Written and maintained by CASRAI Editorial Board
Last updated
Short answer
A stockout is a point at which an item that a facility is expected to have on hand is not available when it is needed. It is a local event, counted at a clinic, pharmacy, store room or lab, and it is different from a shortage, which the FDA assesses nationally. Health programs make stockouts measurable by fixing four things: which item, which facilities, which period and which moment of measurement.
Definition and scope
Health-supply programs turn a stockout into an indicator so the same event is counted the same way at every site. The USAID Data4Impact project measures the number of service delivery points that experienced a stockout of a specific tracer product at any point in a defined period (it gives three, six or twelve months as examples), calculated separately for each product. Its indicator is written for family planning products. The UNAIDS Indicator Registry defines a different measure with the same name: the percentage of treatment sites that had a stock-out of one or more required antiretroviral medicines during a defined period.
Both definitions share one idea that is easy to miss: a stockout is only meaningful against what a site is expected to stock. An item the site never carries is not a stockout, and an item that is physically present but not issued may not count either (see the misunderstandings below; UNAIDS says facilities holding reserve stock they do not issue are not counted). The wider purchasing context sits in CASRAI’s hospital supply chain guide.
How a stockout is measured: two indicators side by side
The two published indicators below use the same word for different calculations. Sources and the date checked are listed at the end.
| Feature | Data4Impact: stockout rate at service delivery points | UNAIDS: antiretroviral medicine stock-outs |
|---|---|---|
| Unit counted | Service delivery points (SDPs) that offer the tracer product | Treatment sites (facilities dispensing) |
| What counts as a stockout | Stocked out of one specific tracer product | Stocked out of one or more required medicines |
| Numerator | Number of SDPs stocked out of a specific tracer product | Facilities dispensing that experienced a stock-out of one or more required medicines |
| Denominator | Total SDPs that reported or were visited and that offer the product | All facilities dispensing in the same period |
| Calculation | (numerator / denominator) x 100, per product | Percentage of sites, reported annually |
| Where the data come from | Site visits, physical inventories, logistics reports, logistics management information system records | Facility stock inventory control reports and requisition forms, plus a list of what each facility is expected to dispense; surveys or site visits if the information system is not operational |
When the stock is checked
Data4Impact bases the numerator on the ending balance of the most recent logistics report or on the day of a site visit. That is a snapshot, not a continuous record. It notes that a 12-month period lets evaluators consider seasonal variation, that six months is less cumbersome because fewer reports need reviewing, and that anything shorter than 12 months calls for investigating seasonality.
Per-product versus any-product
The Data4Impact rate answers “how many sites were out of this product?” The UNAIDS indicator answers “how many sites were out of at least one required medicine?” A site missing one of several required items counts under the second definition, while under the first it is counted only in the figure for that product, so the two figures cannot be compared directly.
Stockout versus shortage versus backorder
These terms are often used interchangeably, but they describe different levels. The FDA, in its drug shortages FAQ (content current as of 12/18/2024), cites the Federal Food, Drug, and Cosmetic Act definition of a shortage as a period of time when the demand or projected demand for the drug within the United States exceeds the supply of the drug. FDA says it evaluates supply and demand on a nationwide level. It describes reports of local supply problems as usually temporary and often involving distribution issues, and says it checks with manufacturers whether national supply meets demand.
So the same product can be stocked out at one clinic while no national shortage exists. Where a supplier tells you an item is delayed rather than unavailable, the status wording matters; CASRAI’s backorder versus discontinued vendor status codes entry covers how those labels differ. The national list itself is described in the FDA drug shortage list guide.
Where it matters for labs, clinics and supply decisions
The two indicators come from public-health commodity programs, so they should not be read as laboratory standards. What they do show is what any stockout count has to specify before it can be trusted:
- The expected item list. UNAIDS requires a list of what each facility is expected to dispense, and Data4Impact counts only sites that offer the product.
- The period. Counts over different windows are not comparable, and a short window can mislead if demand is seasonal.
- The measurement point. An ending balance, a day-of-visit check and a continuous record can give different answers for the same site.
- Record quality. The Data4Impact page says the indicator is only as accurate as the stock records behind it, so it advises checking records against what was planned to be stocked.
Stocking policies address the other side of the problem: how much cover to hold. CASRAI’s safety stock versus buffer stock comparison, the par level entry and the guide to a backup supplier strategy cover those policies. A stockout rate is the outcome measure those policies are meant to move.
Common misunderstandings
“If the shelf has some stock, there was no stockout”
Not necessarily. The UNAIDS registry notes that facilities holding reserve stock that they do not issue are not counted as stocked out, and suggests measuring a “functional stock-out”, meaning the inability to access or use a required medicine. Data4Impact similarly advises checking stock against what was planned to be stocked, to catch rationing that hides a stockout.
“A low stockout rate means supply is healthy”
The UNAIDS registry states that its indicator does not explain why stock-outs occur, which medicines were affected, how long they lasted, or the quality of storage and distribution. A rate is a prompt to investigate, not a diagnosis.
“Brands and substitutes do not matter”
They change the count. Data4Impact says that if brands cannot be interchanged, each brand should be monitored separately, and that where substitution is practiced, brands can be combined. The decision has to be stated with the number.
“A stockout means a national shortage”
No. As above, the FDA evaluates shortages nationally, and local supply problems are often distribution issues.
FAQ
What is the difference between a stockout and a shortage?
A stockout is a facility-level event: an expected item is not available at a site. A shortage, as the FDA uses the term for drugs, is national: demand or projected demand in the United States exceeds supply.
How is stockout rate calculated?
Under the Data4Impact indicator it is the number of service delivery points stocked out of a specific tracer product divided by the total number that offer it, multiplied by 100, over a defined period such as 3, 6 or 12 months. Other programs define the numerator differently, as the UNAIDS indicator does with “one or more required medicines”.
Why does the measurement point matter?
Because stock moves. Data4Impact uses the ending balance of the most recent logistics report or the day of a site visit, so a site can be stocked out between two checks and not be counted.
Sources
- USAID Data4Impact, Stockout rate at service delivery points, checked 2026-10-29.
- UNAIDS Indicator Registry, Antiretroviral medicine stock-outs, checked 2026-10-29.
- U.S. Food and Drug Administration, Frequently Asked Questions about Drug Shortages, content current as of 12/18/2024, checked 2026-10-29.
