Final Report: Health Effects of HAPs Among Inner Urban School Children

EPA Grant Number: R826789
Title: Health Effects of HAPs Among Inner Urban School Children
Investigators: Greaves, Ian , Adgate, John L. , Church, Timothy , Ramachandran, Gurumurthy , Sexton, Ken
Institution: University of Minnesota - Twin Cities
EPA Project Officer: Stacey Katz/Gail Robarge,
Project Period: October 1, 1998 through September 30, 2001 (Extended to September 30, 2003)
Project Amount: $633,044
RFA: Urban Air Toxics (1998) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Air

Objective:

The objectives of this research project were to: (1) determine whether inner urban, poor, minority children attending grades 2-5 in a new environmentally safe school had lower rates of respiratory illness or better learning outcomes than similar children attending a nearby older school; (2) determine whether biomarkers of exposure and health effects differed among children in the two schools; (3) model the relationships between biomarkers and health outcomes; (4) collect and analyze data on covariates and confounders; and (5) provide baseline data for a cohort who can be followed to see if childhood environmental exposures are associated with health problems in later life. Two null hypotheses were tested: the new school has no effect on health outcomes among its students; and biomarkers of exposure are unrelated to health outcomes.

Summary/Accomplishments (Outputs/Outcomes):

The School Health Initiative: Environment, Learning, and Disease (SHIELD) study evaluated young school children’s environmental exposures and respiratory health at a level of detail not previously attempted.

Over a 2-year period, repeated environmental samples were obtained for air quality in homes, schools, and the general environment (samples included allergens and volatile organic compounds [VOC]); blood and urine were obtained on four occasions for biomarkers of exposure to tobacco smoke, heavy metals, VOCs, and pesticides; annual questionnaires were administered for health and exposures; and lung spirometry was performed annually. All of these procedures required the cooperation of parents, teachers, school administrators, the Minneapolis Public School District, community groups, and, most importantly, the children. Because of severe budgetary limitations within the Minneapolis Public School District and loss of key personnel, learning outcome measures were not made available for analysis. For this reason, the study results focused on respiratory health outcomes in relation to the school attended; environmental exposures in the schools, homes and ambient environment; and socioeconomic factors and racial and ethnic characteristics.

A total of 153 out of 270 families (57%) from the two schools agreed to participate in the initial survey. We obtained data from 153 index children—those selected randomly from the school rolls based on grade, gender, school, and whether English was spoken in the home—plus an additional 48 siblings of the index children who wished to participate and also attended the same school. Almost all the children in this study received free school meals, an indicator of low family income, a fact confirmed subsequently by questionnaire responses. At the start of Year 2 of the study, 136 (89%) of the original 153 families had children in these schools, and 107 of those families (70% of the original cohort) participated in the follow-up survey.

Data capture rates for blood and urine samples were high, averaging over 90 percent for at least one sample and about 85 percent for two samples of blood and urine in each year of the study. Participation in lung spirometry was similarly very high, with 90 percent of children completing the test in Year 1 and 96 percent in Year 2. Compliance with peak flow measurements was much lower because these measurements required students to collect and record their own data twice a day, on 3 days in 1 week, for 3 weeks in the year. In Year 1, about 77 percent of students completed 1 week of sampling, but only 32 percent completed all 3 weeks of sampling. Because of the time and frequency needed for this test, class activities were interrupted, and compliance was affected by children’s absences because of field trips, illnesses, etc. The time needed for the peak flow measurements and the relatively low compliance rate in Year 1 prompted us not to repeat these measurements in Year 2; peak flow data were not explored further.

Rates of physician-diagnosed asthma varied by race and ethnicity, with the highest rates occurring among Asians (26%), followed by Whites/Others (21%), Hispanics (13%), Somalis (11%), and African Americans (7%). Atopy, as measured by total IgE levels of more than 100 IU/mL or any specific IgE level greater than 0.35 IU/mL, also varied by race and ethnicity, with Whites/Others having the highest rate (75%), followed by Hispanics (62%), Asians (56%), Somalis (39%), and African Americans (35%).

