Now selecting the first cohort of facilities in Sierra Leone. Request a briefing
Health infrastructure · Starting in Sierra Leone
HealthOS is the shared record beneath hospitals, clinics, insurers and ministries. One patient identity, one clinical history, one claim, one national view — engineered for the connectivity, regulation and scale African health systems actually operate in.
Built for hospitals, clinics, insurers, and public health systems
Engineered to the standards national systems require
HL7 FHIR R4
Clinical data exchange
ICD-11 · SNOMED CT
Coded diagnoses and terminology
OpenHIE architecture
National exchange patterns
ISO 27001 controls
Information security baseline
In-country residency
Data stays under national law
Standards HealthOS is built to. Formal certification follows first production deployment.
The problem
Facilities have systems. What they do not have is a layer beneath those systems that agrees on who the patient is, what happened, and who pays. Every initiative built without that layer starts by rebuilding it.
A patient is a new patient at every facility. History, allergies, results and prescriptions stay in whichever building produced them, so clinicians decide with a fraction of the record.
Care restarts at every door
Claims are assembled by hand after the fact, from notes that were never structured for adjudication. Rejections are discovered weeks later, and the facility carries the float.
Revenue detached from the encounter
Lab, pharmacy, billing and insurance run separate software with no shared contract. Every connection becomes a bespoke project instead of a property of the platform.
Integration as a project, not a property
Capacity, utilisation, disease signal and spend are reassembled by hand each reporting cycle, from sources that disagree. By the time the picture is complete, the decision has passed.
Reporting cycles, not real time
The platform
HealthOS is not a suite of products that integrate. It is one data layer with five surfaces on top — which is why identity, care, money and intelligence stay consistent without an integration programme.
One patient. One record. Everywhere.
A unified patient identifier with deterministic and probabilistic matching across facilities, duplicate detection with human merge review, and a permanent, queryable trail of who accessed what and on whose consent.
The whole clinical day on one record.
Registration through discharge — consultations, admissions, maternity, laboratory and pharmacy — writing to the same record, so nothing is re-entered between departments and nothing is lost between them.
Claims that are complete before they are submitted.
Billing generated from the clinical encounter itself, coded as care is delivered. Invoices reconcile to the encounter that produced them, and a claim is validated against payer rules before it leaves the building.
The operational picture, while it still matters.
Occupancy, throughput, stock, revenue, clinical quality and disease signal computed on live data — the same numbers for the ward manager, the finance director and the ministry, because they come from one source.
One platform, an entire network.
Hospitals, clinics and networks under one hierarchy: shared configuration with local override, consolidated reporting across sites, and role-based access that follows a clinician wherever they work.
That is the whole platform. The usual next question is what it would take to run it across the facilities you already have.
Request a briefingArchitecture
HealthOS sits beneath the systems a country already runs. Facilities keep what works; the record, the identity and the exchange move to a layer everyone can address.
Hospitals, clinics, laboratories and pharmacies — using HealthOS directly, or connected through the API while they keep existing systems.
The master patient index, terminology services, consent and the audit trail. This is the layer that makes a record portable between institutions.
One clinical and financial history per patient, versioned and auditable, with residency and retention governed by national policy.
Payers adjudicating claims, ministries measuring capacity and disease signal, and facilities running their own operations — all reading from the same source.
HL7 FHIR R4 resources and a documented REST API. Connecting a system that already exists must never require a bespoke project.
Facility operations continue through an outage and reconcile on reconnection. Intermittent connectivity is a design constraint, not an edge case.
A network can start with identity alone, or one facility, or one department, and extend. Nothing requires a single-step national cutover.
Security and governance
Patient data is held under national law, not a vendor's terms. The governance model is designed to be inspected — by a hospital board, a data protection authority, or a ministry procurement review.
Deployed in-country where regulation or policy requires it. The sovereign deployment model exists so patient data never has to leave the jurisdiction that governs it.
TLS on every connection, encryption at rest for the record store and backups, with key management separated from application access.
Role-based access scoped to facility and function, with break-glass access recorded and reviewable rather than silently permitted.
Every read and write against a patient record is logged with actor, purpose and consent basis. The trail is queryable and cannot be edited after the fact.
Cross-facility access is governed by recorded patient consent, with scope and expiry, not by whoever holds a login.
A customer can export their full record set in open formats at any time. Lock-in is not a retention strategy for national infrastructure.
Deployment
How HealthOS is hosted is a procurement and sovereignty decision, not a product decision. The platform is the same in all three.
Clinics and hospital groups
HealthOS runs it. Fastest path to production, no infrastructure team required, regional hosting.
