Division of Environmental Health Food Safety and Sanitation Program

Bottled Water Testing

Bottled Water Sampling Requirements

The Alaska Department of Environmental Conservation (ADEC) requires that the operator of establishments that bottle water conduct routine sampling in accordance with 18 AAC 31.740. The information presented below provides basic information only and should not be used as a complete resource on bottled water requirements.

How Often Do I Need to Sample?

Weekly - you must collect samples of each type of finished bottled water for fecal coliforms or E. coli testing.

Quarterly - you must randomly collect samples for aerobic plate count analysis of at least four containers and four closures. The containers and closures are sampled by either using a swab of the container and cap, or a rinse of the container before filling or sealing the bottle.

How Do I Submit a Water Sample?

Check with the laboratory that you will be working with for direction regarding the type of sample containers to use and how much water will need to be collected. For more information please see the Guidance for Quarterly Bottled Water Container and Closure Requirements.

What Are the Levels At Which My Samples Would Not Meet the Requirements?

Weekly samples must be free of coliforms.

Quarterly samples must be free of coliforms, and no more than one of the four samples may exceed:

• one bacteria per milliliter of capacity (preferred and easiest method); or
• one colony per square centimeter of surface area.

What Should I Do If I Get an Unacceptable Result?

If the results of weekly sampling are positive for coliforms you must:

• immediately notify the department
• submit additional samples from the same lot as the initial positive sample (or the next available lot if initial lot if not possible) for analysis within 24 hours or next business day of receiving notice that the initial sample was positive;
• detain the product pending the results of the additional samples;
• notify the department of results of the additional samples; and
• keep a record of any actions taken.

If the results for quarterly bottle and closure samples exceed limits you must:

• review your transportation, storage, and bottling procedures to find the source of the contamination;
• immediately correct any deficiencies found; and
• keep a record of your review.

What Labs Are Certified to Test Bottled Water?

A laboratory to meet the bottled water testing must be certified in both a coliform method and a heterotrophic bacteria method. Find a list of certified laboratories.

Retaining Testing Records

Results of all sampling must be maintained for two years. You may keep the records on-site at your facility or another location but the records must be available within 24 hours after a request is made by the department.

Guidance for Quarterly Bottled Water Container and Closure Requirement

This guidance document describes an example of an approved rinse procedure that meets Alaska Food Code regulatory requirements; which is intended to be protective of public health, less costly to the operator, and easier for the lab to test. This guidance is approved by the Food Safety and Sanitation (FSS) program as an acceptable protocol for the required quarterly container and closure sampling and may be used, as written, without additional consent. Variations and other procedures that could satisfy the regulation are possible, but must be submitted to FSS for review to ensure sampling will be acceptable for compliance purposes.

FSS requires that the operators of establishments that bottle water conduct routine sampling in accordance with 18 AAC 31.740. This includes a quarterly test of at least four random container and closure assemblies. The containers and closures are sampled by either a swab of the container and cap, or a rinse of the container before filling or sealing the bottle‐which is the example described in this guidance document.

The chosen laboratory needs to have certified methods for total coliform and heterotrophic bacteria (HPC). This guidance is suited to laboratories that perform HPC by pour plate or by IDEXX© Multi‐Dose SimPlate® (the rinse procedure described below does not work for unit dose SimPlate®).

Contact the chosen laboratory, ask if they have methods and procedures available to support quarterly bottled water tests, and then request sampling bottles and laboratory specific directions. These bottles will likely look the same as the required weekly coliform bottled water sampling bottle.

Instructions For Bottled Water Processor

You will be required to test at least four assemblies quarterly. For each test, randomly select a container and closure assembly that represents your product line. A container and closure assembly is simply one of your empty bottled water containers you use for packaging your product in.

• The four sampling bottles from your chosen laboratory will arrive sealed and sterile with a preservative inside. Assign one of the lab issued sampling bottles to one container and closure assembly. Clearly label the sample bottle according to your laboratory’s instructions.
• Aseptically fill one lab issued sample bottle to the 100 ml line, taking care not to exceed 102 ml. Use your main water source (where you fill the water bottles during production). An aseptic fill means to fill in a manner that prevents contamination of the container and maintains sample integrity. Don’t put your fingers or any utensils inside the sampling bottle or container and closure assembly. Any amount less than 100ml will not be accepted by the laboratory.
• Pour 100ml of water from the lab issued sample bottle into the container and closure assembly. (Do not discard lab sample bottle, save aseptically for step 5.)
• Place the closure on the container, rinse the water around so that the entire interior surface of the bottle, including the cap, is rinsed with the sample water.
• Remove the closure and aseptically pour the sample water back into the lab issued sample bottle and replace the cap.
• This completes one sample for submission. Repeat the procedure for each of the three remaining container and closure assemblies.
• On the laboratory testing order form (a.k.a. Sample Submission Form or Chain of Custody), you must remember to indicate the volume of the container and closure assembly. Without this information the laboratory may reject the sample or call for clarification which could delay the analysis.

Laboratory Guidance

At the laboratory, this guidance works when using either pour plate or multi‐dose SimPlate® method for HPC (the rinse procedure prescribed above does not work for unit dose SimPlate®).

The HPC should be set up first. This will take about 2 mls from the submitted sample. If it was properly filled to 100ml line, that will leave you just enough (98mls) to run a coliform test. Using an IDEXX© coliform media is the easiest as it will use the same bottle and remaining volume with the least amount of additional steps. A satisfactory coliform result can be reported as ‘Non‐Detect’ or ‘Absent’.

Reporting units for HPC are generally in cfu/ml or MPN/ml depending on method performed. In this example, the HPC units of bacteria/ml, MPN/ml, and cfu/ml will be considered equivalent.

The regulatory units for the rinse of bottled water containers are bacteria per milliliter of capacity, so the retail volume of the container and a conversion is necessary to report for regulatory use.

First, the retail capacity of the container needs to be converted to milliliters (ml).1 gal = 3785 ml5 gal = 18,927 ml

The regulatory limit per 18 AAC 31.740(j)(1): “… more than one bacteria per milliliter of capacity…”

Therefore, for a 1 gal container an HPC result of 38 cfu/ml is unsatisfactory and for a 5 gal container an HPC result of 190 cfu/ml is unsatisfactory.

Regulatory References

• 18 AAC 31.740 Bottled drinking water.

Resources and Contacts