Child Health and Social EcologyFelton EarlsProfessor of Social MedicineHarvard Medical School
Van Leer Panel on Psychosocial Well-being of AIDS Affected Children
Over the past four years, my colleagues and I have been conducting a randomized field trial in a medium size municipality in Tanzania. With young adolescents (age 10 to 14) positioned as change agents, the purpose of the study is to strengthen neighborhood collective efficacy as a means to reduce environmental and personal risks associated with HIV/AIDS. The study is conceived as a structural intervention and informed by ecological theory, the participatory rights of children, and theories of communicative action and social cognition. These ideas and principles support a perspective that views children as agents in promoting the well-being of all children and adults in their communities.
The theory and methods adopted for the Tanzanian study were developed in an epidemiological study of child exposure to violence in Chicago. Employing a multilevel, longitudinal design, the Chicago study demonstrated that both structural and organizational characteristics of neighborhoods exerted an independent, protective effect on violent behavior. The study also produced findings that went well beyond initial expectations by demonstrating neighborhood effects on a wide range of other health and behavior problems, including sexual risk behaviors related to HIV/AIDS. 
Components of the field trial in Tanzania include an extended baseline study of community structure, a child-driven intervention in which neighborhoods are the units of analysis, and pre- and post assessment of neighborhood dynamics and the health and behavior of participating children. The intervention is designed for children to acquire knowledge about the biology (microworld) and social ecology (macroworld) of HIV/AIDS and to disseminate this knowledge through skilled interaction within their neighborhoods. The methods used include geographic mapping, community surveys and drama. Findings of this study will be reported showing that the intervention was successfully mounted and that it produced positive changes in collective efficacy for children and in self-efficacy of children. 
This child-centered structural intervention operates at the interface of civil society and local government.  In the context of this African city, the study has created a new identity for children as “Young Citizens” (or “Raia Chipukizi”). Second, children have proven to be effective agents by imparting knowledge of the biological and social dimensions of HIV/AIDS. Third, organizational changes in everyday life are being produced that increase the “competence” of the community to control the incidence and consequences of HIV/AIDS. Plans to replicate and scale up the Young Citizens intervention will conclude this presentation.
Sampson R, Raudenbush S, Earls F. Neighborhood and violent crime: A multi-level study of collective efficacy.  Science 1997; 277:918-924.
Sampson, Robert J., Morenoff, J, Earls, F. 1999.  Beyond Social Capital: Neighborhood Mechanisms and Structural Sources of Collective Efficacy for Children.  American Sociological Review 1999;64:633-660.
Buka, SL, Brennan RT, Rich-Edwards JW, Raudenbush SW and Earls F. Neighborhood support and the birth weight of urban infants. American Journal of Epidemiology, 2003, 157:1-8.
Rich-Edwards JW, Buka SL, Brennan RT and Earls, F. Diverging associations of maternal age with low birth weight for black and white mothers.  International Journal of pidemiology, 2003:32:83-90. 
Obeidallah D, Brennan R, Brooks-Gunn J and Earls F.  Links between pubertal timing and neighborhood contexts: Implications for girls’ violent behavior.  J Amer Acad Child and Adoles Psychiatry. 2004 (12):1460-1468.
Xue Y, Leventhal T, Brooks-Gunn J, and Earls F. Neighborhood residence and mental health problems of 5- to 11-year olds. Archives of General Psychiatry, 2005, 62:554-563
Bingenheimer J, Brennan R and Earls, F.  Firearm violence exposure and serious violent behavior. Science, 2005, 30, 1323-1326.