lancet report. Project ECHO's operating model (hub-and-spoke, scheduled tele-mentorship) is confirmed via ECHO's own published materials (projectecho.unm.edu); this article did not locate reliable current figures for its global reach (countries, hubs, or clinicians trained) and does not report any. iPath and Collegium Telemedicus are described structurally based on limited available information and their precise current scale is unverified here. The rural US psychiatrist-access figure (roughly 70 percent with no ready referral) is from Bernson et al., PRiMER, 2021 (PMID 35178506), and is used strictly as an analogy about specialist-access mechanics, not as evidence about conditions in any LMIC. Nothing in this article is clinical guidance for any specific patient or case.
