The Power of Information

National Drinking Water Database


National Drinking Water Database - Methodology

Data Sources and Methodology


December 2009

Summary

EWG's National Tap Water Quality Database includes water quality tests conducted by 47,667 utilities nationwide from 2004 to 2009. We obtained the water quality testing data from state health and environmental departments that compile the records from drinking water utilities or associated laboratories. Most state water offices maintain a current record of water utility test data to fulfill their role as primary enforcement agents.

We requested data from all 50 states and the District of Columbia, and received data from 45 states and the district. We compiled the information into an online, interactive drinking water quality database, the largest such resource in existence.

Below we describe data sources, data quality, and the methods of data analysis used to generate facts and figures provided in this study, including the criteria we used to rank drinking water utilities.


Data acquisition - sources and coverage

EWG received tap water contaminant data from state water officials in 45 states and the District of Columbia over a three year period. The overwhelming majority of the data are test results for individual chemicals, from analyses conducted by water utilities to satisfy state and federal test requirements. Less frequently tests in the datasets were conducted by state agencies in short-term monitoring programs.

Some of the states provided data in a standardized format developed by the U.S. Environmental Protection Agency called SDWIS-STATE (Safe Drinking Water Act Information System for States). Certain states and the District of Columbia sent data in non-standard formats unique to each state agency. EWG synthesized all the sources into an integrated database. We posted the data and results from our analyses of it in an online, interactive resource that state health officials, scientists, media and the general public can access free of charge.

Altogether, EWG's drinking water quality database includes nearly 20 million test results, making it the largest such database in existence.

States unable to provide affordable, high quality data

Nine states and the District of Columbia failed to provide affordable, high quality data. EWG researchers requested the data from each of these states' water offices multiple times by email and telephone, between 2007 and 2009.

  • Two states demanded fees for the data that were beyond EWG's budget. Alabama requested a payment of $13,350.60 for its drinking water quality data, while West Virginia set its "data processing" fee at $15,000. EWG's current database contains 2004-2005 test results for utilities in both these states from data the states provided free of charge for our 2005 publication entitled "EWG's National Tap Water Quality Atlas."

  • Water offices in five states failed to respond to multiple requests for data via both email and telephone — Colorado, Georgia, Mississippi, Nevada and Tennessee.

  • Water offices in Kansas and Louisiana committed to sending their water quality data, but did not do so despite multiple additional requests from EWG researchers.

  • Water officials at Washington D.C.'s utility do not maintain electronic records of tests except for disinfection byproduct analyses. The utility provided paper copies of their testing data for a single quarter totaling hundreds of pages. It was beyond EWG's resources to transfer this data to an electronic format suitable for our project. The utility provided limited electronic records for disinfection byproducts, which are included in our database.

Data quality

After assembling our tap water monitoring database, we reviewed the data for inconsistencies, systematically searching for test results outside the expected range or other unique properties indicating potential errors. EWG contacted state water agencies to discuss the potential errors found and corrected many thousands of test results as a result of information supplied by the states.

We also briefed major water utility trade associations (American Water Works Association, or AWWA, and Association of Metropolitan Water Agencies, or AMWA). We provided to their members with a three-week window to review the state agency data we had compiled, as well as the Safe Drinking Water Act violations information from U.S. EPA, and make any necessary corrections.

To expedite the review, EWG analysts flagged individual test results that were 40 times higher than the national average so that utility representatives could confirm or correct information that appeared to fall outside of the expected range.

More than 200 water utilities reviewed the testing and violation data for their system and either verified the accuracy of the data or provided corrections.

Health guidelines

EWG assessed the levels of contaminants found in U.S. tap water against the following legal limits and health guidelines published by government agencies:

  • Maximum Contaminant Level (MCL) — The enforceable standard that defines the highest level of a contaminant that is allowed in drinking water. MCLs are set as close to health-based limits (Maximum Contaminant Level Goals, or MCLGs) as feasible using the best available analytical and treatment technologies and taking cost into consideration. Source: U.S. Environmental Protection Agency, http://www.epa.gov/safewater/contaminants/index.html

  • Maximum Contaminant Limit Goal (MCLG) — A non-enforceable health goal that is set at a level at which no known or anticipated adverse effect on health occurs and that allows an adequate margin of safety. Source: U.S. Environmental Protection Agency, http://www.epa.gov/safewater/contaminants/index.html

  • One-in-10,000 cancer risk limit — The concentration of a chemical in drinking water corresponding to an excess estimated lifetime cancer risk of 1 in 10,000. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • One-in-one-million cancer risk limit — The concentration of a chemical in drinking water corresponding to an excess estimated lifetime cancer risk of 1 in 1,000,000. Based on one in 10,000 (10-4) cancer risk calculated by the EPA. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • California Public Health Goals (PHGs) — Defined by the State of California Office of Environmental Health Hazard Assessment (OEHHA) as the level of contaminant that pose no significant health risk if consumed for a lifetime. For acutely toxic substances, levels are set at which scientific evidence indicates that no known or anticipated adverse effects on health will occur, plus an adequate margin of safety. PHGs for carcinogens or other substances that can cause chronic disease are based solely on health effects without regard to cost impacts and are set at levels that OEHHA has determined do not pose any significant risk to health. Source: http://oehha.ca.gov/water/phg/allphgs.html

