Stock records and medicine expiry in private pharmacies

A dispenser in a retail pharmacy on a trading-centre road reaches to the top shelf for a box of antibiotics, because the lower shelf emptied on Saturday and nobody has been up there since the delivery before last. She turns the carton over and reads the date printed along its edge: it expired in March. There are more behind it. She sets them aside for disposal, notes the loss on the back page of the exercise book she keeps for such things, and serves the next customer, because the queue pays the rent. What she cannot establish, from anything in the shop, is when those boxes arrived, which wholesaler sent them, or whether the rest of that batch is on a shelf she has not reached.

Multiply that shelf across the country, and what comes into view has little to do with dishonesty and everything to do with what a counter can and cannot see about its own stock.

Trap and colleagues, in the Journal of Pharmaceutical Policy and Practice in 2016, record 604 licensed private-sector pharmacies and 6,140 licensed drug shops in Uganda in 2013 — and the licensed trade is not the whole trade. When Buchner and colleagues mapped drug outlets across 284 villages of Kamuli District for PLOS ONE in 2019, they located 215 shops; of the 123 open and willing to be interviewed, 12 held a licence and 93 did not.

Uganda's achievement here deserves to be said plainly before any criticism, because this is where its people actually get their medicines. Obakiro and colleagues, surveying 695 households in Kibuku, Kapchorwa and Mbale for BMC Health Services Research in 2026, found that private pharmacies and drug shops were the main source of medicine for 80.3 per cent of them, and that 72.52 per cent named frequent stock-outs as a barrier to getting medicines. The private counter is what is open when the public shelf is bare, and it is reasonably well stocked: a survey of 126 private outlets by Birabwa and colleagues in 2014 found artemisinin-based combinations in 99 per cent of them.

Eight in ten households surveyed in eastern Uganda buy their medicines at a private counter. Almost none of those counters can say which batch they are selling.

WHAT THE COUNTER CAN SEE

A busy pharmacy has excellent visibility of one thing and poor visibility of the rest. It knows to the shilling what came in today, because the money is counted every evening and a shortfall is noticed at once. Stock is looked at only when somebody looks at it, which means when a customer asks and the dispenser goes to check. Nothing in that routine records how fast a line moves, how much is left, when it arrived or when it dies — and without those facts, ordering is guesswork dressed as experience, corrected only by the two failures it produces: an empty shelf and a written-off carton.

The one careful Ugandan measurement of the practices behind that is old, and the absence of anything comparable since is part of the argument. Nakyanzi and colleagues surveyed 38 medicine outlets in Kampala and Entebbe, 32 of them private, for the Bulletin of the World Health Organization in 2010.

EXPIRY BEGINS AS A RECORDS PROBLEM

Expiry is usually discussed as waste or disposal, and both are the end of the story rather than its cause. In that same study, 18 of the 32 private pharmacies had destroyed medicines and 24 had taken in stock at a reduced price or as a donation, and the authors found the medicines most prone to expiry to be the donated ones, those supplied through vertical programmes, and those with a slow turnover — which is precisely what a counter with no consumption record cannot identify in advance. A discount on stock that is about to die is not a bargain but a loss the shop has agreed to carry without knowing its size.

THE BATCH IS THE UNIT NOBODY KEEPS

The second cost is harder to see and considerably worse. The World Health Organization estimated in 2017 that one in ten medical products circulating in low- and middle-income countries is substandard or falsified. A systematic review of African studies by Mekonnen, Yizengaw and Worku, in 2024, put the pooled prevalence higher, at 22.6 per cent of samples tested. Birabwa and colleagues reached the same territory from another direction, finding brands on Ugandan private shelves that were not on the National Drug Register, among them five unlisted brands of rifampicin.

Little of that can be handled at the counter, which does not hold the one fact that would let it act. A qualitative study in Wakiso District by Musoke and colleagues, in PLOS ONE in 2025, recorded participants describing the alteration of expiry dates on medicines that had already expired, and reported that people identified suspect products by what was missing from them: a quality mark, a seal, a batch number, a label. A batch number written nowhere cannot be traced backwards to a supplier or forwards to a patient. Neither can it be recalled — presenting the National Drug and Health Products Authority Bill, 2025 to Parliament in September 2025, the minister listed drug recalls among the things the 1993 Act does not regulate.

TANZANIA ACCREDITED ITS WAY OUT OF THIS

The instructive comparison is next door. From 2003, Tanzania's Ministry of Health began converting ordinary drug shops into Accredited Drug Dispensing Outlets, a programme documented by Rutta and colleagues in 2015. Accreditation combined dispenser training, business incentives, supervision and enforcement, and made record keeping a condition rather than a virtue: an accredited shop keeps records of the prescription medicines it sells, of customer complaints, and of expired medicines. By mid-2013 it had reached every region of mainland Tanzania, and more than 9,000 shops are now accredited or close to it; in Ruvuma, where it began, unregistered medicines on shop shelves fell from 26 per cent to 2 per cent. The same paper records the model being adapted in Uganda, with 689 shops accredited or awaiting accreditation by July 2015.

WHAT IS THE WAY FORWARD?

Several things would move this, none of them a national platform. Parliament and the regulator should carry the recall power in the 2025 Bill through into a licence condition naming what an outlet records at receiving — product, supplier, batch, expiry, quantity — and then inspect against it, because an inspection indicator is the only version of a records requirement that changes behaviour. The Ministry of Health should take the accredited-outlet approach past pilot scale, and commission a current measurement of private-sector stock practice, given that the best available one describes 38 outlets in 2010. Wholesalers should issue invoices carrying batch and expiry in a form a system can read, not a clerk copy. Providers building software for this market should design around the receiving door rather than the till, work without a connection, and price in shillings. Owners can start before any of that arrives, with one line per delivery.

The carton on the top shelf was lost long before March. It was lost on the afternoon it came through the door and nobody wrote down what it was, who brought it, or when it would die — and it will go on being lost there until receiving becomes the moment something is recorded.