Grantee Research Project Results
Final Report: New Methods for Analysis of Cumulative Risk in Urban Populations
EPA Grant Number: R834582Title: New Methods for Analysis of Cumulative Risk in Urban Populations
Investigators: Scammell, Madeleine Kangsen , Ozonoff, David M.
Institution: Boston University
EPA Project Officer: Hahn, Intaek
Project Period: July 1, 2010 through June 30, 2014 (Extended to June 30, 2015)
Project Amount: $749,226
RFA: Understanding the Role of Nonchemical Stressors and Developing Analytic Methods for Cumulative Risk Assessments (2009) RFA Text | Recipients Lists
Research Category: Human Health
Objective:
This study is the product of many years of research and relationship building among a multidisciplinary team of scientists in the faculty in the Department of Environmental Health, Boston University School of Public Health, and community residents in the City of Chelsea, MA, adjacent to Boston. Chelsea is Massachusetts' second most-densely populated municipality (35,080 residents in 1.8 square miles, 2000 Census). Its median household income is $30,161 compared to $50,502 statewide. Twenty-three percent of people living in Chelsea have incomes below the poverty level, compared to 9.3% for the state. It has a high proportion of non-white residents, with 61% identifying as Hispanic, Black or African American, American Indian, Alaskan Native, Asian and Native Hawaiian or other Pacific Islander, and 39% as White or Caucasian. The Massachusetts Executive Office of Energy and Environmental Affairs has designated every census tract in Chelsea an environmental justice population, the only such municipality in the state. According to a study on environmental justice in Massachusetts conducted by Faber and Krieg at Northeastern University, Chelsea ranks third in Massachusetts for potential hazardous exposures.
Chelsea Creek runs between Chelsea and neighboring East Boston. The land along the Creek on both sides is a Designated Port Area, meaning that development along the Creek must be reserved for marine industrial uses. As a result, the Creek hosts industry, parking lots, a multi-ton salt pile, and fuel storage for industrial and commercial enterprises. Along the Creek one can find Gulf Oil, Global Oil, and Coastal Oil storage facilities containing millions of gallons of petroleum, an animal hide processing plant and numerous state designated hazardous waste sites. The oil storage tanks along the Creek contain all of the jet fuel used at Logan International Airport and 70-80% of the region's heating fuels.
In addition to the numerous sources of air, water, and land pollution located along the Chelsea Creek, residents are exposed to air emissions from traffic on state-designated truck routes. This includes trucks traveling to and from the New England produce market in Chelsea (which had 37,000 truck deliveries in 2010) and vehicles traversing the Tobin Bridge and Route 16, both of which bisect the City of Chelsea. According to modeled concentrations of diesel particulates, Chelsea exceeds the EPA reference concentration by 20%. Massachusetts Department of Public Health data show Chelsea has the highest asthma hospitalization rates in the state and among the highest hospitalization rates for stroke, heart disease, heart attack, major cardiovascular disease, and coronary heart disease.
In 2003, the Chelsea Creek Action Group and U.S. EPA Region 1 published the results of the Chelsea Creek Community Based Comparative Risk Assessment. The purpose of this 2-year effort was for residents in the City of Chelsea and neighboring East Boston to identify environmental issues of greatest concern and make recommendations for improvement. Highest on the list were ambient air quality, water quality, lack of open/green space, asthma and respiratory ailments, noise, and traffic. Key findings were beliefs that current federal, state, and local regulations did not adequately protect the health of urban residents and did not allow for a way to "understand the cumulative impacts of multi-media contamination for local residents." Many residents and community leaders in the City of Chelsea are aware of the shortcomings of current risk assessment methods. Moreover, they expressed concern about the effects of non-chemical stressors related to violence and crime, lack of recreational opportunities for youth, problems of mental health and depression, and issues related to immigration and the economy. These additional burdens were considered to be "part of the environmental mix" and affected the risks of disease. What is not clear is how to put these disparate factors together in a more systematic and analytical way.
Summary/Accomplishments (Outputs/Outcomes):
This study was conducted in partnership with the Chelsea Collaborative, Chelsea, MA. Associate Executive Director, Ms. Roseann Bongiovanni, was our community-based Principal Investigator. Over the course of the study, we trained 15 bilingual English-Spanish speaking staff members and volunteers to recruit and interview residents of Chelsea. Many of the staff or volunteers have photos and bios on the www.chelseastar.org website used as a recruitment and outreach tool.
Specific Aim 1: Use established qualitative and quantitative research techniques to collect, code and characterize information about chemical exposures of concern, social and economic concerns, behavioral risk factors for disease, self-reported health outcomes, and perceptions of environment and quality of life from residents abutting an urban designated port area.
