Date on Master's Thesis/Doctoral Dissertation

8-2015

Document Type

Doctoral Dissertation

Degree Name

Ph. D.

Department

Health Management and Systems Sciences

Degree Program

Public Health Sciences with a specialization in Health Management, PhD

Committee Chair

Thornewill, Judah

Committee Co-Chair (if applicable)

Esterhay, Robert

Committee Member

Roelfs, David

Committee Member

Wilson, Richard

Subject

Public health--Management; Public health--Citizen participation--Kentucky--Louisville

Abstract

Background: Participation in public health networks is important to achieve the goals of public health. This dissertation addresses a common problem in public health management—of how to measure support and predict participation by individuals and organizations in network level collaboratives—through testing an adapted theory producing a newly developed instrument. Purpose: Community health collaboratives have grown in popularity and in funding opportunities in the past couple decades, but they notoriously fail to achieve measureable results. By adapting theory and relevant instruments to a specific public health network context, factors that enable and inhibit participation are identified. Method: A prospective, theory-driven, sequential mixed-methods study was done involving both leaders and members of the Mayor’s Healthy Hometown Movement of Louisville, Kentucky. Two interviews with key network leaders were conducted, along with a focus group of five other central network leaders. Qualitative data was used to adapt highly validated quantitative instruments to this specific network context, and the survey was deployed to more than five thousand individuals on a network-specific listserve. Responses were entered into SPSS, where OLS regression analysis identifies variables correlated with intent to participate in the network for an identified participation opportunity. A second survey to assess actual participation was deployed 3 months later, and binary logistic regression was used to assess correlation with intent to participate. Findings: Several factors were found to be statistically significant using the adapted quantitative instrument. The quantitative data collection obtained responses from 244 total respondents. Regression models were analyzed for two unique participation opportunity types—both of which involved the use of the Healthy Louisville Community Dashboard. For the first—visiting the community dashboard for information—the research identified 6 statistically significant independent variables and a maximum model fit (rr) of .404. For the second—contributing content to the community dashboard—the research identified 5 statistically significant independent variables and a maximum model fit (rr) of .273. A 70.1% response rate to the follow-up survey highlighted a positive and statistically significant correlation between intent to use the dashboard and actual dashboard use, for both visiting the dashboard for information and contributing content to the dashboard. Conclusion: Use of adapted theory and instruments, adapted through the qualitative approach used in this research, produced a way to better understand and predict stakeholder intent to participate in this public health network, related to the identified participation opportunity. These results provide ways for network leadership to better understand network participants and their participation behaviors, as well as provide a framework for future applications in public health and potentially other sectors, in both theory and practice.

Included in

Public Health Commons

Share

COinS