Health Stream Literature Summary - Issue 47 - September 2007

Elevated lead in drinking water in Washington, DC, 2003-2004: The public health response.

Guidotti, T.L., Calhoun, T., Davies-Cole, J.O., Knuckles, M.E., Stokes, L., Glymph, C., Lum, G., Moses, M.S., Goldsmith, D.F. and Ragain, L. (2007) Environmental Health Perspectives, 115 (5); 695-701.

In 2002, lead levels in treated water supplied by the District of Columbia Water and Sewer Authority (DCWASA) began to rise. The increase in lead concentrations occurred following the change from chlorine to chloramine water disinfection treatment on 1 November 2002 . The change in disinfectant was made for reasons of compliance with the pending Disinfection Byproducts Rule. The rise in lead levels was abrupt thereafter and at its peak in early 2004, the 90 th percentile value for homes sampled was 59 ppb. The lead action level (LAL) stipulates that the 90 th percentile of samples cannot exceed 15 ppb, but during this period 68% of homes exceeded 15 ppb on the first draw sample and some exceeded 300 ppb.

Lead service lines are still present in a wide range of older housing types in the District of Columbia . The change in disinfection from chlorine to chloramines altered the leaching of lead from the interior surface of lead service lines and caused lead levels in tap water to rise. Other sources of lead from within households may have included solder in joints between copper pipes, older faucets, and certain types of water meters. In 2003 the DCWASA implemented plans for families living in homes with lead lines or testing above the LAL including: advisories, specific advice to limit exposure, filters, replacement of private segment of lead service lines on their property at cost, replacement of public segments of all lead service lines, low-cost financing to replace private parts of lead service line and free water testing, Lead levels were reduced in the distribution system by adding a commonly used passivating agent to the water. Health advisories associated with the elevated lead levels were lifted in January 2006. The aim of this study was to evaluate the public health implications of the high lead levels by examining blood lead levels in children.

A blood lead screening program to supplement the existing clinical screening program in the District of Columbia was started on the 3 February 2004 and was discontinued on the 2 August 2004 . Two groups were identified, the “target population”, which included children 6 months to 6 years of age and women who were pregnant or nursing and the “outside the target population” which included all others for who testing was requested. Elevated blood lead levels were defined as those greater than 10 microg/dL, the level of concern adopted by the Centers for Disease Control and Prevention. The homes of all of those with elevated blood lead levels were investigated by the District of Columbia Department of Health (DC DOH).

There were 6,834 persons screened for blood lead level. Of these subjects, 2516 were within the target population; 2,342 children less than 6 years of age, 96 pregnant women and 78 women who were nursing. Of the 2,342 children in the target population, 65 (2.8%) had blood lead levels greater than10 microg/dL but all except one had levels less than 45 microg/dL, a level that may be associated with clinically symptomatic lead poisoning. Not all of the target group with elevated blood lead levels were children, two were nursing mothers. None of the 96 pregnant women had elevated blood lead levels. Most of the children who had elevated blood lead levels (70.8%) did not live in homes with lead service lines. The home investigations of those with elevated blood lead levels revealed that in most cases at least one source of lead exposure other than drinking water, usually peeling lead paint and dust, could be identified.

Of those children less than 6 years of age who had blood lead levels less than 10 microg/dL, the blood lead level (mean ± SD) for the 344 children who lived in homes with lead service lines was 3.28 ± 2.05 microg/dL compared with 2.60 ± 1.69 for children living in homes without lead service lines, a statistically significant difference ( p less than 0.05). Of those 4,318 residents who were outside the target population, 4 had blood lead levels greater than 25 microg/dL, the level of concern for adults. Two of these residents had lead service lines and two did not.

Of the 2,482 children less than 6 years of age who were tested for blood lead levels in 2004, first draw water lead concentrations were available for 107. A correlation analysis was performed on this data set and there was no correlation found ( r 2 = -0.03142). A data set was obtained from the DC WASA which included 71 children with blood lead levels of greater than or equal to 10 microg/dL for which paired blood lead levels and water concentrations were available. A correlation analysis was also performed on this data set and there was no correlation ( r 2 = -0.0856 for individuals, r 2 = -0.05639 for all data points and r 2 = -0.09728 for all addresses).

There appears to have been no identifiable public health impact from the elevated lead levels in the drinking water in Washington DC in 2003 and 2004. This may be a reflection on the effectiveness of measures to protect the residents such as compliance with recommendations to filter water. The screening program did reveal the actual situation of a continuing problem with sources of lead in the homes, specifically lead paint. This study cannot be used to correlate lead in drinking water with blood lead levels directly for various reasons: it is based on ecological data instead of individual exposure assessment, the protocol for measuring lead was based on regulatory requirements rather than estimating individual intake, there were interventions introduced to mitigate the effect, exposure from drinking water was confounded by other sources of lead in older homes, the period of potential exposure was not uniform among houses, and actual exposure was variable for individual residents.