Objectives. To detail baseline drinking water sample lead concentrations and features of US state-level programs and policies to test school drinking water for lead in 7 states’ operating programs between 2016 and 2018.

Methods. We coded program and policy documents using structured content analysis protocols and analyzed state-provided data on lead concentration in drinking water samples collected in public schools during initial testing phases.

Results. We analyzed data from 5688 public schools, representing 35% of eligible schools in 7 states. The number of samples per school varied. The proportion of schools identifying any sample lead concentration exceeding 5 parts per billion varied (13%–81%). Four states exceeded 20%. Other program features varied among states. Instances of lead above the state action level were identified in all states.

Conclusions. In 2018, many US public school students attended schools in states without drinking water lead-testing programs. Testing all drinking water sources may be recommended.

Public Health Implications. Initiating uniform school drinking water lead testing programs and surveillance over time could be used to reduce risk of lead exposure in drinking water. (Am J Public Health. 2022;112(S7):S679–S689. https://doi.org/10.2105/AJPH.2022.306961)

One in 5 children and young adults in the United States do not drink water on a given day,1 and more than half may be inadequately hydrated.2 Too many children routinely consume sugary beverages,3 while drinking water in place of sugary drinks can prevent excess weight gain4 and support critical body functions.5 Concerns about school water quality6 may inhibit consumption. Experts agree that there is no safe level of lead exposure,7 and some US school districts have discovered elevated levels of lead in drinking water.810 Drinking water is one way that children are exposed to lead, and lead in drinking water can increase children’s blood lead levels.11

Lead can enter drinking water when service lines that bring water into a building or components of interior plumbing systems like fixtures or fittings contain lead.10,12 Historically, federal oversight and monitoring of the lead concentration in drinking water in public schools has been limited.10 Federal laws regulate both testing and the allowable lead concentration in drinking water at schools that supply water for students via their own sources such as wells or cisterns.13,14 However, most schools (89%)15,16 obtain water from a community water system subject to federal oversight10,17 that guides corrosion-control practices to reduce lead in the water supply but historically did not require testing in schools.10 (Recent rule revisions include requirements for community water systems to test in schools and child care settings constructed before 2014.18) Generally, states provide oversight for ensuring safe drinking water within schools, and state approaches differ.10,13,19

To better understand how state programs monitor and limit potential health-related harms of lead exposure in drinking water in schools and whether programs might be suitable for surveillance over time, we documented and compared aspects of states’ initial water quality testing programs and policies to contemporaneous federal guidance20,21 and public health surveillance guidelines.22 We also characterized available baseline data on the lead content in water samples collected from drinking water outlets (called “outlets” herein) in schools initiating school drinking water testing programs.

We identified state-level programs and policies for drinking water lead testing in operation between January 1, 2016, and February 28, 2018, based on policy scans,10 through online searches and inquiries of state government officials in all 50 states and the District of Columbia (DC). We identified 23 states plus DC with programs and policies for drinking water quality initiatives. Of the 27 remaining states, contacts in 11 (41%) confirmed no program or policy, 1 identified an eligible program, and 1 identified a policy passed beyond our inclusion dates, resulting in 24 states and DC (herein, “25 states”) with eligible programs or policies. We reviewed content on state Web sites, program materials, and policy documents. State government officials provided water testing results or confirmed the status of publicly available data. States (6 states and DC, herein, “7 states”) with compiled, electronic (not paper) numeric water quality test result data from outlet-level samples, collected from no fewer than 15 schools, were included in this analysis (Figure A, available as a supplement to the online version of this article at https://www.ajph.org).

Rationale and Framework for Program Characterization

We documented key areas of policy or program alignment with contemporaneous recommended federal guidelines for lead testing in school drinking water.20,21 These included selection and number of outlets tested, water volume collected, length of time water was stagnant in pipes before the test sample was taken, groups notified of testing results, and remediation strategies recommended when elevated lead concentrations were identified. We identified these features as relevant within the framework of key surveillance system attributes.22 These included the following:

1. data quality based on recommended water testing guidelines noted previously;

2. the structure of operation, the leading agency, and whether the state had a policy or program for water quality testing;

3. acceptability of participation, whether funding was available for sampling, testing, or remediation;

4. sensitivity, the ability to capture an adequate proportion of affected outlets and monitor changes in the cases over time by recurrence of testing, the collection of school and outlet-specific numeric data, and the specified lead-concentration level threshold;

5. representativeness, the ability of the program to describe the distribution of the exposure in the population by place and school, including voluntary or mandatory participation, participation rate, and number and types of outlets sampled; and

6. stability, the ability to make reliable data available (i.e., data, particularly school outlet-level water testing data, online or upon request).