  • EPA human health water quality criteria — Water quality criteria set by the U.S. EPA to provide guidance for states and tribes authorized to establish water quality standards under the Clean Water Act (CWA) to protect human health. These are non-enforceable standards based upon direct exposure by drinking water and the contribution of water contamination to other consumed foods. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/wqctable/index.html

  • Children's health-based limit for 1-day exposure — Concentration of a chemical in drinking water that is not expected to cause any adverse, noncarcinogenic health effects for up to one day of exposure. The One-Day health-based limit (or Health Advisory, HA) is typically set to protect a 10-kg child (approximately 1-year-old) consuming 1 liter of water per day. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • Children's health-based limit for 10-day exposure — Concentration of a chemical in drinking water that is not expected to cause any adverse, noncarcinogenic effects for up to ten days of exposure. The Ten-Day health-based limit (or Health Advisory, HA) is typically set to protect a 10-kg child (approximately 1-year-old) consuming 1 liter of water per day. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • Lifetime health-based limit, non-cancer risk — Concentration of a chemical in drinking water that is not expected to cause any adverse, noncarcinogenic health effects for a lifetime of exposure. The Lifetime health-based limit (or Health Advisory, HA) is based on exposure for a a 70-kg adult consuming 2 liters of water per day. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • National Secondary Drinking Water Regulations (NSDWR) — A National Secondary Drinking Water Regulation is a non-enforceable guideline for contaminants that may cause cosmetic effects or aesthetic effects (such as taste, odor, or color). Some states choose to adopt them as enforceable standards. Source: U.S. Environmental Protection Agency, http://www.epa.gov/safewater/contaminants/index.html

  • Regulated Contaminant Group — The enforceable standard that defines the highest level for a contaminant group that is allowed in drinking water. MCLs are set as close to health-based limits (Maximum Contaminant Level Goals, or MCLGs) as feasible using the best available analytical and treatment technologies and taking cost into consideration. Examples for which MCLs are established for groups of contaminants include the MCL for Total Trihalomethanes, which includes four individual chemicals, and for Total Haloacetic Acids, which includes five chemicals. Also included are breakdown products or forms of other chemicals that are regulated as a group (such as m-xylene or p-xylene). Source: U.S. Environmental Protection Agency, http://www.epa.gov/safewater/contaminants/index.html

  • Drinking Water Equivalent Level (DWEL) — A lifetime exposure concentration protective of adverse, noncarcinogenic health effects that assumes all of the exposure to a contaminant is from drinking water. Source: U.S. Environmental Protection Agency, http://www.epa.gov/waterscience/criteria/drinking/dwstandards.html

  • Health-Based Screening Level (HBSL) — A benchmark concentration of contaminants in water that may be of potential concern for human health, if exceeded. For noncarcinogens, the HBSL represents the contaminant concentration in drinking water that is not expected to cause any adverse effects over a lifetime of exposure. For carcinogens, the HBSL range represents the contaminant concentration in drinking water that corresponds to an excess estimated lifetime cancer risk of 1 chance in 1 million to 1 chance in 10 thousand. Source: U.S. Geological Survey, http://infotrek.er.usgs.gov/traverse/f?p=HBSL:HOME:0

How drinking water utilities were ranked

EWG rated big city (population over 250,000) water utilities based on three criteria:

  • Total number of chemicals detected since 2004.
  • Percentage of chemicals found of those tested for. Some states did not provide data on the full range of chemicals tested for but not detected. For purposes of these rankings, EWG assumed that utilities tested for the 80 contaminants regulated by EPA whenever fewer than that number were reported.
  • The highest average level for an individual pollutant relative to either:
    • legal limits (for regulated chemicals), or,
    • national average concentrations (for unregulated chemicals). We computed national average concentrations using data from utilities that reported detecting the compound.

To construct the final dataset used to rank utilities, EWG supplemented the state databases that underlie this national report with information from utilities' online 2008 water quality reports (the Consumer Confidence Report, or CCR) in two cases. First, we used 2008 CCR data for four common national pollutants - trihalomethanes, haloacetic acids, arsenic, and nitrate - when the state databases contained no tests for these pollutants for a given utility. Secondly, we used all reported detected concentrations from 2008 CCRs when state databases lacked test data more recent than 2005 for a given utility.

EWG calculated the final rankings as follows:

  • We scaled the values calculated for each parameter from 0 to 100 across all utilities assessed.
  • We assigned weights to these scaled values to account for each parameter's relative importance in the ranking system. The highest relative average pollutant level was assigned a weight of 0.5; the total number of contaminants found was assigned a weight of 0.3; and the percentage of chemicals found of those tested for was assigned a weight of 0.2.
  • We calculated final ranks by summing the individual weighted ranks for the three parameters listed above and then ranking the sums.

