SESSION 1A: System Structure & Performance-Partnerships
Room: Thoroughbred 1
Tuesday, April 8, 2014, 2-3:15 pm
Back to Conference Agenda
AUDIO
KC14 Session 1A on System Structure & Performance: Partnerships by NCC for PHSSR
MODERATOR
Joan Reede, M.D., M.S., M.P.H., M.B.A.
PRESENTERS
Lea Ayers LaFave, Ph.D., M.S.N., B.A., RN
Use of Network Analysis to Understand Structures and Collaboration among Regional Networks Providing Tobacco-Related Services
Co-Investigators: Jo Porter, M.P.H., & Stacey Gabriel, B.A.
Background & Research Objective: An emerging body of literature supports the application of systems science in addressing public health issues. Network analysis provides one approach to understanding the context within which public health interventions occur, suggesting that structural network characteristics, such as size and composition, connectivity among partners and positioning of network members can help understand network functionality. As part of a larger study examining how funding and allocation for tobacco-related services relate to connectivity among partner members of local public health systems, this descriptive study explores infrastructural differences among four local public health networks serving diverse geographical and demographic areas.
Data Sets and Sources: The Program to Analyze, Record, and Track Networks to Enhance Relationships (PARTNER), an online social network data collection and analysis tool, was used to collect data from fourteen regional public health networks providing chronic disease prevention services in April 2013. A 42-item survey addressing a broader set of public health issues was sent to 274 partners from regional, multiregional, state level and state governmental organizations, yielding a 64% response rate.
Design and Analysis: Data specific to organizations providing tobacco-related services within a purposive sub-sample of four distinct networks were abstracted from the larger data set providing specific regional tobacco service network descriptors. The four networks selected for this analysis were those that are part of a separate study focused on public health costs for tobacco-related services.
Principle Findings: Each regional network generated a unique set of network characteristics. Although similar in size (n= 7-10), tobacco networks revealed wide variation in other network metrics: density (25-46.7%), degree centrality (52.8-100%), trust (72.2-90.3%), and collaboration levels (cooperative 22.2-45.5%, coordinated 9.1-22.2%, integrated 33.3-55.6%).
Conclusions: These findings provide a lens with which to explore connections between network structure and function, and how those might relate to the delivery of public health services.
Implications: Decreasing resources for public health programming indicate an increasing need for collaboration and resource exchange among diverse public health. Identifying network characteristics and quantifying funding sources and resource exchange can help networks better understand their own infrastructures, and the functioning of local public health systems.
Sharla Smith, Ph.D., M.P.H., B.S.
Public Health System Partnerships: Examining the association of public health systems partnerships and infant mortality
Co-Investigator: Michael Preston, Ph.D., M.P.H, B.S.
Background: Reducing the United States (US) Infant Mortality Rate (IMR) has been and continues to be recognized as an important public health objective. As of 2012, the US IMR was about 6 per 1,000 live births, a historically low rate in the US, but a rate that is still higher than other developing countries.
Research Objectives: The purpose was to examine specific partnership types among local public health systems as well as to explore the association of public health system partnerships (PHSPs) and IMR.
Data Set and Sources: This study used survey response data of local health departments (LHDs) that responded to all three waves (1998, 2006, and 2012) of the National Longitudinal Study of Public Health Agencies (NLSPHA). The NLSPHA data was merged with data from the NACCHO Profile Study, Area Resource File, and National Vital Statistics. Study Design A longitudinal, retrospective cohort research design was used to examine the association between infant mortality rates and PHSPs types among a sample of the nation’s LHDs in 1998, 2006, and 2012.
Analysis: Multivariate regression panel analysis of questions from the NLSPHA (n = 947) was used to determine whether PHSPs were associated with IMRs. Principal Findings PHSPs density and centrality were significantly associated with an increase in infant mortality. A 1% increase in organizational density is associated with a 2.9% increase in infant mortality (p< .05) after controlling for all other factors in the model. In addition, a 1% increase in organizational centrality is associated with a .3 % increase in infant mortality after controlling for other factors in the model.
Conclusions: These findings suggest that public health systems should consider evaluating specific partners’ motivations and expectations of their role in addressing infant mortality in the communities they serve.
Implications: This study and its findings provide guidance to public health policy makers in their efforts to promote population health through public health partnerships. It highlights opportunities to better engage partners through a community based participatory approach. This approach will assist practitioners in developing partnerships in which community organizations play a leadership role in the partnerships.
Susan German, M.P.H.
New Jersey Local Tobacco Control Activity from the MPROVE Study
Co-Investigators: Pauline Thomas, M.D., Natalie Pawlenko, M.S.W., Kevin Sumner, M.P.H, & Margy Jahn, M.P.H
Background: Variation in the volume and reach of tobacco public health activity at the local level is known to exist. Less is known of the correlates and determinants of such variation.
Research Objective: In collaboration with the Public Health Practice-Based Research Network (PH PBRN) National Coordinating Center (NCC), we quantified and characterized tobacco prevention and control activities performed by New Jersey's local health departments (LHDs) as part of the Multi-Network Practice and Outcome Variation Examination (MPROVE) Study.
Data Sets and Sources: Measures of tobacco prevention and control activities were collected from local health departments via an online survey.
Study Design: The study sample, a subset of the multi-state MPROVE Study, launched in 2012, consisted of 69 (73%) of 95 LHDs in New Jersey. A REDCap survey with 18 questions including four regarding tobacco control activities was sent to all LHDs. Analyses included descriptive statistics and will assess relationships between tobacco control activities and demographic and LHD structural characteristics.
Analysis: Descriptive analyses were performed. Relationships between tobacco activities and demographic variables at the LHD jurisdictional level, as well as LHD structure will be assessed.
Principal Preliminary Findings: The most frequent activity pertaining to enforcement of the NJ Smoke-Free Air Act (NJSFAA) was the receipt of one or more complaints (59% of LHDs), followed by one or more inspections conducted (54% of LHDs). The two most common tobacco control activities performed by LHDs were development and dissemination of educational materials (80% of LHDs), and policy development (46% of LHDs); the least frequent activity was tobacco use surveillance (11% of LHDs). Several LHDs believed that they lacked the authority to enforce the NJSFAA.
Preliminary Conclusions: Just over half of the responding LHDs engaged in NJSFAA enforcement. This could indicate either compliance with the NJSFAA in many jurisdictions or a lack of education among the public and several LHDs regarding enforcement.
Implications of Preliminary Findings for Public Health Practice and Policy: There may be a need for increased education of the public and LHDs on NJSFAA enforcement procedures. The additional analytic analyses will be completed by the start of the Keeneland Conference.