Insurers and large networks
Deployed to a national or regional cloud region so data residency requirements are satisfied by architecture, not by policy promise.
Ministries and public health systems
Runs inside government infrastructure, operated by your team with our support. Full control of data, keys and upgrade cadence.
If it runs where you are, the next step is a conversation. Not a procurement process. We are selecting the first cohort of facilities now.
Request a briefingFirst deployment
A platform that claims to serve a continent and starts everywhere serves nobody. HealthOS is being built for one national health system first — its facilities, its payers, its regulator, its connectivity — because a shared record has to work completely somewhere before it can work anywhere.
Expansion follows proof, not funding rounds.
A national health system of this size can reach a genuinely shared record, rather than a pilot that stalls at three facilities.
Intermittent connectivity, mixed-capacity facilities and constrained budgets are the norm here. A platform that works in these conditions travels; one built for ideal conditions does not.
HealthOS is founded and built by Sierra Leoneans. The first deployment is not a market entry — it is home.
Who it serves
Care is delivered by providers, financed by payers and stewarded by government. HealthOS is designed so all three read and write the same record instead of reconciling three versions of it.
Admissions, theatre, wards, laboratory, pharmacy and billing on one record.
Departments stop re-entering what another department already knows.
Fast registration, structured notes and same-day billing without an IT department.
Runs on the staff and hardware the clinic already has.
Structured, coded claims arriving from the encounter that produced them.
Fewer rejections, shorter cycles, a smaller fraud surface.
Capacity, utilisation and spend assembled from live facility data.
A national picture that is current, not a quarterly reconstruction.
Surveillance and programme monitoring on the data clinicians already record.
Disease signal in its first week, not its first month.
Shared standards across sites with local workflow autonomy.
One operating picture across every facility in the group.
The vision
Africa will bring more people into formal healthcare systems over the coming decades than any region in history. Those systems will not be built on paper, and they will not be built on software designed for one hospital in one country.
The countries that get this right will not be the ones that buy the most software. They will be the ones that establish a shared record first, and let everything else — insurance, surveillance, quality measurement, private innovation — be built on top of it.
HealthOS exists to be that layer: the platform connecting providers, insurers, patients and public health systems to a single, secure, sovereign record of care. It starts in Sierra Leone, because infrastructure is proven in one place before it is offered to fifty.
Open standards and a documented data layer, so HealthOS connects to what a country already runs instead of demanding it be replaced.
Residency, encryption and auditable access designed for national systems and the regulators who govern them.
Intermittent connectivity, shared devices and mixed-capacity facilities are the design constraints, not the exceptions.
Hamidu Samuel Mansaray
Founder & CEO
Founded by
Hamidu Samuel Mansaray is a machine learning engineer and startup founder. He arrived at health infrastructure the long way round: a model is only ever as good as the record behind it, and across much of Africa that record is scattered across paper files and disconnected systems, or missing altogether. You cannot put intelligence into a health system that cannot yet reliably remember its own patients. HealthOS is the layer he decided had to be built first.
Questions we are asked
No. A hospital management system automates one building. HealthOS is the layer beneath many buildings: a shared patient identity, a portable record and an exchange that facilities, insurers and ministries can all address. A facility can run its whole operation on HealthOS, but that is a surface on the platform, not the platform.
No. Facilities connect existing systems through the FHIR API and keep them. Most deployments begin with identity and exchange, and adopt clinical or revenue modules only where the current system is the constraint.
Wherever national law requires. HealthOS deploys to managed cloud, to an in-country region, or inside government infrastructure. The sovereign model exists so that data never has to leave the jurisdiction that governs it.
Facility operations continue and reconcile automatically on reconnection. Intermittent connectivity is a first-order design constraint, which is why the record is designed to converge rather than to require a permanent connection.
Consent is a recorded object with scope and expiry, not an implicit property of holding a login. Cross-facility access is granted against that record and every access is written to an immutable audit trail.
Sierra Leone. HealthOS is founded and built by Sierra Leoneans, and the first deployment is at home rather than in whichever market looked most convenient. The constraints here — intermittent connectivity, mixed-capacity facilities, constrained budgets — are the constraints that make a platform portable to the rest of the continent. Expansion follows proof.
Pre-launch. We are selecting a first cohort of hospitals, clinics, insurers and public health partners to deploy with, and we would rather tell you that plainly than imply a customer base we do not yet have.
Talk to us
Tell us which institution you represent and what you are trying to solve. Briefing requests from hospitals, insurers and ministries are answered by the founder directly.
Briefings are a 30-minute working session: your environment, the deployment model that fits, and an honest answer on what HealthOS does not yet do.
Or write directly
hello@healthos.africa