Objective: The nation's 2862 local health departments (LHDs) are the primary means for assuring public health services for all populations. The objective of this study is to assess the effect of organizational network analysis on management decisions in LHDs and to demonstrate the technique's ability to detect organizational adaptation over time.
Design and Setting: We conducted a longitudinal network analysis in a full-service LHD with 113 employees serving about 187 000 persons. Network survey data were collected from employees at 3 times: months 0, 8, and 34. At time 1 the initial analysis was presented to LHD managers as an intervention with information on evidence-based management strategies to address the findings. At times 2 and 3 interviews documented managers' decision making and events in the task environment.
Results: Response rates for the 3 network analyses were 90%, 97%, and 83%. Postintervention (time 2) results showed beneficial changes in network measures of communication and integration. Screening and case identification increased for chlamydia and for gonorrhea. Outbreak mitigation was accelerated by cross-divisional teaming. Network measurements at time 3 showed LHD adaptation to H1N1 and budget constraints with increased centralization. Task redundancy increased dramatically after National Incident Management System training.
Conclusions: Organizational network analysis supports LHD management with empirical evidence that can be translated into strategic decisions about communication, allocation of resources, and addressing knowledge gaps. Specific population health outcomes were traced directly to management decisions based on network evidence. The technique can help managers improve how LHDs function as organizations and contribute to our understanding of public health systems.
The study aimed to assess the effect of organizational network analysis on management decisions in local health departments and to demonstrate the technique&#x0027;s ability to detect organizational adaptation over time.
Department of Biomedical Informatics, Columbia University, New York (Drs. Keeling and Merrill); and Champaign-Urbana Public Health District, Champaign, Illinois (Ms Pryde).
Correspondence: Jacqueline A. Merrill, PhD, MPH, RN, Columbia University School of Nursing, 630 West 168th St, Georgian 226, New York, NY 10032 (firstname.lastname@example.org).
This study was supported by the Robert Wood Johnson Foundation grant ID 59946. Keeling was supported by National Library of Medicine grant T15-LM007079.
The authors thank the employees of the Champaign-Urbana Public Health District for participating in this research.
The authors declare no conflicts of interest.