The practical answer
A useful approach to supporting a finished-beverage dating period is a protocol written before the first pull, using the commercial container and closure, agreed attributes and named methods, and a recorded decision against criteria set in advance. This page describes that approach, not the only legally permitted study design. It does not assign a shelf life to any drink. Who owns the study, how ingredient data relates to the finished drink, and the planning choices for attributes and storage are covered in the shelf life testing plan. The sampling calendar schedules dates; it does not predict results.
Write the protocol before the first pull
The protocol should state the product identity, the batch or lot and the number of batches to be placed on study, the pack, the storage conditions, the pull points, the attributes, the methods, the number of units at each pull, the acceptance criteria and the rule for reaching a decision. Sign and date it before time-zero testing. If a later change is needed, record the amendment and whether already-pulled data remain valid under the new rule.
Name the attributes and the methods
List each attribute with the method, the instrument or laboratory, the sample preparation and the reporting units. For elemental magnesium, name the method and whether the sample is mixed and whether sediment is included, because those choices change the result. For pH, state the temperature, whether the sample is degassed and the procedure used. For appearance, say whether haze is read on a stated turbidimeter, judged against a reference, or both. For flavor, name the panel type and what counts as a meaningful change. Include microbiology when the product, process, distribution risk or food safety plan makes it relevant.
For a conventional beverage with added magnesium, 21 CFR 101.9(g) treats added nutrients in fortified or fabricated foods as Class I, and the nutrient content of the composite must be formulated to be at least equal to the declared value. The magnesium criterion in the protocol should reflect that rule. [1]
Use the commercial container and closure
Place the study in the container, closure, fill volume, headspace and light protection that will be sold, not in a convenient laboratory bottle. The packaging, storage and stability page covers package evidence more generally; this protocol step is specifically to make the study represent the commercial pack.
For a dietary supplement, 21 CFR 111.83 requires reserve samples of each distributed lot to be held in the same container-closure system in which the packaged and labeled supplement is distributed. If distributing supplements for packaging and labeling elsewhere, the reserve-sample container-closure system must provide essentially the same characteristics to protect against contamination or deterioration as the system used to distribute them to that location. Samples must be retained for one year past the shelf life date when a shelf life date is used, or for two years from the date of distribution of the last associated batch if no shelf life date is used. That retention rule is a supplement CGMP requirement. It does not prescribe a shelf-life study method for either a conventional beverage or a supplement. Record the commercial pack and the dated claim together in the study file when the product is a supplement, separately from the reserve-sample obligation. [2]
Set conditions, pulls and replicates in the protocol
The decision condition should be the intended or labeled storage condition, such as ambient storage for a shelf-stable drink or refrigeration for a chilled one. Record temperature, humidity and light, noting humidity especially if the pack can exchange moisture, as well as orientation into the protocol, and record actual chamber or warehouse readings.
ICH Q1A(R2) is pharmaceutical guidance for new drug substances and products. It is not a legal requirement for beverages. Some teams use it only as a reference framework. Its general-case examples include long-term storage at 25 °C / 60% relative humidity or 30 °C / 65% relative humidity, and accelerated storage at 40 °C / 75% relative humidity. It also notes that humidity is not a concern for products in impermeable containers, while aqueous products in semi-permeable containers should be evaluated for water loss. Those examples can inform a beverage protocol for product packed in glass, cans or PET. They do not authorize a beverage date from accelerated data alone. [3]
Write the pull points and the number of replicate units and batches into the protocol. The sampling calendar can turn those points into dates. Keep extra units for retests. In the approach recommended on this page, elevated-temperature results provide early signals and dating is assessed against data at the intended or labeled storage condition. This does not exclude accelerated or bridging evidence whose applicability has been validated for the specific product, failure mechanisms, packaging and conditions.
Record the decision against the criteria you set
At each pull, compare results with the pre-set criteria and record pass or fail, and open an investigation when needed. At the end of the planned period, record whether the data support the proposed date, support a shorter date, or do not support a date, with the owner and the date of the decision. Do not assign a shelf life from a trend or an accelerated-storage result that the protocol did not define as decisive.
For most foods in the U.S., a quality date is not required by federal regulation, except that infant formula must carry a use-by date. The USDA FSIS food product dating page is a general explanation of date labeling; it is not an FDA beverage regulation. When a brand prints a date, relevant stability evidence should support the proposed dating period and the intended label and quality criteria. A written protocol and study file help document that assessment; they are not a guarantee for every unit or the sole legally permitted evidence pathway. [4]
Common questions
Can an accelerated storage condition set the beverage date by itself?
Not automatically. An accelerated result does not by itself establish a valid beverage date. Accelerated or bridging evidence needs a validated basis applicable to the specific product, failure mechanisms, packaging and conditions. This page recommends assessment against intended or labeled storage data and pre-set criteria as one study approach; ICH Q1A(R2) does not establish a universal legal study requirement for beverages.
Does 21 CFR 111.83 tell a conventional beverage how long it may be dated?
No. That section is a dietary-supplement reserve-sample rule. It does not assign a shelf life to a conventional drink.
Sources & further reading
- 21 CFR 101.9(g), nutrition labeling of food, compliance classes
- 21 CFR 111.83, requirements for reserve samples
- FDA, Q1A(R2) Stability Testing of New Drug Substances and Products
- USDA FSIS, Food Product Dating