Two researchers (M. P. K. and L. V.) followed a standardized protocol to independently code the primary program or policy document and sampling protocol to identify these features of each state’s water sampling procedures. A summary profile was then generated and sent to designated state officials for review.19 Twenty-one states (84%) replied with edits and verified profile content.

School and Water Sample Data

To look at representativeness of available numeric data, we summarized data from water samples taken at public schools serving students in grades prekindergarten (PK) to 12 in initial testing waves conducted by programs in operation between January 1, 2016, and February 28, 2018. Researchers identified public schools that served students in any grades from PK through 12 with a physical campus location by matching schools from the state-provided data sets to national data available in the National Center for Educational Statistics (NCES) Common Core of Data (2015–2016) school directory using the school and district names and other public data sources.23

We identified 6375 organizations with water testing data within the 7 states. Of these, we identified 5850 as probable public schools serving grades PK through 12 and were able to match 5688 to schools listed in the NCES national census of public schools. Data for 162 (2.8%) of the probable schools could not be linked. The NCES data were used to characterize schools by level (primary: lowest grade served PK to 3, secondary: lowest grade served 4 to 12), metro status (city, suburban, town, rural), number of PK‒12 students enrolled (sum by grade PK through 12), percentage of students eligible for free or reduced-price meals, and percentage of PK‒12 students in each race/ethnicity category (non-Hispanic White, non-Hispanic Black, Hispanic/Latino, non-Hispanic Asian, non-Hispanic other). We documented the number of “first draw” samples, water taken from the outlet as soon as the fixture is turned on. We specified 3 threshold levels for lead concentration in parts per billion (ppb) in samples based on allowable lead content in bottled drinking water (5 ppb),24 standards specified by the World Health Organization (10 ppb),25 and action levels specified in contemporaneous federal guidance for public water suppliers’ corrosion-control measures (15 ppb).14 The primary outcomes were (1) whether each school had any first-draw sample from an outlet with a lead concentration above each of the 3 specified concentration levels and (2) the mean percentage of water samples within a school above each threshold. Recorded data that indicated that the outlet was not a drinking water source and records with no recorded value of lead concentration were excluded.

Data Analysis

Descriptive statistics summarizing school characteristics and school drinking water lead concentration outcomes were calculated for each state. We compared the characteristics of schools in the states with and without eligible drinking water lead concentration outcome data to describe how representative the states analyzed were among all states with testing programs. Then, within 7 states with eligible outcome data, we compared characteristics of schools that participated in the testing program to those that did not. For each state, we performed the χ2 test to compare school level and metro status, the t test to examine the number of students PK‒12 and percentage of students eligible for free or reduced-price meals, and the estimated global likelihood ratio χ2 test in logistic regressions to examine the percentage of students in each race/ethnicity category.

In post hoc analyses, we assessed whether observed differences in the number of first-draw tests conducted per school across states might have implications for testing outcomes. We selected 2 samples at random (with replacement) from tests conducted at each school 1000 times and calculated the average percentage of schools having any first-draw sample above each specified concentration level across all iterations in each state. We used SAS version 9.4 (SAS Institute Inc, Cary, NC) to perform analyses. We used a 2-tailed significance level of P less than .05.

Among the 7 states, 4 states had programs and 3 states had policies for lead testing in school drinking water. Lead-testing initiatives were often managed by state agencies overseeing environmental health. One state required schools to test for lead in school drinking water, while the other 6 states had nonmandatory programs or policies. All states covered the testing costs, though funding for sampling and treatment varied across states.

States had varied specifications for identifying the number and type of outlets tested per school, per building, or otherwise based on building construction age. All but 1 state had a 1-time, nonroutine testing initiative, meaning there would be limited capacity to evaluate changes over time in water lead concentrations or success of lead mitigation or remediation programs. Most states required some type of action be taken if lead concentrations exceeded 15 ppb at an outlet. The highest action level was 20 ppb, and the lowest was 5 ppb. All states included some level of guidance on remediation measures to be taken when lead concentrations at an outlet exceeded the state’s action level. Additional details on state programs can be found in Tables 1 and 2.

Table

TABLE 1— Program and Policy Features in 6 US States Plus the District of Columbia With Outlet-Level School Drinking Water Lead Concentration Results: 2016–2018

TABLE 1— Program and Policy Features in 6 US States Plus the District of Columbia With Outlet-Level School Drinking Water Lead Concentration Results: 2016–2018