Overall, the rates of asthma and atopy were similar in each school, but rates of allergies to specific antigens differed. Elevated titers for IgE levels to cat and cockroach antigens were substantially greater among students at Whittier than at Lyndale (P < 0.001), whereas titers to dust mite p1 (P < 0.001) were higher for children at Lyndale. Further analysis showed that the differences between schools for allergen sensitivities were explained by differential sensitivities among certain racial groups who happened to be distributed preferentially to each school. Eighty percent of Hispanics, who had high rates of sensitivity to cockroach and cat dander allergens, attended Whittier, whereas the vast majority of Asians and White/Others, with high rates of sensitivity to dust mite allergens, attended Lyndale.

Exposures to allergens in homes were measured directly by sampling floor dust from the living rooms of each home during the late winter and spring months of 2000. For house dust allergens, the highest median exposure levels were seen in Asian/White/Other homes, and the lowest levels in African American homes. Median cockroach allergen levels were highest in the homes of Somalis and Hispanics and lowest in African American homes. Allergen measurements in the homes suggested exposures in that environment may account for Asians/Whites/Others having the highest rates of sensitization to house dust mites, and could explain why Hispanics had a high rate of sensitization to cockroach allergens and Asians/Whites/Others had low rates of sensitization to cockroaches. Sampling for the same allergens in classrooms, we found low concentrations of house dust and cockroach allergens and little difference between the two schools, suggesting that the school environments were not responsible for the different allergic sensitivities of their students.

Further differences were noted between students at the two schools with regard to lung function. Regressions of forced expiratory volume (FEV1) and forced vital capacity (FVC) were constructed using independent variables for age, height, and body mass index (BMI), as well as gender, various racial and ethnic groups (African American, Somali, Hispanic, Asian), and Whittier School. Statistically significant coefficients were obtained for height, each of the racial groups (relative to White/Other children), females (relative to males), and Whittier School (relative to Lyndale). Each of the racial or ethnic groups (African American, Somali, Hispanic, Asian) had significantly lower lung function values rela­tive to the White/Other category; girls had significantly lower lung function than boys; and children who attended Whittier School had larger lung volumes than those attending Lyndale. Adding a further variable to these models for physician-diagnosed asthma showed no overall effect of asthma on lung function.

Differences in lung function of children from different racial and ethnic groups have been reported previously for African American and Hispanic children, but data are unavailable for Somali or Asian children. The present findings are consistent with and amplify similar analyses using National Health and Nutrition Examination Surveys (NHANES) data for African American children’s lung function, expand our understanding of lung function in Hispanic children, and provide the first reports for Somali and Lao/Hmong/Cambodian children.

It was unclear why children attending Whittier had systematically higher lung function than students at Lyndale (on average, Whittier exceeded Lyndale for FEV1 and FVC by about 9 and 12 percent, or 130 and 190 mL, respectively). The school environments did not reveal any appreciable differences in indoor air quality or other risk factors that might explain these differences in lung function. Again, it appeared that school was a surrogate for factors not accounted for in the regression models. Because some racial and ethnic groups distributed preferentially to one or other of the schools, it is possible that the use of a school term may have resulted in confounding between race and school, perhaps leading to overspecification of the model. We continue to explore this odd finding.

Biomarkers of exposure to metabolites of compounds in tobacco smoke, organic solvents, heavy metals, organophosphate pesticides, chlorinated hydrocarbons, and polychlorinated biphenyls were measured in blood and/or urine of these children. These measurements, together with measures of contaminants in ambient air and indoor air levels in homes and schools, were obtained in a companion grant funded by the U.S. Environmental Protection Agency (School-Based Study of Complex Environmental Exposures and Related Health Effects: Part A Exposure, Grant No. R825813).

