Nigerian healthcare operations leaders can use AI agents to request and track HMO authorisation codes, prepare HMO claims, find and fix duplicate patient records and reconcile cash, transfer and HMO payments. Clinical decisions, record merges, deposit waivers and data protection requests stay with people. The Nigeria Data Protection Act 2023 and HMO and health insurance scheme rules apply.
What do healthcare operations leaders in Nigeria spend their time on?
HMOs. Patients covered through employers or the national health insurance arrangements access care through health maintenance organisations. Many services need an authorisation code from the HMO before treatment, by phone, email or portal, and claims must carry the code and the right tariff. HMOs pay on different schedules, query claims and sometimes dispute tariffs, and chasing payment is a major part of the job.
Payment in every form. Many patients pay out of pocket by cash, card or bank transfer, often with deposits before admission. Reconciling what was billed, paid and owed across cash, transfers, HMOs and corporate accounts is a daily task, and transfer confirmations need checking against the bank.
Records. Many facilities run hybrid paper and electronic records. The same patient may have several folders or records under slightly different names, which splits their history and causes billing errors. Cleaning this up is slow work that rarely gets priority.
Where can AI agents help?
- HMO authorisation requests (depth 2 and 3). The agent prepares the request with the patient's enrolment details and proposed service, sends it through the HMO's channel and records the code when received. Requests for high-cost services are drafted for a person.
- HMO claim preparation (depth 2). The agent assembles claims with authorisation codes, tariffs and supporting documents, checks them against each HMO's rules and drafts the batch for the claims officer.
- Duplicate record detection (depth 2). The agent compares names, dates of birth, phone numbers and next-of-kin details across records, and drafts proposed merges with the evidence. Our playbook on fixing data quality issues in customer records with an AI agent covers the method.
- Payment reconciliation (depth 3). The agent matches transfers in the bank account to patient bills, posts clear matches and raises unmatched items. A screenshot from a patient never counts as payment.
- Four-eyes checks on billing changes (depth 2). When staff change a tariff, waive a charge or edit a bill, the agent checks the change against the approval rules and holds anything unapproved. See running four-eyes checks on operational changes with an AI agent.
Duplicate records show why merges stay human. Two records with the same name and similar dates may be one patient or two relatives. A wrong merge can put one person's allergies or results in another's file. The agent proposes, with evidence; a trained records officer merges.
What stays with people, and why?
Clinical decisions stay with clinicians, including whether treatment proceeds while authorisation is pending in an emergency.
Record merges stay with records officers, for the reason above.
Deposit waivers, discounts and decisions on patients who cannot pay stay with people with authority. Disputes with HMOs over tariffs or rejected claims stay with the claims lead. Data protection requests stay with the designated officer.
Which rules apply in Nigeria?
The Nigeria Data Protection Act 2023, enforced by the Nigeria Data Protection Commission, treats health data as sensitive personal data, with stricter conditions for processing. Providers need a lawful basis and a condition for sensitive data, must limit processing to the purpose, keep data secure and respect patients' rights. Facilities that process large volumes of personal data may need to register with the Commission and carry out compliance audits. Sharing patient data with HMOs should be limited to what the claim or authorisation requires.
HMOs and the national health insurance arrangements set their own rules on authorisation, tariffs, claims and timelines, which flow into providers' contracts. Medical confidentiality obligations on health professionals also apply.
Nigeria has no AI-specific law. Providers working with foreign partners, such as international insurers or telemedicine services, should check the rules that apply to those partners.
This is an explanation of the rules, not legal advice; confirm how they apply to you with counsel.
How do you start in the first 90 days?
Month one: start with payment reconciliation. The bank record is the source of truth, and leakage falls quickly. Agree the matching rules for transfer names and references first.
Month two: shadow HMO claim preparation for your two largest HMOs. Compare each batch with the claims officer's, and track queries returned.
Month three: certify reconciliation at depth 3 and claim preparation at depth 2. Begin duplicate detection at depth 2 on one department's records. Follow deploying AI agents: train, shadow, certify, widen. For a market with certified health data hosting and two statutory payers, see our brief on operations agents in French healthcare.
Where Pryme fits
In Pryme, an authorisation code received from the HMO's own channel is signed evidence, and a code a patient reads out over the phone is declared. The claims agent includes only signed codes in a batch. A transfer read from the hospital's bank account is sealed; a patient's screenshot is external and cannot mark a bill paid.
Questions people ask
Can the agent work when the internet is down?
It needs connectivity to reach HMOs and the bank. Plan for outages: queue requests, and let staff use the existing manual process until the connection returns.
Can the agent merge duplicate records?
No. It proposes merges with evidence, and a records officer decides.
Do we need patient consent to send data to HMOs?
You need a lawful basis. For treatment and claims under an HMO plan, that usually exists, but share only what the HMO needs.