Digital Health in Nepal
What "digital health" means here
The term covers more than apps. The WHO framing that Nepali policy documents follow splits digital health technologies into four groups: those aimed at clients, those aimed at health-care providers, those aimed at health-system managers, and those aimed at data services and interoperability.1
That fourth category is where most of Nepal's remaining difficulty sits, and it is the least discussed. A country can have a good maternal health app, a working national reporting system and a hospital EMR, and still be unable to answer a simple question — has this patient been seen before, and where? — because nothing connects them.
The timeline that explains today
Nepal's current position makes more sense read as a sequence than as a snapshot. The peer-reviewed record gives the following milestones:1
| Year | Milestone |
|---|---|
| 1995 | Early ICT for health professionals |
| 1997 | Logistics Management Information System modernised |
| 2002 | Wireless Networking Project introduces connectivity to remote areas |
| 2004 | First telemedicine initiative piloted |
| 2006 | Telemedicine link established at Nangi health post |
| 2011 | HMIS data migrated to the DHIS2 platform |
| 2014 | Electronic health record system piloted |
Two things follow from this. First, Nepal is not early in digital health — it has over a decade of production experience with a national platform. Second, the 2011 DHIS2 migration set the shape of everything after it: the country got very good at aggregate reporting long before it got good at individual-level records. Most of today's interoperability difficulty is a downstream consequence of that ordering.
The policy layer
Three instruments do the governing work:1
- eHealth Strategy (2017) — the sector's first coherent statement that ICT is health-system infrastructure rather than project tooling.
- Digital Nepal Framework (2019) — places health inside a whole-of-government digitisation agenda, which matters because identity, connectivity and payments are not health-sector problems to solve alone.
- Integrated Health Information Management System (IHIMS) Roadmap 2021–2030 — the operative document for anyone building today. Its subject is integration: governing, guiding and improving health information systems as one estate rather than as a portfolio of projects.
The National Health Policy 2019 and the Nepal Health Sector Strategic Plan 2023–2030 both carry commitments to use modern technology for service delivery.1
Architecture perspective
Nepal does not have a policy gap. Read the IHIMS Roadmap and you will find the intent is already correct: integration, standards, a national architecture.
The gap is between a roadmap and an implementable specification. A roadmap says systems should interoperate. An implementable specification says which identifier is authoritative for a patient, which registry is the source of truth for a facility code, what the conformance rules are, and who rejects a non-conformant message. Until that exists and is enforced at the point of integration, every new system re-negotiates the same questions — and every integration is bespoke.
The systems layer
What is actually running matters more than what is planned. The estate an implementer meets in Nepal typically includes:
HMIS on DHIS2 — the national routine reporting backbone since 2011, operating across federal, provincial, local and facility levels. See DHIS2 in Nepal for how it is configured and where its limits are.
Logistics (eLMIS) — supply chain and commodity management, one of the older digitised functions in the sector.
Civil registration and vital statistics (CRVS) — births and deaths, and a persistent source of identity questions for health systems.
EMR/EHR deployments — piloted from 2014 onward and now present in a range of hospitals, with variation in vendor, data model and export capability.
Disease surveillance — event-based and indicator-based reporting, with requirements for timeliness that routine monthly reporting does not have.
Health insurance systems — claims and beneficiary management, which drag identity and facility-registry problems into sharp focus because money depends on the answers.
Read health information systems in Nepal for how these fit together and where the seams are.
The interoperability layer
This is the active frontier, and the most consequential development for implementers is that Nepal now has a national FHIR implementation guide in draft. The Nepal HMIS FHIR Implementation Guide is published as part of the Ministry of Health and Population's integrated HMIS work, built on FHIR R5, and at the time of writing carries a draft ballot status.2
The significance is not that FHIR is new. It is that a national guide changes integration from a bilateral negotiation into conformance against a published target. That is the difference between n² integrations and n.
What a working exchange layer still requires, beyond the guide itself:
- a client registry that can decide whether two records describe the same person
- a facility registry with stable codes that survive administrative reorganisation
- a health worker registry
- terminology services so that codes travel with meaning — ICD, SNOMED CT, LOINC
- an interoperability layer that authenticates, routes, transforms and audits
That list is the OpenHIE pattern, and it is the right one for Nepal's federal structure. Detail in healthcare interoperability in Nepal and HL7 FHIR in Nepal.
Where implementation actually breaks
The published literature is consistent about the barriers: limited access to ICT, electricity and analytically capable staff; difficult terrain; funding and oversight constraints that threaten sustainability; and privacy and data security concerns arising from the volume of personal health data being collected.1 Notably, it also finds that small-scale pilots "often face interoperability issues and sustainability concerns" — the two failure modes appear together because they share a cause.1
From an engineering standpoint, four problems recur:
1. Identity before exchange. Systems that cannot agree on a patient identifier cannot exchange patient data, whatever protocol they speak. National ID linkage helps but does not by itself solve matching for people without documents, or for records created before linkage existed.