Urinary total cotinine levels showed substantial exposures of these children to environmental tobacco smoke (ETS), largely resulting from persons smoking in the homes. African American children had the highest exposure levels, with half of these children having at least 9.0 ng/mL of total cotinine in their urine. The median value among Asian/White/Other children was 2.2 ng/mL. Somali and Hispanic children had substantially fewer detectable samples and overall had lower urinary levels of cotinine than African Americans or Asian/White/Other children. No difference in the average urinary cotinine level was observed between children attending the two schools. Total urinary cotinine is often considered the most direct, and therefore the best, indicator of ETS exposures. The present findings also indicated that questionnaires or time-activity data can predict urinary cotinine levels well and can be used epidemiologically to describe children’s exposure to ETS.

Metabolites of the tobacco-specific lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) were used as an index of children’s exposure to carcinogens in ETS. The metabolites of NNK are 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and its glucu­ronide (NNAL-Gluc). Of 74 samples for which there was sufficient urine to measure the metabolites of NNK, either NNAL or NNAL-Gluc was detected in 62 (84%). These metabolites were present in 52 (96%) of the 54 samples with total urinary cotinine greater than or equal to 5 ng/mL and in 10 (50%) of 20 samples with total cotinine less than 5 ng/mL. Levels of NNAL, NNAL-Gluc, and total cotinine were not significantly different in samples collected from the same children at 3-month intervals.

Very limited data are available concerning carcinogen uptake by children exposed to ETS. The present findings therefore add substantially to our understanding of children’s exposures resulting from inhaling carcinogens in sidestream tobacco smoke. Levels of the combined metabo­lites (NNAL + NNAL-Gluc) in this study were comparable to those observed in previous studies of nonsmoking adult women living with a spouse who smoked. There was a 93-fold range of NNAL + NNAL-Gluc values in the exposed children. The results demonstrated widespread and considerable uptake of the tobacco-specific lung carcinogen NNK in this group of elementary schoolchildren, raising important questions about potential health risks. The data further indicated that objective biomarkers of carcinogen uptake are important in studies of childhood exposure to ETS and may be relevant to the development of tobacco-related cancer later in life.

Children’s exposures to VOCs were measured with personal sam­ples indoor at home and school and with fixed samples for outdoor air. In addition, biomarkers of VOC exposures were measured in blood samples obtained from children attending the two schools. Time-activity data were collected for the children who wore personal sampling devices: the majority of time was spent indoors in the home (65%) and at school (25%).

Personal samples obtained in the winter and spring months of 2000 demonstrated that more than 75 percent of children had detectable levels of exposure on each occasion to the following com­mon VOCs: benzene, carbon tetrachloride, chloroform, p-dichlorobenzene, ethylbenzene, δ-limo­nene, methylene chloride, α- and β-pinene, styrene, tetrachloroethylene, toluene, trichloroethyl­ene, m/p- and o-xylene. These exposures did not necessarily translate into detectable levels of VOCs in blood: detectable levels were found in the blood of 50 percent or more children for the following compounds: benzene, 1,4-dichlorobenzene, ethylbenzene, styrene, tetra­chloroethyl­ene, toluene, and m/p- and o-xylene. Median blood concentrations showed very few differences between children attending the two schools; all median VOC concentrations were less than 0.5 μg/L, indicating very low exposures overall among this population of inner urban children.

Organochlorine pesticide exposures were assessed in 2000 and 2001 by measuring blood levels of the parent compound or a metabolite. The most prevalent compound was p,p-DDE (present in 96-100% of samples), followed by β-HCCH (65-72%), p,p-DDT (51-61%), t-NONA (25-46%), and oxychlor (20-30%). Median blood levels of p,p-DDE were in the range of 1.0-1.4 μg/L. Median exposure levels for all other organochlorine pesticides in blood were less than 0.1 μg/L—an order of magnitude less than DDE. No difference was found in the median levels of these metabolites for students attending each school.