AZ CA MA NV RI UT DC
Policy or programa Program Policy Program Program Policy Program Policy
Agency responsible for program management Arizona Department of Environmental Quality California State Water Resources Control Board Massachusetts Department of Environmental Protection Nevada Division of Environmental Protection Rhode Island Department of Health Utah Department of Environmental Quality District of Columbia Department of General Services
Mandatory or voluntary Voluntary Voluntary Voluntary Voluntary Voluntary Voluntary Mandatory
Funding source State government and city partners Not specified State government State government using a federal grant State government using a federal grant State Not specified
Funding covers cost of sampling Not specified Not specified Covered Covered, up to a certain amount Covered Covered Not specified
Funding covers cost of testing Covered Covered Covered Covered Covered Covered Covered
Funding covers cost of treatment and remediation Not specified Not specified Not covered Covered Not specified Not covered Covered
Groups notified of test results School staff, students, parents/guardians Parents/guardians School staff, parents/ guardians, general public Nevada Division of Environmental Protection receives the results and notifies school and school district Schools and general public School staff, students, parents/ guardians, government, general public General public
Strategies for remediation EPA-recommended strategies and referral to the Arizona School Facilities Board EPA-recommended strategies EPA-recommended strategies Short-term EPA-recommended strategies EPA-recommended strategies EPA-recommended strategies Program includes long-term remediation measures

Note. EPA = US Environmental Protection Agency.

aPolicy for water quality testing defined as a policy mechanism used to establish the program via state statute, executive order, or funding appropriation. Program for water quality testing defined as initiated by a state agency or department pursuant to an existing directive or grant of authority.

Table

TABLE 2— Program and Policy Data Collection and Sampling Protocol Features in 6 US States Plus the District of Columbia With Outlet-Level School Drinking Water Lead Concentration Results: 2016–2018

TABLE 2— Program and Policy Data Collection and Sampling Protocol Features in 6 US States Plus the District of Columbia With Outlet-Level School Drinking Water Lead Concentration Results: 2016–2018

AZ CA MA NV RI UT DC
Recurrence 1 time 1 time 1 time 1 time 1 time 1 time Routine, surveillance
Action level (ppb)a 15 15 15 20 15 15 5
Recurrence of testing process 1 time 1 time 1 time 1 time 1 time 1 time Annually
Settings Public schools (K‒12) Public schools, private schools, charter schools (K‒12) Public schools, charter schools (K‒12), licensed early childcare centers on school grounds Public schools, and state-sponsored charter schools (all elementary schools and any school with prekindergarten or kindergarten) All public schools that responded to the program invitation (pre-K‒12) Public schools (K‒12) Public schools (grades not specified)
Type of outlet tested Drinking Regularly used drinking fountains, cafeteria and food preparation areas, or reusable water bottle filling stations Drinking, cooking, medicinal, and other (e.g., ice maker, home economics, service connector, bathroom, and classroom sinks) Drinking and cooking Drinking, cooking, and faucets used by in school childcare Drinking, cooking, medicinal, and other (e.g., home economic and special education rooms) Drinking
Proportion or number of outlets tested per school or building 2 per building constructed before Jan 1, 1988, 1 per building constructed after 1–5 per school All outlets except those clearly marked as not used for drinking, cooking, or medicinal care 2 outlets per school At least 3 outlets per building 2 outlets per school All outlets used for drinking
Volume of water in sample collected 250 mL 1 L 250 mL 250 mL or 1 L 250 mL 250 mL Not specified
Stagnation periodb before sample collection At least 6 h Not specified 8–18 h At least 6 h At least 6 h 8–18 h Not specified

Note. ppb = parts per billion.

aAction level is the concentration of lead in ppb found in a drinking water sample that triggers a response.

bStagnation period is the recommended amount of time that water remain motionless, or stagnant, in the pipes before samples are drawn.

States With Lead Concentration Data

Of the 25 states with a school drinking water testing initiative in operation between January 1, 2016 and February 28, 2018, 18 states did not have eligible data (i.e., compiled electronic numeric data from outlet-level samples collected in more than 15 schools). Compared to these 18 states, the 7 states with eligible data had more city schools (37% compared to 26%), more suburban schools (41% compared with 35%), and fewer rural schools (13% compared with 26%), a higher proportion of students eligible for free or reduced-price meals (56% compared with 46%), lower proportions of White students (34% compared with 57%) and Black students (7% compared with 16%), but higher proportions of Asian students (9% compared with 5%) and Hispanic students (46% compared with 18%) (P < .001 for all comparisons). The estimates presented are a summary across all public schools in the state.

Schools With Lead Concentration Data

Characteristics of the schools in the 7 states that met our data inclusion criteria are reported in Table 3. Overall, the 5688 public schools from 690 local education agencies with testing data represent 35% of all eligible schools in those states based on criteria identified in state program testing documents. In all 7 states, more than 60% of the schools with testing data were primary schools. In every state, public schools participating in the state’s testing program were significantly different in several ways from schools eligible but not participating. In most states (n = 6), participating schools had higher student enrollment compared with eligible nonparticipating schools, and suburban schools were overrepresented in water quality data in 5 states. Participating schools in 3 states had lower percentages of students eligible for free or reduced-price meals compared with eligible nonparticipating schools, while participating schools in 2 states had higher percentages (Table 3).

