Where private hospitals in Ghana lose 20% of revenue

Most private hospitals in Ghana lose between 15% and 25% of potential revenue to operational leakage they cannot identify. We know this not from a survey, but from sitting in their reception areas, their pharmacies, their finance offices, and watching where the cedis quietly go.

The leakage isn't dramatic. It isn't fraud, in most cases. It happens between people doing their best, with tools that can't keep up. What follows is a field guide to the four most common kinds — what they look like, what they cost, and what it actually takes to close each one.

1. Missed charge capture

The largest single category of leakage is also the most embarrassing to discuss. A clinician performs a procedure, dispenses a medication, applies a dressing, orders a test — and the action does not become a line on a bill. The work is done. The cost is incurred. The revenue evaporates.

We've measured this in pilot facilities by manually reviewing 200 randomly sampled encounters and comparing what was clinically documented to what appeared on the bill. The gap is consistently 6–12% of total billable activity, even in well-run hospitals.

The reasons are quotidian. Triage was busy. The receptionist forgot. The doctor wrote it on a paper slip that didn't make it to billing. The lab ran the test before the order was formally entered, and the bill was generated from the order rather than the result. The end-of-day reconciliation didn't reconcile.

What closes it: structurally tying every clinical action to a bill line at the moment of action — not at the moment of billing. The clinical workflow becomes the billing workflow. There is no separate "billing step." A test ordered is a test billed. A medication dispensed is a medication charged. The encounter itself cannot close until every order is either billed or explicitly waived with a reason.

2. Inventory shrinkage

Run a stock count of any pharmacy or central store in any private hospital in Ghana. Compare the count to the system's expected count. The variance — across drugs, consumables, lab reagents — is typically 3–6% of stock value over a quarter.

Some of this is unavoidable. Drugs expire. Consumables get damaged. Records have small errors. But a meaningful portion is operational: medications dispensed without a corresponding charge, supplies removed from the shelf without a requisition, stock that left the building.

The pattern is usually invisible because it never crosses anyone's desk. The pharmacy thinks the wards have it. The wards think the pharmacy still has it. The auditor sees a gap at year-end and writes "stock variance — investigate."

What closes it: a single source of truth for stock across the whole facility. Every movement — from central store to pharmacy, from pharmacy to ward, from ward to patient — recorded as a transaction. Dispensing tied to a prescription tied to a bill. Reconciliation that runs nightly and surfaces variances the next morning, not at month-end.

3. Pricing drift

A hospital sets its price book. The price book is fine. Then a payor updates their tariff and the new rates don't propagate. A specific procedure becomes an exception that the cashier handles manually. A new medication arrives without a price; "the doctor will tell you" becomes the policy. Within a year, the operative price book in practice no longer matches the price book in the system.

The cost is hard to measure but real. Each individual mis-priced item is small; in aggregate, against thousands of encounters per month, the drift compounds. Most CFOs we've spoken to underestimate this category by an order of magnitude.

What closes it: version-controlled price books with effective dates. Payor-specific pricing as first-class. Mandatory price for every billable item — no manual overrides without a documented reason. And, critically, a price-book audit cycle that someone owns.

4. Claim cycle losses

For hospitals that accept insurance — most do — a meaningful portion of revenue depends on claims being submitted accurately, on time, and resubmitted correctly when they're rejected. In practice, the claim cycle in most facilities looks like this:

The encounter ends. The bill is generated. A claim form is printed or assembled. The claim is sent to the payor — sometimes that day, sometimes that week, sometimes that month. The payor accepts it, or rejects it for one of fifteen possible reasons. If rejected, the bill goes back to a queue. The queue gets attention when finance has time, which is rarely. After 90 days, many claims simply do not recover.

The leakage here is 2–4% of billable revenue, with significant variance — facilities with strong claim discipline lose almost none; facilities without it lose substantially more.

What closes it: auto-generated claim forms at the moment of encounter close, coded correctly the first time. Submission queues with SLA tracking. Rejection-reason analysis that surfaces patterns (your top three rejection reasons account for 70% of rejections; fix them). Resubmission workflows that are owned by a specific person, not "the team."

What it adds up to

The four leaks together account for 13–25% of revenue in the typical private hospital we've measured. For a facility doing GHS 500,000 in monthly billable activity, that's GHS 65,000–125,000 a month. Annualised, GHS 780,000–1.5M.

That money already left your hospital. You just didn't see it leave.

Closing the leaks

You can attack each leak with a separate solution — a billing module, an inventory system, a price book tool, a claims tracker. We've watched hospitals do exactly that and end up with four systems that don't talk to each other. The integration cost exceeds the leakage cost.

The alternative is one operating system that holds the four loops together: clinical action tied to billing tied to stock tied to claims, with the whole thing visible on one screen for the people running the hospital. That is the bet ACOS is making.

If you want to see what it looks like when the loops are closed, we'd be happy to show you. Talk to us.