Children’s exposures to organophosphate pesticides were assessed by measuring urinary metabolites: diethyl-dithiophosphate, diethylphosphate, diethyl-thiophosphate, dimethyl-dithiophosphate, dimethylphosphate, and dimethyl-thiophosphate. More than 50 percent of children had detectable levels of diethylphosphate, diethyl-thiophosphate, dimethylphosphate, and dimethyl-thiophosphate, with 96-98 percent having detectable levels of both diethylphosphate and dimethyl phosphate in their urine. Median urinary levels of diethylphosphate and dimethylphosphate were 6.1 and 7.6 μg/L, respectively. No difference was found in the median levels of these metabolites for students attending each school.

Urinary levels of phenol pesticides showed a substantial number of compounds present in the urine of these children in 2001. More than 50 percent of children had detectable levels of 1- and 2-naphthol; 2,4-dichlorophenol; 2,4,5-trichlorophenol; 2,5-dichlorophenol; 3,5,6-trichloropyridinol; or o-phenylphenol. Median exposure levels were generally in the range of 1-9 μg/L for these various compounds. No difference was found in the median levels of these compounds for students attending each school.

Levels of polychlorinated biphenyls (PCBs) in blood showed very few compounds were detectable in these children. The most prevalent were PCB153 (68-85%), PCB118 (52-56% of samples), and PCB180 (31-37%). Median levels of blood concentrations were uniformly less than 0.05 μg/L. In general, PCB levels in students at each school were similar.

Urinary levels of phthalates were detected in high fractions of students at each school in 2001. More than 75 percent of children had detectable levels for the following monophthalate compounds: 2-ethyl-5-hydroxyhexyl phthalate (100% of samples), 2-ethyl-5-oxohexyl phthalate (99-100%), 2-ethylhexyl phthalate (94-96%), benzyle phthalate (99-100%), methyl phthalate (58-76%), and n-butyl phthalate (100%). Among these frequently detected monophthalate compounds, median urinary concentrations were generally in the range of 20-30 μg/L, with the exception being 2-ethylhexyl phthalate where median concentrations were about 4 μg/L. These values are lower than data reported recently based for the NHANES survey of 1999-2000. No difference was found in the median levels of these compounds for students attending each school.

Blood lead and mercury levels were measured in these children in 2000 and 2001. Lead was found in 97-100 percent of samples and mercury was found in 40-72 percent. The median levels for lead were 2.0-2.6 μg/dL, and the median levels for mercury were 0-0.3 μg/L. These are relatively low values, and again no difference was found in the median levels of these metals for students attending each school. The concentrations of elements in urine showed virtually all children had detectable levels of antimony, barium, cesium, cobalt, lead, molybdenum, thallium, and tungsten. Smaller fractions had detectable levels of chromium (68%), platinum (49%), and uranium (91%). The highest concentrations were observed for molybdenum, with a median level of 75 μg/L. All other median concentrations were less than 6 μg/L. No difference was found in the median levels of these elements for students attending each school.

Exposure-response models are being analyzed that relate respiratory health outcomes to the various biomarkers of exposure measured in this study. To guide these analyses, we have constructed a causal model that considers genetic, socioeconomic, educational, and environmental exposures as risks factors and modifiers for the outcomes of interest—asthma, atopy, and lung function.

Preliminary assessments suggest that metabolites of organophosphate pesticides and of nicotine (cotinine) are associated with abnormalities in lung function and increased levels of circulating IgE. Additional compounds under consideration as possibly affecting respiratory health include mercury, styrene, polonium, and perhaps one or more of the PCBs.

Conclusions

Few differences in school exposures to hazardous agents were found between students attending a newly constructed, environmentally conscious school and those attending a traditional school built in the 1970s. Exposures in each school were generally very low and did not pose an environmental health risk to students. This is good news for public school districts and school administrators who are faced with difficult choices as to whether they should renovate old schools or replace them with new ones. At least in some cases, older schools can be maintained at a level that provides a healthy and safe environment equivalent to a newly constructed school.