Table

TABLE 3— Characteristics of Schools With Outlet-Level Drinking Water Lead Concentration Data in 6 US States Plus the District of Columbia: 2016–2018

TABLE 3— Characteristics of Schools With Outlet-Level Drinking Water Lead Concentration Data in 6 US States Plus the District of Columbia: 2016–2018

AZ CA MA NV RI UT DC
Total no. of schools with water test results 1274 2202 947 308 209 640 108
% of eligible schools 58 22 51 78 69 63 49
School level, no. (%)
 Primary 875 (69)a 1474 (67)a 610 (64)a 304 (99) 130 (62) 430 (67) 78 (72)
 Secondary 399 (31) 728 (33) 337 (36) 4 (1) 79 (38) 210 (33) 30 (28)
 χ2 Pb < .001 < .001 .003 .28 .69 .97 .052
Metro status, no. (%)
 City 570 (45) 1005 (46) 148 (16) 150 (49) 14 (7) 75 (12) 108 (100)
 Suburban 334 (26) 1003 (46) 705 (74) 115 (37) 161 (77) 368 (58) 0 (0)
 Town 182 (14) 91 (4) 19 (2) 20 (6) 0 (0) 86 (13) 0 (0)
 Rural 188 (15) 103 (5) 75 (8) 23 (7) 34 (16) 111 (17) 0 (0)
 χ2 Pb .003 < .001 < .001 < .001 < .001 < .001 .33
Mean no. of students pre-K‒12 (SD)c 707 (525)a 688 (523)a 536 (366)a 631 (226)a 469 (304) 668 (452)a 427 (249)a
Pd < .001 < .001 .004 .015 .38 < .001 < .001
Mean % of students eligible for free or reduced-price meals (SD)e 53 (34)f 56 (29)f Not reported 65 (26) 42 (26)a 43 (23)f 80 (36)a
Pc .003 <.001 < .001 < .001 < .001 .4
Mean % of students by race and ethnicity, (SD)g
 Non-Hispanic White 37 (27) 28 (24) 62 (28) 33 (23) 67 (28) 74 (21) 12 (21)
 Non-Hispanic Black 5 (5) 7 (10) 7 (11) 10 (10) 7 (8) 1 (2) 69 (32)
 Non-Hispanic Asian 2 (3) 13 (16) 6 (9) 4 (4) 3 (3) 2 (2) 2 (3)
 Hispanic 46 (28) 45 (27) 20 (23) 44 (23) 18 (21) 18 (16) 15 (21)
 Non-Hispanic other race/ethnicity 10 (20) 7 (5) 4 (3) 9 (5) 5 (3) 6 (10) 2 (2)
Ph < .001 < .001 < .001 < .001 < .001 .23 .011
Mean no. of first-draw water samplesi taken per school (SD) 9.8 (8.0) 4.6 (0.9) 37.8 (28.9) 2.1 (0.4) 3.5 (2.5) 2.3 (1.2) 56.6 (35.9)

aHigher percentage or mean in schools with water test results compared with eligible schools without water test results.

bP value based on a χ2 test for the comparison between schools with water test results and all eligible schools without water test results within a state with respect to the characteristic.

cNot all schools reported enrollment by grade in the following states: AZ (1262 schools reporting), CA (2201 schools reporting).

dP value based on a t test for the comparison between schools with water test results and all eligible schools without water test results within a state with respect to the characteristic.

eNot all schools reported enrollment numbers necessary to calculate percentage of students eligible for free or reduced-price meals in the following states: AZ (1091 schools reporting), CA (2198 schools reporting), DC (107 schools reporting), MA (0 schools reporting), NV (305 schools reporting), UT (633 schools reporting).

fLower percentage or mean in schools with water test results compared with eligible schools without water test results.

gNot all schools reported enrollment numbers necessary to calculate percentage of students by race/ethnicity in the following states: AZ (1256 schools reporting), CA (2199 schools reporting), UT (634 schools reporting).

hP value based on a global likelihood ratio χ2 test for the comparison between schools with water test results and all eligible schools without water test results within a state with respect to the characteristic.

iA sample was identified as a first draw if it was recorded as such in the state’s database (recorded as “pre-flush,” “first draw,” “stagnation draw,” “stagnant”), not recorded as an additional or follow-up sample (“flush,” “retest,” “resample,” “30 sec draw”), or if it was dated as the earliest recorded sample at a particular water outlet identified by a location identification code or location name.

