Toward A Better Understanding Of Community Mobilization For Maternal Health In Rural Zambia

Community mobilization is a process oriented solution for complex societal problems, including interventions addressing maternal health. Community mobilization (CM) is defined as a capacity-building process through which stakeholders engage in participatory and specific activities to improve and sustain their condition. Despite frequent use, measurement of complex outcomes from the process or a description of the domains involved is lacking. Little is known about how individuals are impacted as a result of the CM process or how community members perceive the function of CM. This mixed methods study evaluated change in individual and community level domains of CM, using longitudinal data from a three year intervention study employing CM prior to introducing a maternity waiting home intervention in rural Zambia. A CM survey was collected from three groups (women, health workers, community members; n=1,202) with associated focus groups (n=76) at two timepoints. Data were collected from ten intervention sites and ten comparison sites. Factor analysis refined factors used to assess change over time using a multivariable model. Focus group data were analyzed using content analysis with the CM domains as a priori themes and an unconstrained categorization matrix to guide our interpretation and allow for latent themes. Data were merged using a convergent mixed methods approach. Factor analysis maintained the domains of governance, collective efficacy, self-efficacy, and power in relationships. Focus group data indicated social cohesion as key to the CM intervention, with health workers interfacing between the health system, relatives, and pregnant women. The CM domains of self-efficacy, power in relationships, and governance each showed significant change over time in the multivariable models (p< 0.001). Interaction effects showed increases in the self-efficacy factor isolated within intervention communities (0.34, p< 0.001). Women had significantly lower factor scores for self-efficacy, governance, and collective efficacy compared with health workers or community members. Those with less education had lower self-efficacy (p< 0.001), collective efficacy (p< 0.01), and governance (p< 0.01) scores. Content analysis revealed barriers and facilitators to healthcare for pregnant women. Barriers were structural (e.g., poor road infrastructure) and cultural (e.g., traditional gender roles). Facilitators included the CM domains of collective action, governance, social cohesion, and self-efficacy. Merged data indicated domains of CM had varying group dependent impact, influenced by cultural and structural realities. Our analyses showed CM had a significant impact on various individual and community based CM domains in rural Zambia influenced by structural and cultural context. Future work should employ mixed methods to plan for implementation of CM interventions and contextualize their results. Evaluations of CM benefit from comprehensively investigating domains across various groups, linking impact to health outcomes.PhDNursingUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/163158/1/dcbeck_1.pd