Data sources on drinking water contaminant toxicity

EWG created a core database of chemical hazards, regulatory status and study availability by pooling the data of more than 50 databases and sources from government agencies, industry panels, academic institutions, and accredited professional associations. Individual toxicity data sources we compiled are listed below:

Primary references - Known and probable carcinogens, reproductive and developmental toxicants

  • ACGIH (American Conference of Governmental Industrial Hygienists) 2008. ACGIH cancer classification system. www.acgih.org
  • California EPA (California Environmental Protection Agency). 2008. Office of Environmental Health Hazard Assessment (OEHHA). Safe Drinking Water and Toxic Enforcement Act of 1986. Chemicals known to the State to cause cancer or reproductive toxicity. http://www.oehha.ca.gov/prop65/prop65_list/Newlist.html
  • European Commission's Joint Research Centre. 2008. Classification and Labelling: Chemicals: Annex I of Directive 67/548/EEC. http://ecb.jrc.ec.europa.eu/
  • EPA (U.S. Environmental Protection Agency). 2008. Integrated Risk Information System (IRIS). Evidence for human carcinogenicity based on 1986-2005 guidelines. http://www.epa.gov/iris
  • IARC (International Agency for Research on Cancer). 2008. Overall Evaluations of Carcinogenicity to Humans, as evaluated in IARC Monographs Volumes 1-99 (a total of 935 agents, mixtures and exposures). http://monographs.iarc.fr/ENG/Classification/index.php
  • NTP (National Toxicology Program). 2005. Eleventh Report on Carcinogens, Eleventh Edition; U.S. Department of Health and Human Services, Public Health Service, National Toxicology Program. http://ntp.niehs.nih.gov/go/roc

Secondary references - Known and probable carcinogens, reproductive and developmental toxicants

  • CHE (The Collaborative on Health and the Environment). 2006. Toxicant and Disease Database. http://database.healthandenvironment.org
  • Environmental Defense Fund. 2006. Scorecard — The Pollution Information Site. http://www.scorecard.org
  • EPA (U.S. Environmental Protection Agency). 2006. EPA Water Disinfection By-Products with Carcinogenicity Estimates.http://www.epa.gov/nheerl/dsstox/sdf_dbpcan.html
  • Grandjean P and PJ Landrigan. 2006. Known human neurotoxins from: Developmental neurotoxicity of industrial chemicals. Lancet 368(9553): 2167-78.
  • NIOSH (National Institute for Occupational Safety and Health). 2006. NIOSH Carcinogens List (Potential occupational carcinogens). http://www.cdc.gov/niosh/npotocca.html
  • NTP (National Toxicology Program) 2006. Chemicals Associated with Site-Specific Tumor Induction in Mammary Gland. http://ntp.niehs.nih.gov/go/10908

Other human and environmental health databases (neurotoxicity, immunotoxicity, endocrine toxicity, ecotoxicity, etc.)

  • AOEC (Association of Occupational and Environmental Clinics). 2006. Asthmagen compilation - AEOC exposures codes. http://www.aoec.org/tools.htm
  • Environment Canada. 2008. Domestic Substances List Categorization. Canadian Environmental Protection Act (CEPA) Environmental Registry. http://www.ec.gc.ca/CEPARegistry/subs_list/dsl/dslsearch.cfm
  • EPA (U.S. Environmental Protection Agency). 1987 & 2005. Office of Pesticide Programs. Inert (other) Pesticide Ingredients in Pesticide Products - Categorized List of Inert (other) Pesticide Ingredients.
  • European Commission on Endocrine Disruption. 2007. EU (European Union)- Strategy for Endocrine Disrupters 2007. Commission on endocrine disruption requested by the European Parliament in 1998. http://ec.europa.eu/environment/endocrine/strategy/substances_en.htm
  • Illinois EPA (Illinois Environmental Protection Agency), 2000. Preliminary list of chemicals associated with endocrine system effects in animals and humans or in vitro. In EPA (U.S. Environmental Protection Agency), 2000. Handbook for Non-Cancer Health Effects Valuation, Appendix C.
  • NLM (National Library of Medicine). 2006. HazMap - Occupational Exposure to Hazardous Agents.
  • NTP (National Toxicology Program). 2006. NTP Center for the Evaluation fo Risks to Human Reproduction (CERHR). NTP-CERHR Reports and Monographs. http://cerhr.niehs.nih.gov/reports/index.html
  • RTECS (Registry of Toxic Effects of Chemical Substances). 2007. Developed by National Institute for Occupational Safety and Health (NIOSH), currently maintained by Elsevier/MDL.
  • Mammary Carcinogens Reviews Database. 2007. Rudel RA, KR Attfield, JN Schifano and JG Brody. Chemicals Causing Mammary Gland Tumors in Animals Signal New Directions for Epidemiology, Chemicals Testing, and Risk Assessment for Breast Cancer Prevention. CANCER Supplement. 2007 109(12): 2635-2666. http://sciencereview.silentspring.org/mamm_browse.cfm