The average number of first-draw samples taken at each school varied across states, as did the variability in the number of first-draw samples taken across schools within a state. States with small numbers of first draws per school and low variability in the number of first draws across schools included Nevada (mean = 2; SD < 1), Utah (mean = 2; SD = 1), Rhode Island (mean 4; SD = 3), and California (mean = 5; SD = 1). States with greater numbers of first draws per school and higher variability in number of first draws across schools included Arizona (mean = 10; SD = 8), Massachusetts (mean = 38; SD = 29), and DC (mean = 57; SD = 36).

School Drinking Water Lead Concentrations

All states collected, analyzed, and reported data on lead in measured concentrations as low as 5 ppb, and 3 states (Massachusetts, Nevada, and Utah) routinely collected, analyzed, and reported data on lead in measured concentrations as low as 0 to 1 ppb for all samples. Lead concentrations in drinking samples ranged from 0 to 42 000 ppb, with a median of 1.7 ppb and 90th percentile value of 9.7 ppb. Figure 1 depicts the percentage of schools with any first draws found to have lead concentration levels above 5, 10, and 15 ppb (Figure 1a).

Results varied by state. Nevada had the lowest percentage of schools with any first draws above 5 ppb (13%), and Massachusetts had the highest (81%). The medians for the 7 states were 37%, 20%, and 15%, by lead concentration level, respectively. For the mean percentage of first draws per school above 5, 10, and 15 ppb, the medians among states were 10%, 3%, and 2%, respectively (Figure 1b). California had the lowest mean percentage of first draws per school above 5 ppb (4%), while Massachusetts had the highest (25%). If 2 outlets per school were sampled at random (with replacement) from among all the outlets tested in a school (i.e., the lowest average number of outlets tested in the states), the estimated percentage of schools with any first draws having lead concentrations above each of the 3 specified levels (see Figure 1c) would have been lower than the observed percentage (Figure 1a) in every state. The median estimated percentages of schools with any first draws having lead concentrations above 5, 10, and 15 ppb using 2 random samples per school were 15%, 5%, and 2%, respectively. States with the biggest differences between the observed percentage of schools with any first draws above the specified concentration levels, and the estimated percentage using 2 random samples per school were DC, Massachusetts, and Arizona, while Nevada and Utah had the smallest differences.

In 2018, many US public school students attended schools in states without drinking water lead-testing programs. In available data from 7 states, the proportion of schools identifying any sample lead concentration exceeding 5 parts per billion varied (13%–81%), and 4 states exceeded 20%. Federal policy dictates that schools provide potable drinking water at no charge in the areas where meals are served during mealtimes if a school participates in the National School Lunch Program.26 Ninety-five percent of US public schools participate in this program.27 No federal policy requires that these schools conduct testing to ensure the quality of that drinking water, though rule revisions in December of 202118 expanded testing requirements for some schools served by community water systems. As a result, in 2018, nearly half of US public school students attended schools in states without drinking water lead-testing programs that tested all outlets.19 Within states, programs were often not structured to collect data consistent with standards for representativeness (e.g., voluntary participation of some school types limits generalizability, limited outlets tested within a school may not represent actual lead concentration at all water outlets). In addition, many did not maintain data in a centralized digital repository or make them publicly available, key features of data collection system stability that enable use for surveillance to support public health.

The voluntary nature of most state testing programs, together with lack of specificity in sampling protocols, compromises data representativeness, quality, and sensitivity, key characteristics of quality surveillance. In the 7 states with sufficient data available to characterize outlet-level numeric lead testing results, there were several notable findings. Overall, testing was conducted in just over one third of all public schools serving grades PK through 12, and school participation in testing in many states was not representative based on measures of student demographics or by school locales.

In addition, while there is no generally recognized standard for an allowable concentration of lead in school drinking water, the extent to which testing identified concentrations of lead in drinking water exceeding the 5-ppb standard used for bottled drinking water varied by state. In 4 of the 7 states, more than 20% of schools had at least 1 outlet where the lead concentration exceeded 5 ppb. Furthermore, widespread testing of all drinking water outlets in schools was not common. In the 2 states in which all outlets likely to be used for human consumption were tested, approximately 80% of schools had 1 or more outlets that exceeded 5 ppb. Notably, we found that the use of limited testing locations may not accurately represent the prevalence of lead in drinking water within all outlets at a school, potentially inhibiting appropriate responses. Other research confirms that the concentration of lead found in drinking water can vary considerably within the same building from outlet to outlet and even at the same outlet depending on use and infrastructure.28 Recent simulation research suggests the utility of a 5-sample, 90th-percentile approach to identify schools with increased lead in drinking water for further action.12

In addition to expanding programs to all states to ensure stable access to data of sufficient quality, other areas for improvement are suggested by data collected from existing state-directed programs. The sensitivity of the surveillance system includes its ability to detect a problem and to monitor changes over time. Only 3 states (Massachusetts, Nevada, and Utah) collected, analyzed, and reported data on lead in measured concentrations of 0 to 1 ppb, the maximum concentration of lead in school drinking water recommended by the American Academy of Pediatrics.29 Few states collect data on drinking water lead concentrations over time in a way that includes a central repository for data; clear documentation of the date, school, and physical location of the outlet from which water was obtained; and the type of sample tested. These details are critical for maintaining an accurate accounting of data collected and monitoring of water quality improvements or issues over time.

