A hold time study in pharmaceutical manufacturing demonstrates how long an in-process material, intermediate or bulk product can wait between one production step and the next without its quality being affected. It is one of the points inspectors check most consistently: when a blend waits days for the tablet press, or a batch of bulk tablets sits in quarantine for weeks before packaging, the company must be able to produce data — not habits — to support that waiting time. The international reference is the WHO guidance in Technical Report Series No. 992, Annex 4 (2015), which this practical guide turns into an operational protocol.
What a hold time study in pharmaceutical manufacturing is, and when it is needed
A hold time study establishes, with experimental data, the maximum period during which a material may be held under defined conditions — container, closure, environment — before being processed in the next step. The WHO guidance (TRS 992, Annex 4) is written primarily for non-sterile oral solid dosage forms, but its principles also apply to liquids, creams and ointments. Outside its scope: cleaning validation hold times (dirty and clean hold time), APIs and biologicals.
The underlying principle is clear-cut: intermediates and bulk products should normally not be stored beyond the established hold time, and hold times should be determined before the product is marketed, with the chosen maximum holding period supported by relevant data. In other words: the waiting time is a validated process parameter, not a by-product of shop-floor scheduling.
The regulatory references: WHO TRS 992, EU GMP and FDA
Three sources frame the topic, each with a different role:
- WHO TRS 992, Annex 4 (2015) — the only guideline entirely dedicated to hold time studies: it defines hold points, study design, time points and suggested tests per stage.
- 21 CFR 211.111 (FDA) — requires, where appropriate, time limits for the completion of each phase of production; deviations from established limits are acceptable only if they do not compromise quality, and must be documented and justified.
- EU GMP Annex 15 (2015) — covers hold times in a different context: §10.8 requires dirty hold time and clean hold time to be defined for the cleaning process. Do not mix the two levels: production hold time concerns the material, cleaning hold time concerns the equipment.
In an audit, the typical question is not "do you have hold times?" but "how did you justify them?". A solid answer combines the three references: defined limits (FDA), a documented study (WHO), and no improper overlap with cleaning hold times (Annex 15).
Topics like this one — turning a guideline into a protocol that stands up in inspection — are the daily bread of "The Pragmatic GMP", GuideGxP's free weekly newsletter. Subscribe here to receive a hands-on deep dive like this every week.
Designing the protocol: samples, conditions and time points
The WHO guidance does not impose a single method, but it lists the elements a defensible protocol must contain:
- A risk-based number of batches. The study may be run on one or more batches: the number must be justified through a risk-based approach (product criticality, process variability, history).
- Representative samples in the real container. The sample is held in the production pack or in a simulated container equivalent in material and closure. Where headspace matters (oxidation risk), test the worst case, including part-filled containers.
- Real storage conditions. Temperature and humidity must reflect the quarantine or production area where the material actually waits, not an ideal climatic chamber.
- A sampling plan with a baseline. Intervals and quantities are defined; results at each time point are compared with the initial data of the control sample, with statistical analysis where appropriate.
- Predefined acceptance criteria. The material specifications at the stage concerned, approved before the study starts.
Hold points, time points and tests: the stage-by-stage map
For a coated tablet, the WHO guidance identifies hold points along the whole chain: from binder solution to granulation, from dried granules to lubrication, from blend to compression, from cores to coating, down to packed bulk. The table summarises the suggested time points and typical tests:
| Stage / hold point | Suggested time points | Recommended tests (examples) |
|---|---|---|
| Binder solution | Initial, 2, 5, 8 hours (starch: initial, 2, 5 hours) | Appearance, microbial test, viscosity if applicable |
| Dispersions (granulation pastes, coating solutions) | Initial, then 12 to 72 hours in 12-hour steps | Appearance, specific gravity, viscosity, sedimentation, pH, microbial test |
| Dried granules | Initial, 15, 30, 45 days | Description, assay, related substances, LOD/water content, particle size distribution, density |
| Final blend | Initial, 15, 30, 45 days | Blend uniformity, LOD, particle size, density, microbial test |
| Cores and coated tablets in bulk | Initial, 30, 45, 60, 90 days | Description, hardness, friability, disintegration, dissolution, assay, degradation products, dose uniformity, microbial test |
An often overlooked detail: if hold time study data show adverse trends, finished product batches manufactured from intermediates held for long periods may need to be added to the long-term stability programme.
Typical mistakes and inspection expectations
- "Inherited" hold times with no data: limits copied from similar products or from shop-floor practice, with no study and no documented rationale.
- Study container different from the real one: a study in laboratory glassware does not defend a bulk waiting in polyethylene drums with an inner liner.
- Cumulative time ignored: each stage respects its own limit, but nobody assesses the cumulative effect along the chain up to packaging.
- Unmanaged deviations: exceeding a hold time is a deviation to be documented and justified with a quality impact assessment, as 21 CFR 211.111 requires — not an event to patch up afterwards in the batch record.
- Silent batch records: hold start/end dates and times not recorded, making it impossible to demonstrate compliance with the limits.
GuideGxP recommendation
Treat hold times as process specifications: map the real waiting points of your flow (including weekends and planned shutdowns), assign each a provisional risk-based limit, and plan the studies starting from the most critical stages — final blend and unpacked bulk first. Include hold time studies in the Validation Master Plan with responsibilities and deadlines, and carry the approved limits into batch records with mandatory date/time fields. At the next inspection, the difference between "we have always done it this way" and a signed protocol is the difference between a finding and a question closed in two minutes.
To structure your whole validation programme — hold time studies included — our operational guide Validation Master Plan (VMP) in GMP provides templates, a validation inventory and risk-based criteria ready for audit, aligned with PI 006-4.
