Mary E. Northridge, PhD, MPH, Joanne Yankura, MPH, Patrick L. Kinney, ScD,. Regina M. ... ume and the high density of diesel-powered vehicles ... structed directly across 133rd Street from an ... The school is not near a bus depot and thus.
Commentary
Diesel Exhaust Exposure Among Adolescents in Harlem: A Community-Driven Study
..... ....
L
A
;; b
Mary E. Northridge, PhD, MPH, Joanne Yankura, MPH, Patrick L. Kinney, ScD, Regina M. Santella, PhD, Peggy Shepard, BA, Ynolde Riojas, MS, Maneesha Aggarwal, PhD, Paul Strickland, PhD, and the Earth Crew Diesel exhaust contributes to air pollution in general and to ambient particulate pollution in particular." 2 Increased cardiopulmonary morbidity and mortality have been associated with elevated levels of particulate matter less than 10 mm in diameter (PM,O).3? Exacerbation of respiratory symptoms and deterioration in lung function is known to occur with increased PM0o concentrations, particularly in individuals with preexisting chronic conditions such as chronic obstructive pulmonary disease and asthma.1-16 The latter is a special concern among youth in poor urban communities such as Harlem (located in Northem Manhattan, New York City) as a result of the documented rise in asthma morbidity and mortality over the past 15 years. 17-19 Moreover, several studies have suggested that PMIO generated during diesel fuel combustion may play a role in the etiology of asthma, although a causal relationship has not been firmly established.2-25
Institute of Environmental Health Sciences (NIEHS) Center for Environmental Health in Northern Manhattan (both part of the Joseph L. Mailman School of Public Health at Columbia University) to help them determine the possible impact of diesel exhaust on the health of the community. In particular, they were concerned about the respiratory health of young people, a potentially more exposed and vulnerable population. With initial funds from a pilot grant that was competitively awarded from the NIEHS Center for Environmental Health in Northern Manhattan, a partnership was formed between WE ACT, researchers at Columbia University, and health care providers at Harlem Hospital Center and Columbia Presbyterian Medical Center. A long-term goal of the partnership is to better understand the relationship between the environment and the health of the residents of Northem Manhattan and to help promote needed interventions. The first step, reported here, was to collect
Six of the 7 bus depots in Manhattan are located in Northern Manhattan, largely in disadvantaged African American and Latino communities. In addition, trucks are prohibited from using the highways that run along the east and west sides of Manhattan; instead, major truck routes cut through the center of Harlem and other poor Northern Manhattan neighborhoods. Residents are exposed to diesel exhaust as a result of heavy traffic volume and the high density of diesel-powered vehicles that travel these major streets. Harlem residents expressed concern when the Manhattanville Bus Depot was constructed directly across 133rd Street from an intermediate school in West Harlem. Members of West Harlem Environmental Action (WE ACT), a community-based environmental advocacy group, approached researchers at the Harlem Center for Health Promotion and Disease Prevention and the National
Mary E. Northridge and Joanne Yankura are with the Harlem Center for Health Promotion and Disease Prevention, New York, NY. Mary E. Northridge, Patrick Kinney, Regina M. Santella, Ynolde Riojas, and Maneesha Aggarwal are with the National Institute of Environmental Health Sciences (NIEHS) Center for Environmental Health in Northern Manhattan, New York, NY. Peggy Shepard and the Earth Crew are with West Harlem Environmental Action, New York, NY. Paul Strickland is with the Johns Hopkins School of Hygiene and Public Health, Baltimore, Md. The members of the Earth Crew are listed under Acknowledgments. Requests for reprints should be sent to Mary E. Northridge, PhD, MPH, NIEHS Center for Environmental Health in Northern Manhattan, 60 Haven Ave, Level B- 1, New York, NY 1003 2 (e-mail: menI 1@ columbia.edu). This paper was accepted March 27, 1999. Editor's Note. Michael Greenberg served as the responsible editor for this paper and approved the final version. July 1999, Vol. 89, No. 7
Commentary
data on the levels of diesel exhaust exposure and lung function among Harlem youth. The study is also important as a method for responding to community-initiated health concerns and involving community members in every phase ofthe research, from formulation of the study hypothesis to dissemination of the findings.
Methods Study Design In the spring of 1997, 26 seventh-grade students from Thurgood Marshall Academy in Harlem participated in a study to assess diesel exhaust exposure and lung function. The school is not near a bus depot and thus provides a "background" assessment of the chosen exposure and outcome among Harlem adolescents. An in-person interview was developed on the basis of standard items from previously validated questionnaires.26'27 The surveys were administered by trained high school students from WE ACT's Earth Crew Youth Leadership Program, a communitybased outreach initiative for youth leadership development through environmental education and community service. Among the items asked about were age, gender, race, ethnicity, personal smoking history, exposure to maternal smoking, and history of asthma. Spot urine samples were collected and assayed for l-hydroxypyrene, a metabolite of pyrene (a polycyclic aromatic hydrocarbon), as a marker of diesel exhaust exposure28 and for cotinine, a metabolite of nicotine, as a marker of tobacco smoke exposure. Lung function was assessed with computerassisted spirometry.
