Overview
This field study tested flushing as a lead control strategy at 50 New Jersey public school buildings. Each building contributed two drinking fountain samples at three time points: first draw after overnight stagnation, after a 10-minute morning flush, and at lunchtime (3 to 5 hours after the flush). First-draw lead concentrations were high enough to trigger concern, but a 10-minute flush produced a modest reduction in median lead. By lunchtime, lead had rebounded to near-first-draw concentrations, demonstrating that a one-time morning flush provides no lasting protection for children using fountains later in the day.
Key numbers
Lead concentrations (mg/L; note: 1 mg/L = 1,000 µg/L):
- First draw: range below detection to 0.075 mg/L (0 to 75 µg/L); median 0.010 mg/L (10 µg/L)
- 10-minute flush: range below detection to 0.135 mg/L (0 to 135 µg/L); median 0.007 mg/L (7 µg/L)
- Lunchtime: range below detection to 0.074 mg/L (0 to 74 µg/L); median 0.007 mg/L (7 µg/L)
The lunchtime median was not statistically different from the first-draw median, indicating full rebound within 3 to 5 hours.
By water corrosivity (Aggressive Index): corrosive water (AI below 10) produced significantly higher first-draw lead (median 0.015 mg/L, 15 µg/L) versus moderately corrosive water (median 0.008 mg/L, 8 µg/L). Flushed concentrations did not differ significantly by corrosivity category.
Methods (brief)
Sampling at 100 drinking water fountains across 50 schools in New Jersey. First-draw samples collected before the school day; 10-minute flush then conducted; lunchtime samples collected 3 to 5 hours later. Furnace atomic absorption (EPA Methods 239.2 and 220.2). Aggressive Index calculated from pH, total dissolved solids, and temperature. The EPA school action level at the time was 0.020 mg/L (20 µg/L); the community water system action level was 0.015 mg/L (15 µg/L).
Evidence fitness
Supports: characterizing lead concentrations at school drinking water fountains in NJ in the early 1990s; demonstrating that a single morning flush does not provide all-day protection because lead rebounds from fixture and pipe surfaces during subsequent stagnation.
Does not support: generalizing to modern school plumbing (post-2014 lead-free fixture requirements); quantifying lead at non-fountain outlets (kitchen sinks, laboratory taps); extrapolating to schools outside NJ or schools with different plumbing age profiles.
Limitations
Data are from the early 1990s; plumbing materials, corrosion control practices, and water chemistry have changed substantially since. The study measured fountains only, not all fixture types. Stagnation times between the morning flush and the lunchtime samples are approximated (3 to 5 hours). No adjustment for partial flushing volumes or flow rate variability. Detection limits not explicitly stated in the paper text; below-detection values treated as zero.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.