The goal was to interview up to 500 randomly recruited residents who would reflect the demographics of the city in five census tracts (see ChelseaSTAR Study Exit for a map). In each census tract we identified particular census blocks and established a target number of recruits for each block. An example is census tract 1604 (CT1604), which is large and mostly industrial, where we selected only the residential or mixed land use areas near the Tobin Bridge for recruitment. Using 2000 census data, we estimated the population size of our recruitment area to arrive at a target recruitment number for each neighborhood that reflects their relative population size (e.g., for CT1601 our goal was 206 participants distributed between 33 census blocks, so our goal was to recruit six people in each block).
We invited participants and informed residents of the study at community meetings and via flyers, informational ads in the local papers (English and Spanish language publications), and via a bilingual conversation between the Chelsea STAR project coordinator, Ms. Rosa Maria Olortegui, and Dr. Scammell, which was filmed at the local cable television station and repeatedly played on cable TV. In all these media, interested residents were invited to call Ms. Olortegui to determine eligibility and set up an interview.
Our recruitment strategy relied primarily on door-knocking in each census block, taking on one census tract at a time. If no one answered the door, we left a letter asking people to call Ms. Olortegui at the Chelsea Collaborative. The letter was signed by Dr. Scammell and Ms. Roseann Bongiovanni, both Chelsea residents. We kept a written record of each home we invited to participate so that if after a week we had not obtained a response, study staff returned to the same location at a different time for a second attempt. We returned to the home twice (a total of three visits), before abandoning that home and moving to another home in the same census block. Recruits were limited to one person per household.
Interviews were conducted by teams of two: the primary interviewer, and a note-taker. Upon entering the home of the participant, the lead interviewer consented the participant before beginning the interview. Written documentation of consent was waived by the Boston University Medical Center IRB given the sensitive nature of our questions. However, the Boston University Medical Campus IRB required extensive training of all study staff prior to entering the field to recruit participants or conduct interviews. In addition to the NIH training in human subjects protection, we developed the following training sessions and held them for study staff on a twice yearly basis:
- Recruitment and Interviews: This training covered the process of how to recruit and consent participants, conduct interviews, and take notes, as well as the protocol for protecting confidentiality. This covered the essentials of the study protocol, with the confidentiality of participants as the primary concern. These trainings were led by the PI and project coordinator, and used team work, role playing (scheduling and conducting an interview), and team building. The emphasis was on the ethical and responsible conduct of research.
- Preventing and Managing Difficult Situations: This training focused on identifying, preventing, and managing potentially difficult situations that might arise during the recruitment and scheduling of interviews and was led by detectives Sammy Mojica and Rosie Medina at the Chelsea Police Department. They discussed the types of crime experienced in Chelsea, geographic areas of concern, cautions to take while walking in various neighborhoods, and an orientation to the police services and responses in Chelsea in the event a staff member needed to call on police. Tips were provided on personal and staff safety for door-knocking and entering homes. A protocol for communication among the staff was established, which was reviewed in detail during the training.
- The third training was developed and led by Joanne Timmons, MPH, domestic violence (DV) program coordinator at Boston Medical Center and a social worker for the Chelsea Schools, Ms. Anita Mercado. Ms. Timmons discussed strategies for preventing and managing difficult emotional and potentially dangerous situations for team members and interviewees, before and after the consent process. Ms. Mercado trained staff on recognizing signs of abuse of an elder, child, or disabled person. Our staff were not considered mandated reporters by law, but we trained them to respond appropriately if they observed such abuse. Ms. Timmons provided a basic introduction to domestic violence and abusive relationships, including tactics of the perpetrator and indicators of high risk DV situations. She trained staff on ways to manage a situation if a person gets upset or makes threats, or if a person becomes very sad and emotional. The training also discussed how to be professional and kind while not taking on the role of therapist or friend. This training included role playing.
Study protocol approval by the IRB was an unusually lengthy process, in part due to the fact that we proposed community-based and participatory research in a community widely viewed has having high-crime and risk. Once the IRB approved our protocol, Dr. Scammell co-wrote an article for a Boston Medical Campus publication on how to navigate such IRB challenges (see publications).
In Winter/Spring 2011, the project coordinator, Ms. Olortegui, and project staff presented the study objectives and received input on the types of questions to be asked during the interviews with the following meetings or committees sponsored by or organized with the Chelsea Collaborative:
Chelsea Collaborative leadership retreat (January), Chelsea Collaborative staff meeting (February), Chelsea Green Space committee (January and February), Chelsea Latino Immigrant Committee (February), Chelsea United in Defense of Education (February and March), City-Wide Tenants Association (February), Environmental Chelsea Organizers (March), Vecinos Unidos / Neighbors United (February and March), and the Shanbaro Community Association of Somali Bantu in Chelsea (March). We also met with members of the City of Chelsea Board of Health (March), the City of Chelsea Tree Committee (April), and staff of the Massachusetts General Hospital Chelsea Healthcare Clinic (May) to discuss our research and data sharing.
