Stability testing and shelf life

Stability studies for every new formula, in its final pack

Shelf life testing puts measured data behind the overage, the pack and the best before date of every custom and private label supplement we develop and make in Leicester for premium wellness brands. Every new formula goes on accelerated and real-time stability studies in its final pack from the pilot batch, with photostability for liquids in clear packs and light-sensitive actives, and in-use studies for multidose packs such as powder tubs, liquid bottles and gummy jars.

Shelf life testing proves that a supplement still matches its label on the last day of its shelf life, not only on the day it was made. Marketplaces check label claims, as our testing and batch release page explains, so we make the stability file part of the formula.

What your stability file contains

• The protocol, with batches, conditions, time points and acceptance criteria
• Results at every time point, with each method and the laboratory's accreditation status for that test
• Trend analysis, the shelf-life justification and the overage rationale for each active and market
• Label wording for storage, the best before date and the once-opened period

Five stresses behind supplement stability, and the control for each

Supplement stability testing starts with chemistry. Five stresses drive degradation: moisture, oxygen, heat, light and reactions between ingredients, such as Maillard browning of reducing sugars with amino groups. Each has a control, and the study proves it works.

Heat, and the limits of accelerated data

The Arrhenius equation links a reaction's rate constant to absolute temperature through its activation energy: rates climb steeply as temperature rises, which is why storage at 40°C exposes weak points early. The forecast holds only while the same reaction runs at both temperatures. ICH Q1A(R2) itself notes that some physical changes, such as the melting of creams, "may be expected under accelerated conditions". Supplements do too: gelatine gels melt close to body temperature, and amorphous powders turn sticky once absorbed water drops their glass transition below storage temperature. So accelerated data rank prototypes and forecast; real-time data at the label's storage condition confirm the date we print.

Water activity and moisture

Moisture content, by loss on drying or Karl Fischer titration, is how much water a product holds. Water activity is how available that water is: the ratio of the product's water vapour pressure to that of pure water at the same temperature, on a scale of 0 to 1. FDA notes that most foods sit above 0.95, enough for bacteria, yeasts and moulds to grow. Dried supplements sit far lower, so water activity there mainly governs chemistry and texture, with yeasts and moulds the remaining microbial watch-point in gummies.

Water moves from high water activity to low, not from wet to dry by weight. Gelatine capsule shells hold 13% to 16% moisture against 4% to 6% for HPMC (Al-Tabakha, 2010), so a dry, hygroscopic fill can pull water from gelatine and leave the shell brittle. For probiotics, a review of production practice advises a total water activity below 0.2, ideally below 0.15 for long shelf lives (Fenster and colleagues, 2019). Maillard browning runs fastest at intermediate water activity, close to where many gummies and soft chews sit.

Overages set from data

An overage is active added above the declared amount so the label still holds at the end of shelf life. As of October 2026, the European Commission's guidance on tolerances accepts vitamins in food supplements from 20% below to 50% above the declared value, and minerals from 20% below to 45% above, measurement uncertainty included. Legal maximum amounts take priority over the upper tolerance. With the US floor of 100% for added nutrients, a product sold in both markets under one declared value needs each added vitamin between 100% and 150% of that value, and each added mineral between 100% and 145%, from release to its last day. We size each overage from the loss rate measured in your product and pack, the variability of the test method and those windows, never from habit.

Packaging barrier

The pack is part of the formula, so studies run in the final pack.

• Moisture: HDPE is a better moisture barrier than PET, glass and aluminium foil are close to complete barriers, and desiccants hold the headspace dry.
• Oxygen: PET and glass resist oxygen far better than HDPE, and nitrogen flushing lowers headspace oxygen.
• Unit doses: PVC blisters let moisture through, PVDC coatings or PCTFE laminates slow it, and cold-form aluminium blisters stop it.

Barrier data come from water vapour transmission testing to ASTM F1249, oxygen transmission to ASTM D3985 and container moisture permeation to USP <671>. EU rules require recyclable packaging from 2030 at the earliest, and a move from foil laminate to a recyclable mono-material film changes the barrier. Every pack change goes back through stability, which is where our packaging development work starts.

Accelerated or real-time: how shelf life testing sets your date

We design each stability protocol in-house and arrange storage and testing with accredited partner laboratories working to ISO/IEC 17025. Every assay is checked for your product's matrix and is stability-indicating: it separates the active from its breakdown products, so a falling result is real.

Accelerated stability testing and real-time studies

Our benchmark is ICH Q1A(R2), the current ICH stability guideline for medicines; as of October 2026, a consolidated ICH Q1 guideline to replace it is planned but not yet adopted. For brands selling into hot, very humid climates, we add long-term storage at 30°C and 75% relative humidity.

For a medicine, ICH Q1A(R2) counts as significant change a 5% change in assay from its initial value, any degradation product exceeding its acceptance criterion, or a failure on appearance, physical attributes, functionality, pH or dissolution. We apply the same triggers as early warnings.

How a study runs

Samples from the pilot batch, in the final pack, are labelled by condition and time point and logged into climate chambers at our partner laboratories, which run under continuous monitoring with alarms. At each pull, samples are tested by the release methods and trended against the release result, and any out-of-trend result is investigated and closed with QA and Operations sign-off.

