Overview
This cross-sectional study measured water lead levels in 151 households with private domestic wells in three Illinois counties (Kane, Peoria, and Jackson). Private wells are unregulated for lead at the federal level; testing is rare and national prevalence data are limited. Lead was detected above the analytical quantification limit of 0.76 µg/L in 48.3 percent of first-draw stagnation samples. The strongest predictor was home age combined with corrosive water: homes built in 1986 or before with corrosive water had an odds ratio of 11.07 (95% CI: 3.47 to 35.31) for detectable lead relative to newer homes with non-corrosive water. Flushing (7th-liter samples) reduced detection prevalence markedly, indicating the lead source is primarily in-home plumbing rather than the aquifer.
Key numbers
Water lead level detection and exceedance (first-draw 1-L stagnation samples, µg/L):
- Detectable above LOQ (0.76 µg/L): 48.3 percent of homes (73/151)
- Exceeded 15 µg/L (US EPA community water system action level): 3.3 percent of homes (5/151)
- Exceeded 5 µg/L (Health Canada MAC as of 2019): not separately reported but can be derived from distribution
Water lead levels after 7th-liter flush:
- Detectable above LOQ: 22.5 percent of homes (34/151)
- Exceeded 15 µg/L: 0 percent of homes (0/151)
Screening strategy performance (older home AND corrosive water): sensitivity 88 percent, specificity 42 percent, positive predictive value 58 percent, negative predictive value 80 percent.
Blood lead dose context (Ettinger et al. 2014, as cited): a 1 µg/L increase in water lead corresponds to a 0.042 µg/dL increase in blood lead.
Methods (brief)
Cross-sectional study. First-draw 1-L samples after 6 h or more stagnation; 7th-liter flushed samples (after discarding 6 L). Graphite furnace atomic absorption spectrometry (EPA Method 200.9; Agilent 240Z with Zeeman background correction). MDL and LOQ: 0.76 µg/L. Non-detects replaced with LOQ divided by the square root of 2 for analysis. Larson-Skold Index and chloride-to-sulfate mass ratio (CSMR) calculated from water chemistry data to assess corrosivity.
Evidence fitness
Supports: establishing lead prevalence in private domestic wells in Illinois; identifying older housing and corrosive water as the dominant risk predictors; demonstrating that flushing reduces but does not eliminate detection; motivating testing recommendations for unregulated private well households.
Does not support: estimating lead concentrations in private wells in other regions (Illinois geology and infrastructure age profiles are specific); predicting individual home risk from water chemistry alone; applying community water system regulatory frameworks to private wells.
Limitations
Private well recruitment via press releases and partnerships with county health departments may oversample households with existing water quality concerns, introducing selection bias. The study was conducted in three Illinois counties; well aquifer chemistry, infrastructure age, and plumbing materials vary regionally. Stagnation samples were collected with the standard 6-hour protocol; shorter stagnation times (common in frequently occupied homes) would likely yield lower detection rates. The 7th-liter flush is more aggressive than the 5-minute flush tested in community system studies; direct comparison should account for this protocol difference.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.