Are we trying to reinvent the wheel with pesticide testing in cannabis when other agricultural industries have been doing this for decades?

The requirement of pesticide and heavy metal screening on all intermediate and final products is unnecessary. The protection of public health could be achieved more efficiently, both financially and practically, by the random sampling of producer/processors at the farm level along with the random sampling of packaged product. Washington State could establish a voluntary testing protocol (with associated official label) to encourage producer/processors to have their product tested. This would also provide incentive for labs to become certified for pesticide and/or heavy metal testing. Data from a voluntary testing program would generate scientific information for the validation and standardization of the testing of cannabis. This program would also provide crucial economic data for small business impact analysis.

The science of pesticide testing is still very young and hasn’t been researched to the level required for the determination of action levels or validity of all results. It is premature based on the state of knowledge of testing the cannabis matrix for these compounds. This is demonstrated by the results from the Fall 2018 ILC/PT (Emerald Scientific) for 58 of the pesticides listed in WAC 314-55-108. 31 participating labs had difficulty accurately determining pesticide levels in a spiked hemp matrix. Composite analysis of all reported values shows a standard deviation range of 41% to 285% and the percent recovery range of 13% to 11,707% for individual pesticides. 29% of reported mean measurements were <80% or >120% of the assigned value. In addition, a study by the Smith et al. report showing 78% false negatives on pesticide testing in the cannabis matrix. (B.C. Smith, P. Lessard, and R. Pearson, Cannabis Science and Technology 2(1), 48-53 (2019)).

Alternative Pesticide and Heavy Metal Testing Program

Given the current state of the industry, we’ve outlined substitutes for each level of testing that we believe better service consumers and testing labs.

Random sampling of Producers (farms): Soil, plants, hydroponic system, spraying equipment. These tests would help detect banned pesticides.

Random sampling of packaged product: Packaged product collected from producers/processors before shipping to retain outlets to maintain chain of custody. Retail sampling could be added but chain of custody may be more difficult to maintain. These samples will test for banned pesticides and determine that approved pesticides are below action limits.

Random sampling would allow for a data base to be generated that could be analyzed by a workgroup or a third-party to determine the ability of labs to detect pesticides in the cannabis matrix. This would allow for the accurate determination of appropriate action levels. It would also provide information on frequency of false negatives and positives.

Enhanced enforcement, audit, validation of:

  • 134-55-108 (action levels, sampling by WSDA, LCB)
  • 134-55-087 (pesticide logbooks and application procedures.)

Sampling, possible alternatives:

  • Sampling is performed by State agency (LCB, WSDA, WDoE)
  • Third-party sampling
  • Producer/processor sampling

Cost of sampling:

  • Producers/processors pay a predetermined licensing fee equivalent to the estimated cost of testing
  • Producers/processors pay testing lab directly
  • Samples processed by State or certified lab (bidding process; even distribution of samples to certified labs)


Results reported directly to the LCB or managing State agency for determination of actions required (recall, retest, no action). This would limit the liability of testing labs and the State takes responsibility of required action.

State accredited pesticide/heavy metal compliant label:

The LCB establishes a voluntary testing program for lot level testing. Producers/processors who choose to have this test performed and then pass, can place an official State label on their product (e.g. Pesticide Compliant).

This will create a market to encourage more certified labs to participate in pesticide/heavy metal testing. Producers/processors and retailers will have a value-added product to help pay for the cost of testing.


All samples that fail initial screening for pesticides should be re-tested to confirm. A new sample should be obtained. One option would be to send the sample to a different lab or State lab.

State should validate and certify results of labs before issuing a product recall.

QA Sample Packaging:

Due to ease of cross-contamination of samples by pesticides, new stringent QA sampling packaging rules should be in place.