We identified limited instances in which funding was provided for all aspects of sample collecting, processing, or treatment and remediation. The lack of dedicated funding for testing, treatment, and remediation is a recognized barrier at the local level.10 Resources and supports for schools and local partners can promote acceptability of data collection for use in monitoring and surveillance. Recently, the US Environmental Protection Agency provided funding to each state and DC to support or establish programs for testing school drinking water for lead.30 This funding provides an opportunity to develop, require, and implement recommendations for key features of testing programs that align with established criteria for surveillance systems. These recent federal funding initiatives30 that support instituting testing programs may increase feasibility and acceptability.

In 2016, just over one third (37%) of school districts provided funding for or offered training for custodial or maintenance staff on school drinking water quality, and fewer than 1 in 5 (19%) provided funding or offered professional development for school staff on how to implement school drinking water quality initiatives.31,32 Thus, in addition to enhanced federal guidance for school water quality testing,21 strategies to improve knowledge of school staff and increased technical and financial support for the implementation of more widespread testing programs may increase the acceptability of testing programs among schools with varying constraints and prompt further action to reduce students’ exposure to lead in school drinking water.

Limitations

This study describes the features of 7 state programs testing school drinking water for lead; it does not quantify students’ exposure to lead in drinking water, nor potential health impacts. We do not know what proportion of outlets was sampled in schools in states without universal testing programs, nor details on how these outlets were identified and selected, which may have implications for program implementation.33 In the 2 states where program guidance specified sampling all outlets (Massachusetts and DC), the differences between the observed percentage of schools with any results above specified lead concentration levels and the estimated percentage obtained using 2 random samples per school were the largest. This would suggest that if only a portion of outlets in these schools had been tested, officials would not have identified all outlets with water quality issues. We cannot estimate the actual potential for misclassification in other states without data from all outlets.

This study presents the available data on the baseline prevalence of lead identified in drinking water in states that were early adopters of drinking water quality assessment programs. It does not represent the current status in schools, nor document the actions taken to remediate at outlets that exceeded state action levels. Recent changes to relevant federal rules,18 guidance,34 and new funding programs30 to support testing will contribute to additional availability of data to conduct ongoing surveillance.

Public Health Implications

Ensuring that children attend schools where the quality of drinking water is verified is a public health priority. Broader implementation of school drinking water lead testing programs will benefit from recently enacted financial and technical supports. State school drinking water lead testing programs could be strengthened to provide surveillance data capable of allowing officials to prevent potential adverse exposures through timely detection, treatment, and remediation to limit students’ potential exposure to lead in drinking water while at school.

ACKNOWLEDGMENTS

This research was supported by Healthy Eating Research, a National Program of The Robert Wood Johnson Foundation (grant 280-0799) and in part by a grant from the Centers for Disease Control and Prevention (U48DP006376). M. K. Poole was supported by a T32 training grant in nutrition (DK 007703–22) from the National Institutes of Health.

Note. The findings and conclusions in this report are those of the authors and do not represent the views of any funding organization or agency.

CONFLICTS OF INTEREST

The authors declared no conflict of interest.

HUMAN PARTICIPANT PROTECTION

The institutional review board of the Harvard T. H. Chan School of Public Health determined that this study was not human participant research.