Among the populations of poor, inner urban schoolchildren studied here, the greatest risk of hazardous exposures to a wide variety of toxic agents (e.g., allergens, ETS) occurred in the homes, where children were found to spend, on average, about two-thirds of their time. This is not a new finding but serves to emphasize again that children living in low-income, inner urban areas experience most of their environmental health risks in their homes.

The population of children studied here was of mixed racial and ethnic composition, providing an opportunity to examine exposures and responses to environmental risks in relation to race and ethnicity. A number of racial and ethnic differences were observed in relation to respiratory health, asthma, and allergies. We also documented racial and ethnic differences in lung function among these children. The changes seen among African American children add to previous studies in confirming that African Americans have smaller lungs than whites. The findings in African American children and the other racial and ethnic groups will help clinicians and epidemiologists to better understand the respiratory health of minority and immigrant children and help inform future studies of environmental exposures and respiratory health in minority populations.

Exposures to a host of environmental toxics were assessed in this study, and few differences were found between the two schools. Preliminary data relating these exposures to lung function, atopy, and asthma suggest that environmental tobacco smoke and perhaps organophosphate pesticides may be associated with adverse effects on the respiratory health of these children.

The baseline data reported here provide an opportunity to study the cohort as it ages and to examine respiratory and other health outcomes in relation to the exposures that have been documented in earlier life.


Journal Articles on this Report : 11 Displayed | Download in RIS Format

Other project views: All 11 publications 11 publications in selected types All 11 journal articles
Type Citation Project Document Sources
Journal Article Adgate JL, Church TR, Ryan AD, Ramachandran G, Fredrickson AL, Stock TH, Morandi MT, Sexton K. Outdoor, indoor, and personal exposure to VOCs in children. Environmental Health Perspectives 2004;112(14):1386-1392. R826789 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Journal Article Greaves IA, Sexton K, Blumenthal MN, Church TR, Adgate JL, Ramachandran G, Fredrickson AL, Ryan AD, Geisser MS. Asthma, atopy, and lung function among racially diverse, poor inner-urban Minneapolis schoolchildren. Environmental Research 2007;103(2):257-266. R826789 (Final)
  • Abstract from PubMed
  • Full-text: Science Direct Full Text
    Exit
  • Other: Science Direct PDF
    Exit
  • Journal Article Hecht SS, Ye M, Carmella SG, Fredrickson A, Adgate JL, Greaves IA, Church TR, Ryan AD, Mongin SJ, Sexton K. Metabolites of a tobacco-specific lung carcinogen in the urine of elementary school-aged children. Cancer Epidemiology, Biomarkers & Prevention 2001;10(11):1109-1116. R826789 (2000)
    R826789 (2001)
    R826789 (2002)
    R826789 (Final)
    R825813 (2001)
  • Abstract from PubMed
  • Full-text: AACR Journals-Full Text HTML
    Exit
  • Other: AACR Journals-PDF
    Exit
  • Journal Article Needham LL, Sexton K. Assessing children's exposure to hazardous environmental chemicals: an overview of selected research challenges and complexities. Journal of Exposure Analysis and Environmental Epidemiology 2000;10(6 Pt. 2):611-629. R826789 (2002)
    R826789 (Final)
    R825813 (2001)
  • Abstract from PubMed
  • Abstract: Nature - Abstract
    Exit
  • Other: Nature PDF
    Exit
  • Journal Article Ramachandran G, Adgate JL, Banerjee S, Church TR, Jones D, Fredrickson AL, Sexton K. Indoor air quality in two urban elementary schools—measurements of airborne fungi, carpet allergens, CO2, temperature, and relative humidity. Journal of Occupational and Environmental Hygiene 2005;2(11):553-566. R826789 (Final)
    R825813 (2001)
  • Abstract from PubMed
  • Abstract: Taylor & Francis - Abstract
    Exit
  • Journal Article Sexton K, Greaves IA, Church TR, Adgate JL, Ramachandran G, Tweedie RL, Fredrickson A, Geisser M, Sikorski M, Fischer G, Jones D, Ellringer P. A school-based strategy to assess children's environmental exposures and related health effects in economically disadvantaged urban neighborhoods. Journal of Exposure Analysis and Environmental Epidemiology 2000;10(6 Pt. 2):682-694. R826789 (2000)
    R826789 (2001)
    R826789 (2002)
    R826789 (Final)
    R825813 (2000)
    R825813 (2001)
  • Abstract from PubMed
  • Other: Nature PDF
    Exit
  • Journal Article Sexton K, Adgate JL, Church TR, Greaves IA, Ramachandran G, Fredrickson AL, Geisser MS, Ryan AD. Recruitment, retention, and compliance results from a probability study of children's environmental health in economically disadvantaged neighborhoods. Environmental Health Perspectives 2003;111(5):731-736. R826789 (2002)
    R826789 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Journal Article Sexton K, Waller LA, McMaster RB, Maldonado G, Adgate JL. The importance of spatial effects for environmental health policy and research. Human and Ecological Risk Assessment: An International Journal 2004;8(1):109-125. R826789 (2002)
    R826789 (Final)
    R825241 (Final)
  • Abstract: Informaworld Abstract
    Exit
  • Journal Article Sexton K, Adgate JL, Church TR, Hecht SS, Ramachandran G, Greaves IA, Fredrickson AL, Ryan AD, Carmella SG, Geisser MS. Children's exposure to environmental tobacco smoke: using diverse exposure metrics to document ethnic/racial differences. Environmental Health Perspectives 2004;112(3):392-397. R826789 (2002)
    R826789 (Final)
    R832734 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Full-text: Environmental Health Perspectives full text
  • Journal Article Sexton K, Adgate JL, Church TR, Ashley DL, Needham LL, Ramachandran G, Fredrickson AL, Ryan AD. Children's exposure to volatile organic compounds as determined by longitudinal measurements in blood. Environmental Health Perspectives 2005;113(3):342-349. R826789 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Journal Article Sexton K, Adgate JL, Fredrickson AL, Ryan AD, Needham LL, Ashley DL. Using biologic markers in blood to assess exposure to multiple environmental chemicals for inner-city children 3-6 years of age. Environmental Health Perspectives 2006;114(3):453-459. R826789 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Supplemental Keywords:

    school children, inner urban, low-income, indoor air pollution, airborne particles, heavy metals, volatile organic compounds, tobacco smoke, pesticides, allergies, asthma, learning, questionnaires, blood tests, urine tests, lung function, minority groups, African-American, Somali, Hispanic, Asian, Cambodian, Laotian,, RFA, Health, Scientific Discipline, Air, Toxics, Geographic Area, air toxics, Environmental Chemistry, Health Risk Assessment, Epidemiology, State, VOCs, Risk Assessments, Susceptibility/Sensitive Population/Genetic Susceptibility, Biochemistry, Children's Health, genetic susceptability, indoor air, Ecology and Ecosystems, asthma, urban air, pesticide exposure, monitoring, ambient air quality, atmospheric, risk assessment, sensitive populations, urban air toxics, building related illness, emission inventory, urban monitoring sites, Minnesota, MN, air pollutants, biological sensitivities, infants, inner urban school children, lung, buildings, health risks, urban school children, airway disease, measuring childhood exposure, respiratory problems, ambient air, HAPS, hazardous air pollutants, pesticides, susceptible populations, exposure, Human Health Risk Assessment, air pollution, children, emissions, pulmonary toxicity, urban air pollutants, ethnic groups, assessment of exposure, childhood respiratory disease, children's vulnerablity, inhalation, human exposure, allergic, toxicity, pulmonary, sick building syndrome, urine and blood samples, environmental toxicant, harmful environmental agents, urban air pollution, inhaled, schools, biological markers, indoor air quality, human health, sensitive population, allergen, allergies, disease, Minneapolis-St.Paul Metropolitan area, respiratory, VOC sensitivity, air quality, autoimmunity

    Progress and Final Reports:

    Original Abstract
    2000 Progress Report
    2001 Progress Report
    2002 Progress Report