Wilma Brakefield-Caldwell

Community action against asthma: examining the partnership process of a community-based participatory research project.

AbstractBACKGROUND: Community Action Against Asthma (CAAA) is a community-based participatory research (CBPR) project that assesses the effects of outdoor and indoor air quality on exacerbation of asthma in children, and tests household- and neighborhood-level interventions to reduce exposure to environmental asthma triggers. Representatives of community-based organizations, academia, an integrated health system, and the local health department work in partnership on CAAA’s Steering Committee (SC) to design and implement the project. OBJECTIVE: To conduct a process evaluation of the CAAA community-academic partnership. DESIGN: In-depth interviews containing open-ended questions were conducted with SC members. Analysis included established methods for qualitative data, including focused coding and constant comparison methods. SETTING: Community setting in Detroit, Michigan. PARTICIPANTS: Twenty-three members of the CAAA SC. MEASUREMENTS: Common themes identified by SC members relating to the partnership’s ability to achieve project goals and the successes and challenges facing the partnership itself. MAIN RESULTS: Identified partnership accomplishments included: successful implementation of a complex project, identification of children with previously undiagnosed asthma, and diverse participation and community influence in SC decisions. Challenges included ensuring all partners’ influence in decision-making, the need to adjust to “a different way of doing things” in CBPR, constraints and costs of doing CBPR felt by all partners, ongoing need for communication and maintaining trust, and balancing the needs of science and the community through intervention. CONCLUSIONS: CBPR can enhance and facilitate basic research, but care must be given to trust issues, governance issues, organizational culture, and costs of participation for all organizations involved.

Effect of geocoding errors on traffic-related air pollutant exposure and concentration estimates

Exposure to traffic-related air pollutants is highest very near roads, and thus exposure estimates are sensitive to positional errors. This study evaluates positional and PM2.5 concentration errors that result from the use of automated geocoding methods and from linearized approximations of roads in link-based emission inventories. Two automated geocoders (Bing Map and ArcGIS) along with handheld GPS instruments were used to geocode 160 home locations of children enrolled in an air pollution study investigating effects of traffic-related pollutants in Detroit, Michigan. The average and maximum positional errors using the automated geocoders were 35 and 196 m, respectively. Comparing road edge and road centerline, differences in house-to-highway distances averaged 23 m and reached 82 m. These differences were attributable to road curvature, road width and the presence of ramps, factors that should be considered in proximity measures used either directly as an exposure metric or as inputs to dispersion or other models. Effects of positional errors for the 160 homes on PM2.5 concentrations resulting from traffic-related emissions were predicted using a detailed road network and the RLINE dispersion model. Concentration errors averaged only 9%, but maximum errors reached 54% for annual averages and 87% for maximum 24-h averages. Whereas most geocoding errors appear modest in magnitude, 5% to 20% of residences are expected to have positional errors exceeding 100 m. Such errors can substantially alter exposure estimates near roads because of the dramatic spatial gradients of traffic-related pollutant concentrations. To ensure the accuracy of exposure estimates for traffic-related air pollutants, especially near roads, confirmation of geocoordinates is recommended.