2. Facility codes that move. Federal restructuring changed administrative boundaries. Any system whose facility identifiers encoded geography inherited a migration problem, and any indicator time series across that boundary needs documented breaks.
3. Pilot-shaped funding. A pilot is funded to demonstrate; production is funded to persist. Systems built to demonstrate rarely carry the operational scaffolding — monitoring, support rota, upgrade path — that production needs. This is why "the pilot worked" and "the system is running" are unrelated statements.
4. Data quality is produced at the point of entry. No integration layer repairs data that was never recorded properly. The strongest quality lever remains making the data useful to the person entering it — which is a workflow design problem, not a database problem.
What Nepal should do next
If I had to sequence the next three years of national work, I would order it like this — and the order matters more than the pace.
Registries first. Facility registry, then health worker registry, then client registry. They are the least contested politically and unblock everything downstream. A country that builds an exchange layer before its registries ends up rebuilding the exchange layer.
Conformance second. Publish the FHIR profiles, then stand up a public validator and a test server with synthetic data. Integrators should be able to prove conformance before they are anywhere near production data. A guide nobody can test against is documentation, not governance.
Procurement third. The cheapest interoperability intervention available to government is a contract clause: systems purchased with public money must export their data against the national profiles, and must publish a capability statement. This costs nothing and changes vendor behaviour permanently.
Then the exchange layer. Once there are registries, conformance and at least two consumers who actually need to exchange, the interoperability layer has a job to do. Built earlier, it is infrastructure in search of a use case.
AI, honestly
There is real potential in language technology for a country with Nepali-language clinical documentation, low specialist density and a large volume of unstructured text. There is also a strong temptation to skip the boring layer and announce AI.
An AI system needs the same things an exchange layer needs: identity, structured data, terminology, and governance about who may use what. Nepal's constraint on clinical AI today is not model access — it is the availability of representative, labelled, lawfully usable data, and the regulatory position of software that influences care. That argument is developed in AI in healthcare in Nepal.
Practical digital health engineering
Analysis is more credible when someone has had to make it work. Amakomaya is a Nepal-built digital health platform — maternal and child health products, a DHIS2-based reporting stack, and an HL7 FHIR interoperability layer currently being upgraded toward FHIR v6 — and it is used on this site as evidence rather than as the subject.
The engineering write-up, including what did not work, is in building an interoperable maternal health platform.
Frequently asked questions
- What is the status of digital health policy in Nepal?
- Nepal enacted an eHealth Strategy in 2017 and the Digital Nepal Framework in 2019, and works to the Integrated Health Information Management System Roadmap 2021–2030. The National Health Policy 2019 and Nepal Health Sector Strategic Plan 2023–2030 also carry technology commitments.
- Does Nepal use DHIS2?
- Yes. Health Management Information System data migrated to the DHIS2 platform in 2011, and DHIS2 remains the national platform for routine health reporting across federal, provincial, local and facility levels.
- Is HL7 FHIR used in Nepal?
- A national FHIR implementation guide for Nepal's HMIS exists in draft, published under the Ministry of Health and Population's integrated HMIS work and built on FHIR R5. Adoption by individual systems varies and should be checked per system rather than assumed.
- What is the main barrier to digital health in Nepal?
- Published research points to infrastructure, connectivity and workforce capacity, alongside funding sustainability and data privacy. From an implementation standpoint the recurring blocker is foundational: patient identity, facility registries and enforced conformance, without which systems cannot exchange data reliably.
- Who governs digital health in Nepal?
- The Ministry of Health and Population, with the Department of Health Services and the integrated HMIS programme, sets direction for national health information systems; the Digital Nepal Framework places this within wider government digitisation.
Sources
- Paudel S, Paudel D, Boucher F. Digital Health in Nepal: A Perspective on Overcoming Challenges and Leveraging Opportunities. Kathmandu University Medical Journal. 2025; 91(3): 386–91. PDF
- Nepal HMIS FHIR Implementation Guide (draft, v0.0.1-ballot), Ministry of Health and Population, Government of Nepal. fhir.hmis.gov.np
- Ministry of Health and Population, Government of Nepal — mohp.gov.np
- Department of Health Services, Government of Nepal — dohs.gov.np
- Nepal HMIS — hmis.gov.np
- WHO. Global strategy on digital health 2020–2025. Geneva: World Health Organization; 2021. who.int