EDF’s assessment of a health-based benchmark for lead in drinking water

Tom Neltner, J.D.is Chemicals Policy Director

Health professionals periodically ask me how they should advise parents who ask about what constitutes a dangerous level of lead in drinking water. They want a number similar to the one developed by the Environmental Protection Agency (EPA) for lead in dust and soil (which is the primary source of elevated blood lead levels in young children). I usually remind them that EPA’s 15 parts per billion (ppb) Lead Action Level is based on the effectiveness of treating water to reduce corrosion and the leaching of lead from plumbing; it has no relation to health. Then I tell them that EPA is working on one and to hold tight. Admittedly, that is not very satisfying to someone who must answer a parent’s questions about the results of water tests today.

On January 12, EPA released a draft report for public comment and external peer review that provides scientific models that the agency may use to develop potential health-based benchmarks for lead in drinking water. In a blog last month, I explained the various approaches and options for benchmarks that ranged from 3 to 56 ppb. In another blog, I described how EPA’s analysis provides insight into the amounts of lead in food, water, air, dust and soil to which infants and toddlers may be exposed. In this blog, I provide our assessment of numbers that health professionals could use to answer a parent’s questions. Because the numbers are only a start, I also suggest how health professionals can use the health-based benchmarks to help parents take action when water tests exceed those levels.

EDF’s read on an appropriate health-based benchmark for individual action on lead in drinking water

When it comes to children’s brain development, EDF is cautious. So we drew from the agency’s estimates calculated by its model to result in a 1% increase in the probability of a child having a blood lead level (BLL) of 3.5 micrograms of lead per deciliter of blood (µg/dL).

EDF's assessment of a health-based benchmark for individual action on lead in drinking water
Age of child in home and type of exposureHouses built before 1950¹Houses built 1950 to 1978²Tests show no lead in dust or soil³
Formula-fed infant3.8 ppb8.2 ppb11.3 ppb
Other children 7 years or younger5.9 ppb12.9 ppb27.3 ppb

As you consider the appropriate number, keep the following in mind:

  • There is no safe level of exposure to lead in children’s blood. The public health goal is always 0 ppb but it is difficult to translate into action on an individual home. The health-based benchmarks are designed to help public health professionals provide practical advice to parents.
  • Since a health professional is typically focused on actions an individual should take, we selected a number representing a child having a less than 1% chance of having an elevated BLL.
  • We used 3.5 µg/dL instead of the Reference Value of 5.0 µg/dL established by the Centers for Disease Control and Prevention (CDC) in 2012 because CDC is expected to lower the level to 3.5 µg/dL later this year. This CDC Reference Value is referred to by EPA as the elevated BLL.
  • The levels are for the amount of lead in drinking water actually consumed. Water test results can vary based on a number of factors, including the sampling method. If a resident conducted the water test for a utility to ensure compliance with the Lead and Copper Rule, the sample was the first liter of water coming from the cold water tap at the kitchen sink after it sat in the faucet overnight. This sample most likely overestimates the exposure. However, if the home has a lead service line (LSL) or lead pipes in internal plumbing, actual exposure may be more or less than the level found in the first liter depending on the condition of the pipe. Finally, lead levels change with the seasons and water chemistry; a single test will miss these changes.
  • The numbers come from a draft report. EPA is seeking public comment and convening an expert peer review panel to provide guidance. When it issues proposed revisions to the Lead and Copper Rule, hopefully by the end of 2017, the numbers may very well change.
  • One of the reasons the health-based benchmarks for children between 1 and 7 years old are much higher than those for formula-fed infants is that they drink much less – less than half – the amount of tap water.

EDF’s suggestions for how health professionals can use the health-based benchmark

If the water tests are above the appropriate health-based benchmark, we suggest health professionals take the following steps.

Step 1: Always consider paint. Lead-based paint remains the primary source of lead exposure to children with elevated BLLs. Low income and minority children remain at the greatest risk. When the paint is disturbed or deteriorated, it may contaminate soil and dust that a child can ingest. If the home is built before 1978, educate the resident on the hazards and suggest testing the soil, floors and window sill. Provide them EPA’s Protect Your Family from Lead in Your Home pamphlet. It is far from perfect, but it is useful.

Step 2: Determine whether the home has an LSL. Call the drinking water utility to see if they know or check their website as more and more communities are making the information available on-line. When you are in the home, if you can, check the water line as it comes into the house. National Public Radio has a good demonstration of how to do this check. If you have a portable X-ray fluorescent (XRF) lead paint analyzer, you can use it instead of scratching the pipe. If the home has or may have an LSL, advise the resident to work with the utility to have it safely removed since it can unpredictably release high levels of lead that may be missed by a one-time water test. Some communities are offering innovative ways to help offset the replacement costs. Until the LSL is replaced, suggest the resident periodically test the water and consider a filter to remove lead in the drinking water, especially if young children, pregnant women, or a formula-fed infant lives there. Flushing is a lower cost alternative, but may take several minutes to clear the water that has been sitting in the LSL. If there is no LSL, only a few seconds of flushing may be needed.

Step 3: Understand the sample method and results. Testing is typically performed on water samples taken from the first liter coming from the cold water tap at the kitchen sink after it sat in the faucet overnight. This method is how utilities evaluate compliance with EPA rules and, unless there is an LSL, may overestimate typical lead levels in the water. Consider taking another sample after flushing the line and letting it sit for 30 minutes. It should have non-detectable levels of lead and demonstrate the benefits of flushing before use. You may also want to test the water from the bathroom where children may get their water during the night.

Step 4: Provide basic educational materials to help residents reduce lead in their drinking water. The National Drinking Water Alliance has useful materials on its website including excellent fact sheets for renters and condo owners and for homeowners.


Testing water can help residents better understand how much lead may be in the water they drink. The results are most useful when compared to a health-based benchmark that is relevant to the resident’s situation. Health professionals can help individuals understand the health risks associated with the lead levels, and provide useful suggestions for them to take appropriate steps to reduce their exposures and be more confident that the water they and their children drink is safe.

¹ For homes without lead in dust or soil, we used Exhibit 22 with a lead in soil and dust of 0 µg/g.

² For pre-1950 homes, we used EPA’s Exhibit 50, which is based on the geometric mean of lead in soil of 221 µg/g (Exhibit 7) and dust of 134 µg/g (Exhibit 8).

³ For homes built after 1950, we used Exhibit 22 with a lead in soil of 37 µg/g and dust of 72 µg/g. This is similar to the levels reported by EPA for housing built after 1950 of 23 µg/g and dust of 63.7 µg/g.

This entry was posted in Drinking Water, Emerging Science, Flint, lead and tagged , , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.