Monday, August 24, 2009

Letter to City Council - Why the Variance tests are designed for failure

August 24, 2009
Via email

Dear Commissioners –

I am writing to share with you the latest information obtained by the Mt. Tabor Neighborhood Association Land Use Committee regarding the year-long, high volume, finished drinking-water study Portland completed in May. This information, and the study’s report of findings yet to be published, should bolster our efforts seeking relief from LT2 and its unnecessary build requirements. It should also serve to caution us as the Water Bureau moves forward with any Variance testing options offered by the EPA.

This multi-year study, which Portland joined a year ago, is being managed by the Water Research Foundation (see the project abstract online at: ). It was designed to examine the finished drinking water for the presence of infectious Cryptosporidium.

As we understand it, participants in this current WRF study were to be immediately notified if at any time during the data collection process infectious Crypto was found in a water sample. The City of Portland was never notified of any such presence, so we should be able to extrapolate that no infectious Crypto was found in our finished drinking water samples. While we won’t have the final report on this study for some months, we do have a preliminary final report from November 2008. After repeated document requests from a member of our neighborhood we finally have access to this report (just last week) which the Water Bureau has seemingly had since November 2008 (report is attached as PDF). I would like to highlight for you three items from this report:

1) The report discusses the fact that there are many types of Crypto in the natural environment, but that not all genotypes are infectious to humans. We know infectious Crypto genotypes to be from human and domesticated animal sources - not the wild animals present in the Bull Run Watershed.
The genus Cryptosporidium contains at least 16 recognized
species that infect a variety of vertebrates… However, most cases of human
cryptosporidiosis are attributed to C. Parvum and C. Hominis.

There are
currently two species of Cryptosporidium that cause the majority of human
infections, C. parvum and C. hominis. However, the source of contamination of
environmental waters is often livestock or feral animals that can shed species
of oocysts that are not infectious to humans and so represent minimal public
health risk.
2) The report affirms that appropriate testing protocols must gather more information than just the presence/absence of Crypto oocysts when analyzing water and assessing public health risks. Appropriate testing must confirm the genotype, whether the oocyst is even active (alive), as well as the overall condition (or health) of the oocyst. Counting oocysts without adjusting for these other factors will not produce accurate public health data.
The condition of the oocysts is also very important in determining the risk of infection. Oocysts are exposed to many conditions in the environment that can reduce their infectivity… The length of time post-shedding from the carriage animal, water temperature, and the amount of ultraviolet (UV) exposure from sunlight can reduce oocyst infectivity… In addition, surface
waters are exposed to natural UV irradiation in sunlight which may damage oocyst
DNA thereby inhibiting DNA replication and reducing infectivity.

3) But perhaps most interestingly, this report specifically calls attention to the chief criticism of EPA’s LT2-Variance testing protocols – these test methods set forward by the EPA (and agreed to by the Water Bureau) wrongly ignore an analysis of the many factors (genotyping, condition, etc.) that appropriately define the public health risk of detected Crypto oocysts.
However, since many oocysts in surface waters belong to
species other than C. hominis and C. parvum, the public health benefits of the
risk assessment framework underlying the LT2 ESWTR, based solely on
FITC-positive oocysts with no specification or genotyping may be questioned.

The current methods of Cryptosporidium detection in untreated surface
water (Method 1622 and 1623; US EPA, 2005) use an antibody based detection
method to identify oocysts. This method only provides presence/absence detection
of oocysts…The detection of non-infectious oocysts or oocysts belonging to a
species that is not infectious for humans could cause unwarranted concern for a
contaminant that may not be a significant public health risk.

We should proceed with caution into any Variance testing options offered by the EPA that do not allow us to genotype or assess the condition of an oocyst if detected in a test sample – at present, all testing options offered by EPA to date DO NOT allow for genotyping or conditional analysis. The Water Bureau is poised to begin the Variance tests in just two weeks, using the very EPA method -- Method 1623 -- criticized in the WRF study (quote above). Crypto is common in the environment, but infectious Crypto is limited and infectious Crypto is what all of our public health agencies should care about. Any test that does not allow us to genotype or assess the condition of an oocyst is a test designed to produce failure through false positives.

S. Stewart
Mt. Tabor Neighborhood Association Land Use Co-Chair