What we measure at each time point

Photostability follows ICH Q1B, whose confirmatory test exposes samples to at least 1.2 million lux hours of overall illumination and at least 200 watt hours per square metre of near ultraviolet energy. We run it for liquids in clear packs and light-sensitive actives such as riboflavin and vitamin B12.

How shelf life is assigned

Shelf life is the period over which every attribute stays within specification. ICH Q1E sets the statistics: for an attribute that falls over time, such as assay, shelf life ends at the earliest time the lower one-sided 95% confidence limit for the mean crosses the acceptance criterion. Where long-term and accelerated data show little change and little variability, Q1E lets a room-temperature medicine's shelf life run up to twice as long as, but not more than 12 months beyond, the period covered by long-term data. We hold supplements to the same discipline: an initial shelf life from accelerated and early real-time data, confirmed by real-time data that run to the end of the printed date. In the UK and EU, that date appears as a best before date. FDA does not require an expiry date on supplements, but expects data to support any date a label carries.

In-use stability

In-use stability covers the opened pack, which takes in moist air, oxygen and microbes with every dose. We follow the European Medicines Agency's note for guidance on in-use stability testing, written for medicines in multidose containers: at least two batches, one towards the end of its shelf life; doses removed at intervals that simulate real use; storage as the label directs; and tests for appearance, actives, preservative and antioxidant content, and total viable count. The result sets the "once opened" statement on your label.

What fails first in gummies, liquids, probiotics and powders

Gummies

Gummies put actives through heat, acid, water and sugar at once. The cook exposes actives to heat, and high-methoxyl pectin gummies set at a pH of about 2.8 to 3.5. We add sensitive actives late and track water activity, texture, actives and breakdown products, such as creatinine in a creatine gummy, at every time point. In June 2025, SuppCo reported that 4 of 6 creatine gummies tested at an ISO/IEC 17025-accredited laboratory failed identity and potency testing, two with no detectable creatine. Every gummy batch also gets an active assay as standard. See custom gummy manufacturing.

Liquids, shots and drinks

Vitamin C oxidises in solution and speeds the breakdown of vitamin B12, and benzoates with ascorbic acid can form trace benzene, faster with heat and light (FDA), so a vitamin C liquid gets another preservation route. We confirm preservation by challenge testing to European Pharmacopoeia 5.1.3 or USP <51>, repeated on a stability batch at the proposed shelf life, as ICH Q1A(R2) sets for medicines. See liquid supplement manufacturing.

Probiotics

The active is a live count. Water activity matters most, followed closely by storage temperature (Fenster and colleagues, 2019), so blends stay below 0.2 total water activity, in low-moisture HPMC capsules and high-barrier packs with desiccant. The same review rules out PET bottles. Real-time studies run at the storage condition printed on the label, and the label count is the count at the end of shelf life. See probiotic supplement manufacturing.

Powders and capsules

Powders often fail physically before they fail chemically: hygroscopic mineral salts and fruit powders cake, spray-dried flavours turn sticky, and reducing sugars brown with protein, which also costs lysine. We test flow, caking, dispersion and colour alongside assay. Capsules trade water between shell and fill, and gelatine can cross-link with aldehydes over shelf life, slowing dissolution (Al-Tabakha, 2010), so we test disintegration to USP <2040> at each time point.

From the target shelf life to the date on your label

Stability work starts at the brief. Our research and development process runs it in five steps:

1. Brief. We agree the target shelf life, markets, climates, storage statement and pack.
2. Formulation. We choose stable forms and grades from our ingredient library, set water activity targets and plan overages.
3. Bench screening. Short accelerated holds rank prototypes, and sensory checks catch flavour fade early, part of our flavour development and taste masking work.
4. Pilot batch in the final pack. Formal accelerated and real-time studies start, and the initial shelf life and overages follow from their data.
5. Production. Every batch is tested by accredited partner laboratories as part of every order, and quality control releases every batch against its specification. Real-time data run to the printed date, and a change of supplier, pack or formula goes back through stability before your label changes.

Your formula is yours, and so is the stability file behind it. We keep three retained samples per batch for shelf life plus one year. Claims are reviewed alongside stability, because a claim's conditions of use depend on the dose that survives to the last day; see claims, evidence and regulation by market. R&D, including stability work, is a paid stage, included with an order of the right size. British VitaTech is ISO 9001, HACCP and GMP certified by BSI, and FDA registered. Start your project.

References: ICH Q1A(R2), stability testing of new drug substances and products (2003); ICH Q1B, photostability testing (1996); ICH Q1E, evaluation of stability data (2003); European Medicines Agency, note for guidance on in-use stability testing, CPMP/QWP/2934/99 (2001); European Commission, guidance on tolerances for nutrient values declared on a label (2012); US FDA, Small Entity Compliance Guide for dietary supplement CGMP (2010); Fenster and colleagues, Microorganisms (2019); Al-Tabakha, Journal of Pharmacy and Pharmaceutical Sciences (2010).