Investigators from the Department of Environmental Health at BUSPH offered knowledge and expertise in qualitative and quantitative interview methods and in large national surveys (BRFSS and NHANES). We sought to develop an interview guide that would build from existing guides and that also would include questions specific to the Chelsea community to ensure the guide was appropriate to the population we were sampling.
Dr. Scammell and Ms. Olortegui drafted the interview guide with input from all of the above committees, all co-investigators, as well as investigators at the BUSPH Department of Community Health Sciences with expertise in exposure to violence and neighborhood effects on health. The final interview guide includes components of the Collective Efficacy Survey, the Multi-group Ethnic Identity Measure (MEIM), National Health and Nutrition Examination Survey (NHANES), and the Behavioral Risk Factor Surveillance Survey (BRFSS). Sections of the interview guide focus on: Social Environment, Ethnic Identity, Physical Environment, Individual Health, Life Stressors, Community Involvement. The guide includes both open and closed ended questions.
Questions that came specifically from our meetings with community committees focus on: community lighting of streets and walkways, odors, trash pickup and recycling, relaxation and recreation, use of parks, religious affiliations and community participation, parents’ participation in schools, language barriers, cultural adaptation, and other immigrant issues. The questionnaire then was translated into Spanish and piloted with Spanish-speaking members of the Chelsea Collaborative. We recently developed an annotated version of the Guide for use in future studies and submitted it for peer review publication in the journal Community Engaged Scholarship for Health (CES4H) in April 2016.
In June 2013, we completed the interview data collection process, recruiting a total of 354 residents. Also in 2013, we established a collaboration with the Puerto Rican Health Study, an NIH-funded cohort study conducted out of Tufts University. Dr. Scammell met with their team three times to discuss plans for setting a route for mobile monitoring of air pollutants in the City of Chelsea and for the location of a stationary monitor. The route includes all census tracts that are part of the Chelsea STAR study. Data now are available and they are being analyzed.
Judy Ou, the PI’s doctoral student at BUSPH, wrote her thesis using Chelsea STAR data. It was titled, Neighborhood Determinants of Obesity in Chelsea, MA. She graduated in Spring 2015 and now is a Research Associate in Pediatric Hematology/Oncology at the University of Utah. The first of three papers was published in late 2015 and two additional papers, also using qualitative and quantitative data from the study, are in preparation for submission.
Specific Aim 2: Use already developed research software implementing Galois lattices (also called Formal Concept Analysis) to examine the hierarchical and structural relationships of quantitative and qualitative data elements.
For a week in June or July in each year of our funding period, Drs. Ozonoff and Scammell worked with Dr. Alex Pogel (New Mexico State University) to use Formal Concept Analysis software to analyze the interview results. Numeric, non-text responses were used to identify areas for improvements of the software. The result was auto inclusion of descriptive labels on lattice nodes, implementing additional methods for drawing lattices (including nested line diagram capability), and labeling of concepts with the cardinality of their extents (i.e., automatically including the number of elements within each set on the node labels). Each year we identified additional features to be added to the software, including aspects related to the scaling of non-binary responses and calculation and display of nested lattices (for example, viewing lattices within all males and females, or other potentially confounding variables that we would normally stratify). The epidemiological meaning of the mathematical representation was an additional focus. This work is continuing.
Specific Aim 3: Use the lattice as a technique for cumulative risk assessment by examining the relationships revealed by computation.
Early explorations of the data for the purpose of testing software functionality revealed surprising patterns in questions related to types of violence and food security. Using responses to questions normally used to measure exposure to violence in the community by aggregating score, we observed that knowledge of neighborhood gang violence was disjoint from knowledge of robbery. Almost all individuals with experience of robbery thought they had enough food and the kinds of food they wanted but none of these had concerns about gang violence, or rape/sexual assault. In contrast, all people who reported that sexual assault or rape, and/or gang fights occurred in their neighborhood also reported not having enough food or the types of food they wanted. The most commonly reported events were robbery or mugging, and this was most often reported by people for whom food security was not a concern, nor were gang fights, rapes, or sexual assault. Thus, the lattice diagram view revealed that lumping these groups together by aggregating their scores obscures an important difference.
We examined an additional data set with the same or similar variables kindly provided us by Jane Clougherty (an EPA STAR grantee) of the New York City Health indicators. We compared results obtained by exploratory factor analysis and lattice visualization to determine if similar clustering of variables occurred. These and similar comparisons are ongoing.
Specific Aim 4: Share results of analysis with community members and public health officials and make attempts to share the software with public health practitioners and epidemiologists to use as an additional and practical tool for data analysts.