Data Collection No refusals were recorded among the students' parents/guardians or among the students themselves. Twenty-four students were in attendance on the day of the study (all provided questionnaire data and all but 3 provided a urine sample). Concurrent spirometry was scheduled but was performed at a later date owing to equipment malfunctioning. Because of individual absences on either test day, 3 students provided questionnaire data but no spirometry measurements and 2 students provided spirometry measurements but no questionnaire data. Complete data were obtained on 18 students. Computer-assisted spirometry was used to measure lung volumes and to calculate several parameters, including predicted and measured forced vital capacity, forced expira-
July 1999, Vol. 89, No. 7
tory volume in the first second of exhalation (FEV1), peak expiratory flow achieved during exhalation, flow rate achieved during the 25th to 75th quartiles of the expiratory phase 75), and FEV, expressed as a percent(FEF25 age of forced vital capacity (FEVI/FVC). Finally, an FEF2, -7 as a percentage of the predicted value was calculated for each student.29 1-Hydroxypyrene was assessed with monoclonal antibody 8E1 1 for immunoaffinity purification followed by synchronous fluorescence spectroscopy. The limit of detection for the assay used is 0.05 pmol/mL. Creatinine assays (Sigma Chemicals, St. Louis, Mo) were performed to correct for the spot nature of the specimens. Cotinine levels were measured with an enzyme-linked immunosorbent assay (ELISA) to account for any possible contribution of tobacco-related polycyclic aromatic hydrocarbons to the 1-hydroxypyrene levels obtained (STC Technologies, Bethlehem, Pa).
StatisticalAnalysis Statistical analyses of questionnaire items, urine assay levels, and spirometric parameters were performed with SAS statistical software, version 6.1 1.30 Descriptive frequencies of the categorical data and distributions of the continuous data were first examined. To demarcate "abnormal" from "normal" spirometry results, FEV1 and FEF2%_75 values expressed as percentage of the predicted value were divided into 2 categories on the basis of clinical criteria and usual practice: 80% or less and greater than 80%. A third ("borderline") category was created for FEF25-75 values greater than 80% but less than 90% of the predicted value for final analysis and presentation of results. Since the distribution of 1-hydroxypyrene levels was highly skewed, 4 categories were created, each containing approximately equal numbers of subjects. Spirometry results, specifically FEVI/ FVC and FEF2%_75, were then examined by asthma history, smoking status, and urinary 1-hydroxypyrene levels. Stratified analyses were also conducted to see whether any of these parameters varied by sex or selfreported smoking status.
Results Descriptive Statistics The 24 students who provided questionnaire data were demographically quite similar (Table 1). All participants were in the seventh grade (mean age = 12.8 years, age range= 12 to 14 years) and all students con-
sidered themselves to be African American. One student also identified herself as being of Hispanic/Spanish origin. Slightly more than half of the sample (58%) were male. Nine of the 24 questionnaire respondents reported having mothers who currently smoked cigarettes (38%), consistent with other prevalence estimates of smoking among Harlem adult women.27 Only 2 students, neither of whom reported a history of asthma, said that they currently smoked cigarettes. Nine students (38%) responded yes to the question "Have you ever had asthma?" All of the measured FEVy/FVC levels were greater than 80%, that is, within the normal range. Nonetheless, 3 students had FEF2%-75 values of 800/o or less ofthe predicted value and 4 students had FEF25 75 values greater than 800/o but less than 90% of the predicted value, suggestive ofpossible lung function impairment. The mean concentration of urinary 1-hydroxypyrene was 0.43 pmol/mL, although the distribution was not normal. One quarter of the urine samples had 1-hydroxypyrene levels of 0.05 pmol/mL (the limit of detection) or below. Another quarter of the samples had assayed levels of 0.06 to 0.33 pmol/mL; the next quarter had values that ranged from 0.34 to 0.73 pmol/mL, and the final quarter had concentrations ranging from 0.74 to 1.40 pmol/mL. None of the urine samples assayed had detectable levels of cotinine by the enzymelinked immunosorbent assay (ELISA) method (.50 ng/mL), meaning that there was no significant contribution from tobacco smoke to the measured 1 -hydroxypyrene levels in this study. Although the ELISA method works well with heavy smokers, it is not sensitive enough to detect low or secondhand tobacco smoke exposure. As the purpose here was to rule out major contributions from tobacco smoking to the measured 1 -hydroxypyrene levels, the ELISA method was deemed acceptable. There was no apparent relationship between FEF25 75 values and urinary 1-hydroxypyrene levels (Table 2). The 13 students with normal FEF25-7, results had urinary 1-hydroxypyrene levels that varied considerably. In fact, urinary 1-hydroxypyrene concentrations for this group were almost equally divided among the 4 1-hydroxypyrene quartiles. Of the 3 students with FEF2%75 values of 80% of the predicted value or below, 2 had urinary 1 -hydroxypyrene concentrations below the limit of detection and the third had a measured level of 0.34 pmol/mL. Among the 4 individuals with "borderline" FEF25_75 values (that is, in the >800/o-75 Expressed as Percentage of Predicted Value 80-
