Hospital Operations & Patient Flow Platform
A private hospital or multi-doctor clinic with outpatient consultation, a pharmacy, a laboratory and inpatient beds. Patient records on cards in a file room. Billing computed at each point of service. Some patients on HMO schemes, most paying directly.
The reality
A private hospital or multi-doctor clinic runs outpatient consultation, a pharmacy, a laboratory and inpatient beds. Patient records live on cards in a file room. Billing gets worked out at each point of service. Some patients are on HMO schemes; most pay directly. The patient's identity, their clinical history and their financial position live in three separate places that only reconcile when someone walks between rooms.
Where it breaks
- A doctor sees a patient without knowing their allergies or history until a paper file physically arrives.
- A charge gets worked out from memory at the cash point instead of being raised by the clinical action that incurred it, so billing and care drift apart.
- An HMO claim gets rejected for a reason nobody wrote down, so the same avoidable rejection happens again on the next claim.
- Nobody can say who has looked at a given patient's record, or when.
- A new patient record gets created for someone who already has one, because nothing checks before it happens.
The Afivox approach
Afivox doesn't start by digitising the file room. It starts by mapping where a patient's identity, their clinical record and their bill actually diverge today, and builds the system to keep those three things as one continuously reconciled record instead of three that occasionally get compared.
The system
Users
Receptionist, nurse, doctor, pharmacist, lab scientist, cashier
Application
Registration, appointments and queue, clinical
Services
Pharmacy, laboratory, billing, HMO and claims
Data
Patient → Encounter → Note, Order → Result, Account → Charge → Payment
Integrations
HMO scheme eligibility and claims
Security
Roles, break-glass logging, access log
What it looks like
Concept mockups — illustrative interfaces, not a built system.
Operations dashboard
14
In queue
22 min
Avg wait
31/40
Beds occupied
9
Claims pending
6
Lab results due
3
Pharmacy alerts
Appointment calendar
| Time | Patient | Doctor | Status |
|---|---|---|---|
| 09:00 | T. Adeyemi | Dr. Bello | |
| 09:30 | M. Chukwu | Dr. Nnamdi | |
| 10:00 | F. Yusuf | Dr. Bello |
Reception queue
Clinician workspace
Patient journey
Billing status
| Patient | Charges | HMO portion | Patient portion | Status |
|---|---|---|---|---|
| T. Adeyemi | ₦24,500 | ₦18,000 | ₦6,500 | |
| M. Chukwu | ₦12,000 | ₦0 | ₦12,000 | |
| S. Okonkwo | ₦41,200 | ₦41,200 | ₦0 |
Audit screen
| User | Action | Record | Time |
|---|---|---|---|
| Dr. Bello | Viewed | T. Adeyemi — Note | 09:22 |
| N. Eze (Pharmacist) | Dispensed | S. Okonkwo — Prescription | 09:31 |
| Dr. Nnamdi | M. Chukwu — Note | 09:40 |
How it flows
Secure by design
- Roles for receptionist, nurse, doctor, pharmacist, laboratory scientist, cashier, records officer and administrator, each seeing only what their role needs.
- Clinical notes are visible only to clinical roles; billing is visible only to billing roles.
- A clinical note is never destructively edited — corrections are addenda with an author and a timestamp.
- Break-glass emergency access is permitted, but every use is logged loudly and surfaced for review.
- Every view of a clinical record, not only every edit, is logged and attributable.
Designed to improve
- Continuity of care — the doctor sees allergies and history without waiting for a file.
- Billing accuracy — charges appear on the account as care is given, not reconstructed at the cash point.
- Claims performance — a claim is assembled from the encounter, validated before submission, with its rejection reason recorded when one occurs.
- Accountability — every view of a clinical record is attributable.
Before
- A patient's identity, history and bill live in three places that only agree when someone physically checks each one.
- A doctor prescribes without seeing an allergy that's written down somewhere else.
- A charge is worked out from memory at the cash point, sometimes after the patient has already left.
- A rejected HMO claim gets resubmitted with the same mistake, because nobody recorded why it was rejected the first time.
After
- One record per patient, found before a second one is ever created.
- The doctor sees allergies and history without waiting for a file.
- Charges appear on the account as care is given.
- A claim's rejection reason is recorded, so the same avoidable rejection stops recurring.
Your operation probably has a workflow like this.
One session. We look at what's breaking, and what we'd build around it — whether or not you hire us afterward.
Tell us where yours breaks