We drafted an 80-page report that includes all descriptive analyses of our data. That report is in final stages of editing and will be posted on our website and available for download. We also are creating a brochure to be shared at community meetings that summarizes key findings regarding asthma, food security, pest problems, and housing concerns.
One concern raised among interviewees was the safety of parks and drug use in parks. Dr. Scammell represented the Chelsea Collaborative on a Community Leadership Team (CLT) in the City of Chelsea convened to address issues of substance abuse. The subcommittee on which she served focused on "hot spots," including parks and "tot lots" identified by our study participants. The goal of the CLT was to reduce drug abuse and to create a safer and healthier community in the City of Chelsea. We shared our data on parks and park use and apparent barriers to park use due to concerns regarding violence identified in our study.
In the process of conducting the study, our investigators met many community leaders, researchers and EPA officials interested in using the interview guide for their own purposes. Recently, Boston University School of Public Health and Harvard T.H. Chan School of Public Health jointly launched a Center for Research on Environmental and Social Stressors in Housing Across the Life Course (CRESSH). CRESSH investigators are developing interview guides for measuring social and environmental stressors related to housing. The research is focused on housing and air quality, and will recruit 100 residents in the city. The Chelsea STAR project maps, questionnaire, and recruitment strategy have been adapted by the new study, and a stationary air monitor will be located at the Chelsea Police Station. As a partial result of our research, we have engaged in discussions with Massachusetts Port authority regarding noise and air pollution (attended by Dr. Scammell, Ms. Bongiovanni, and our state representative Sal Di Domenico,) and the need for greenspace in the city. Dr. Scammell also has engaged faculty in the BUSPH Department of Epidemiology in discussions regarding the use of our data as pilot data for additional funding focused on immigration challenges and handgun violence. Our data are unique in the information they provide on immigration status and associated social and environmental stressors within individuals.
The Chelsea Collaborative has used data from the interviews about street lighting to advocate for improved public lighting in several neighborhoods. Their efforts culminated in the contracting of 1,627 new LED streetlights in neighborhoods across the city in 2013.
Data obtained in the interviews also was shared with representatives of the Chelsea Police Department and the City of Chelsea Board of Health at their leadership retreat. Furthermore, the data were used by the Board of Health to launch a new home inspection ordinance to address structural problems that could result in the exacerbation of pest problems. Analyses have been conducted from the interview results regarding the use of parks and exposure to community violence, physical activities, and depressive symptoms for residents of Chelsea.
Journal Articles on this Report : 3 Displayed | Download in RIS Format
| Other project views: | All 17 publications | 4 publications in selected types | All 3 journal articles |
|---|
| Type | Citation | ||
|---|---|---|---|
|
|
Ou JY, Levy JI, Peters JL, Bongiovanni R, Garcia-Soto J, Medina R, Scammell MK. A walk in the park: the influence of urban parks and community violence on physical activity in Chelsea, MA. International Journal of Environmental Research and Public Health 2016;13(1):E97 (12 pp.). |
R834582 (Final) |
Exit Exit Exit |
|
|
Payne-Sturges DC, Korfmacher KS, Cory-Slechta DA, Jimenez M, Symanski E, Carr Shmool JL, Dotson-Newman O, Cloughtery JE, French R, Levy JI, Laumbach R, Rodgers K, Bongiovanni R, Scammell MK. Engaging communities in research on cumulative risk and social stress-environment interactions: lessons learned from EPA's STAR Program. Environmental Justice 2015;8(6):203-212. |
R834582 (Final) R834576 (Final) R834577 (Final) R834578 (Final) R834579 (2014) R834579 (Final) R834580 (Final) R834581 (Final) |
Exit Exit Exit |
|
|
Scammell MK, Montague P, Raffensperger C. Tools for addressing cumulative impacts on human health and the environment. Environmental Justice 2014;7(4):102-109. |
R834582 (Final) |
Exit Exit |
Supplemental Keywords:
ambient air, water, health effects, human health, sensitive populations, age, race, diet, ethnic groups, toxics, particulates, metals, solvents, public policy, observation, preferences, social science, epidemiology, mathematics, modeling, monitoring, analytical, Northeast, EPA Region 1, transportation, petroleum, environmental justice;Relevant Websites:
Boston University School of Public Health news article on partnership between Chelsea Collaborative and BU researchers: Health Disparities: narrowing the gap through community partnerships (PDF) (18pp, 2.12 MB) Exit
Boston University School of Public Health, Faculty websites:
- Madeleine Scammell | Boston University School of Public Health Exit
- David Ozonoff | Boston University School of Public Health Exit
The Cumulative Impacts Project Exit, Sponsored by the Science & Environmental Health Network and the Collaborative on Health and the Environment
Several EPA STAR grantees working on cumulative risk analysis in a webinar, which can be downloaded here: Cumulative Impacts on Health: New Community-Based Research Projects, Part 2 Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.