References

1. Rosinger AY, Bethancourt H, Francis LA. Association of caloric intake from sugar-sweetened beverages with water intake among US children and young adults in the 2011‒2016 National Health and Nutrition Examination Survey. JAMA Pediatr. 2019;173(6):602604. https://doi.org/10.1001/jamapediatrics.2019.0693 Crossref, MedlineGoogle Scholar
2. Kenney EL, Long MW, Cradock AL, Gortmaker SL. Prevalence of inadequate hydration among US children and disparities by gender and race/ethnicity: National Health and Nutrition Examination Survey, 2009‒2012. Am J Public Health. 2015; 105(8):e113e118. https://doi.org/10.2105/AJPH.2015.302572 LinkGoogle Scholar
3. Bleich SN, Vercammen KA, Koma JW, Li ZH. Trends in beverage consumption among children and adults, 2003‒2014. Obesity (Silver Spring). 2018;26(2):432441. https://doi.org/10.1002/oby.22056 Crossref, MedlineGoogle Scholar
4. Muckelbauer R, Gortmaker SL, Libuda L, et al. Changes in water and sugar-containing beverage consumption and body weight outcomes in children. Br J Nutr. 2016;115(11):20572066. https://doi.org/10.1017/S0007114516001136 Crossref, MedlineGoogle Scholar
5. Popkin BM, D’Anci KE, Rosenberg IH. Water, hydration, and health. Nutr Rev. 2010;68(8): 439458. https://doi.org/10.1111/j.1753-4887.2010.00304.x Crossref, MedlineGoogle Scholar
6. Patel AI, Bogart LM, Klein DJ, et al. Middle school student attitudes about school drinking fountains and water intake. Acad Pediatr. 2014;14(5):471477. https://doi.org/10.1016/j.acap.2014.05.010 Crossref, MedlineGoogle Scholar
7. Environmental Protection Agency. Integrated science assessment for lead (third external review draft, Nov 2012). EPA/600/R-10/075C. 2012. Available at: https://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=242655. Accessed June 24, 2022. Google Scholar
8. Triantafyllidou S, Edwards M. Lead (Pb) in tap water and in blood: implications for lead exposure in the United States. Crit Rev Environ Sci Technol. 2012;42(13):12971352. https://doi.org/10.1080/10643389.2011.556556 CrossrefGoogle Scholar
9. Triantafyllidou S, Le T, Gallagher D, Edwards M. Reduced risk estimations after remediation of lead (Pb) in drinking water at two US school districts. Sci Total Environ. 2014;466-467:10111021. https://doi.org/10.1016/j.scitotenv.2013.07.111 Crossref, MedlineGoogle Scholar
10. US Government Accountability Office. K‒12 education: lead testing of school drinking water would benefit from improved federal guidance. GAO-18-382. 2018. Available at: https://www.gao.gov/assets/gao-18-382.pdf. Accessed February 14, 2022. Google Scholar
11. Zartarian V, Xue J, Tornero-Velez R, Brown J. Children’s lead exposure: a multimedia modeling analysis to guide public health decision-making. Environ Health Perspect. 2017;125(9):097009. https://doi.org/10.1289/EHP1605 Crossref, MedlineGoogle Scholar
12. Rome M, Estes-Smargiassi S, Masters SV, et al. Using the lead and copper rule revisions five-sample approach to identify schools with increased lead in drinking water risks. Environ Sci Technol Lett. 2022;9(1):8489. https://doi.org/10.1021/acs.estlett.1c00845 CrossrefGoogle Scholar
13. Subchapter XII - Safety of public water systems. Fed Regist. 1974:964‒1043. To be codified at 42 USC §300f et seq. Google Scholar
14. Maximum contaminant level goals and national primary drinking water regulations for lead and copper. Fed Regist.1991:26460‒26564. To be codified at 40 CFR §141 and 142. Google Scholar
15. Centers for Disease Control and Prevention. State-level school health policies and practices: a state-by-state summary from the School Health Policies and Programs Study 2006. Atlanta, GA: US Department of Health and Human Services; 2007. Available at: https://www.cdc.gov/HealthyYouth/SHPPS/2006/summaries/pdf/State_Level_Summaries_SHPPS2006.pdf. Accessed February 14, 2022. Google Scholar
16. Jones SE, Fisher CJ, Greene BZ, Hertz MF, Pritzl J. Healthy and safe school environment, part I: results from the School Health Policies and Programs Study 2006. J Sch Health. 2007;77(8):522543. https://doi.org/10.1111/j.1746-1561.2007.00233.x Crossref, MedlineGoogle Scholar
17. Subpart I - Control of lead and copper. Fed Regist. 1991;56(110):26548‒26562. To be codified at 40 CFR §141. Google Scholar
18. National primary drinking water regulations: Lead and Copper Rule revisions. Fed Regist. 2021;8(10):4198‒4312. To be codified at 40 CFR §141 and 142. Google Scholar
19. Cradock A, Hecht C, Poole MK, Vollmer L, Flax C, Barrett J. Early adopters: state approaches to testing school drinking water for lead in the United States. Boston, MA: Prevention Research Center on Nutrition and Physical Activity at the Harvard T. H. Chan School of Public Health; 2019. Available at: https://www.hsph.harvard.edu/prc/projects/school-research/early-adopters. Accessed February 14, 2022. Google Scholar
20. Environmental Protection Agency. 3Ts for reducing lead in drinking water in schools: revised technical guidance. 2006. Available at: https://nepis.epa.gov/Exe/ZyPDF.cgi/P100O90T.PDF?Dockey=P100O90T.PDF. Accessed February 14, 2022. Google Scholar
21. Environmental Protection Agency, Office of Ground Water and Drinking Water. 3Ts for reducing lead in drinking water in schools and child care facilities: a training, testing and taking action approach. EPA 815-B-18-007. 2018. Available at: https://www.epa.gov/sites/default/files/2018-09/documents/final_revised_3ts_manual_508.pdf. Accessed February 14, 2022. Google Scholar
22. German RR, Lee LM, Horan JM, Milstein RL, Pertowski CA, Waller MN. Updated guidelines for evaluating public health surveillance systems: recommendations from the Guidelines Working Group. MMWR Recomm Rep. 2001;50(RR-13): 135, quiz CE31–CE37. MedlineGoogle Scholar
23. Glander M. Documentation to the 2014–15 Common Core of Data (CCD) universe files. Washington, DC: National Center for Education Statistics, US Department of Education; 2016. NCES 2016-077. Available at: https://nces.ed.gov/ccd/pdf/2016077_Documentation_062916.pdf. Accessed February 14, 2022. Google Scholar
24. Subpart B-Requirements for specific standardized beverages. Fed Regist. 2016:676‒695. To be codified at 21 CFR §165.110. Google Scholar
25. Guidelines for Drinking-Water Quality: Fourth Edition Incorporating the First Addendum. Geneva, Switzerland: World Health Organization; 2017. Google Scholar
26. National School Lunch Program and School Breakfast Program: nutrition standards for all foods sold in school as required by the Healthy, Hunger-Free Kids Act of 2010. Fed Regist. 2016;81:5013150151. MedlineGoogle Scholar
27. Food Research and Action Center. FACTS: National School Lunch Program. November 2016. Available at: https://frac.org/wp-content/uploads/cnnslp.pdf. Accessed February 14, 2022. Google Scholar
28. Doré E, Deshommes E, Andrews RC, Nour S, Prévost M. Sampling in schools and large institutional buildings: implications for regulations, exposure and management of lead and copper. Water Res. 2018;140:110122. https://doi.org/10.1016/j.watres.2018.04.045 Crossref, MedlineGoogle Scholar
29. AAP Council on Environmental Health. Prevention of childhood lead toxicity [errata in Pediatrics. 2017;140(2):e20171490 and Pediatrics. 2020;145(6):e20201014]. Pediatrics. 2016;138(1): e20161493. https://doi.org/10.1542/peds.2016-1493 MedlineGoogle Scholar
30. Environmental Protection Agency, Office of Ground Water and Drinking Water. State lead testing in school and child care program drinking water grant implementation document: Water Infrastructure Improvements for the Nation (WIIN) Act. EPA 815-B-19-011. June 2019. Available at: https://www.epa.gov/sites/default/files/2019-06/documents/implementation_document_for_wiin_2107_testing_in_schools_june_19_2019_final.pdf. Accessed February 14, 2022. Google Scholar
31. US Department of Health and Human Services, Centers for Disease Control and Prevention. Results from the School Health Policies and Practices Study. 2016. Available at: https://www.cdc.gov/healthyyouth/data/shpps/pdf/shpps-results_2016.pdf. Accessed February 14, 2022. Google Scholar
32. Cradock AL, Everett Jones S, Merlo C. Examining differences in the implementation of school water-quality practices and water-access policies by school demographic characteristics. Prev Med Rep. 2019;14:100823. https://doi.org/10.1016/j.pmedr.2019.100823 Crossref, MedlineGoogle Scholar
33. Umunna IL, Blacker LS, Hecht CE, Edwards MA, Altman EA, Patel AI. Water safety in California public schools following implementation of school drinking water policies. Prev Chronic Dis. 2020;17:E166. https://doi.org/10.5888/pcd17.200366 Crossref, MedlineGoogle Scholar
34. US Environmental Protection Agency. 3Ts for reducing lead in drinking water: module 7: recordkeeping. 2018. Available at: https://www.epa.gov/ground-water-and-drinking-water/3ts-reducing-lead-drinking-water. Accessed February 14, 2022. Google Scholar

Related

No related items

TOOLS

SHARE

ARTICLE CITATION

Angie L. Cradock, ScD , Jessica L. Barrett, MPH , Mary Kathryn Poole, MPH , Chasmine N. Flax, MPH , Laura Vollmer, MPH, RD , and Christina Hecht, PhD Angie L. Cradock, Jessica L. Barrett, and Chasmine N. Flax are with the Prevention Research Center on Nutrition and Physical Activity, Department of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, Boston, MA. Mary Kathryn Poole is with the Department of Nutrition, Harvard T. H. Chan School of Public Health. Laura Vollmer is with the Cooperative Extension, University of California, Division of Agriculture and Natural Resources, Davis. Christina Hecht is with the Nutrition Policy Institute, University of California, Division of Agriculture and Natural Resources, Oakland. “Lead Concentrations in US School Drinking Water: Testing Programs, Prevalence, and Policy Opportunities, 2016‒2018”, American Journal of Public Health 112, no. S7 (September 1, 2022): pp. S679-S689.

https://doi.org/10.2105/AJPH.2022.306961

PMID: 36179297