The Structure of Policy Networks for Injury and Violence Prevention in 15 US Cities

Abstract

Objectives:

Changes in policy can reduce violence and injury; however, little is known about how partnerships among organizations influence policy development, adoption, and implementation. To understand partnerships among organizations working on injury and violence prevention (IVP) policy, we examined IVP policy networks in 15 large US cities.

Methods:

In summer 2014, we recruited 15 local health departments (LHDs) to participate in the study. They identified an average of 28.9 local partners (SD = 10.2) working on IVP policy. In late 2014, we sent survey questionnaires to 434 organizations, including the 15 LHDs and their local partners, about their partnerships and the importance of each organization to local IVP policy efforts; 319 participated. We used network methods to examine the composition and structure of the policy networks.

Results:

Each IVP policy network included the LHD and an average of 21.3 (SD = 6.9) local partners. On average, nonprofit organizations constituted 50.7% of networks, followed by government agencies (26.3%), schools and universities (11.8%), coalitions (11.2%), voluntary organizations (9.6%), hospitals (8.5%), foundations (2.2%), and for-profit organizations (0.7%). Government agencies were perceived as important by the highest proportion of partners. Perceived importance was significantly associated with forming partnerships in most networks; odds ratios ranged from 1.07 (95% CI, 1.02-1.13) to 2.35 (95% CI, 1.68-3.28). Organization type was significantly associated with partnership formation in most networks after controlling for an organization’s importance to the network.

Conclusions:

Several strategies could strengthen local IVP policy networks, including (1) developing connections with partners from sectors that are not well integrated into the networks and (2) encouraging indirect or less formal connections with important but missing partners and partner types.

Keywords

injury and violence prevention organizational network policy

Each year, millions of US residents experience violence, injury resulting from violence, or unintentional injury.^1,2 In 2010, 176 000 people in the United States died as a result of injury or violence; together, injury and violence are the leading cause of death among those aged 1-30 in the United States.² Violence and violence-related intentional injury occur as a result of child maltreatment, youth violence, intimate partner violence, suicide, and sexual violence. The most common causes of unintentional injury are motor-vehicle incidents, poisoning, falls, suffocation, drowning, and fire.²

Public health policy—or the laws, regulations, and practices followed by government and other organizations to protect and improve public health—has been demonstrated to be effective in reducing many types of violence and injury. For example, implementing and enforcing seatbelt use policies has reduced the number of injuries and deaths in traffic accidents³; required building inspections have reduced the number of lead poisoning incidents⁴; and integrating fall prevention into state policy has led to a reduction in the number of fall-related hospital admissions.⁵ Requiring background checks for gun and ammunition purchases has reduced the number of homicides, suicides, and unintentional deaths by guns,^6

–11 and a policy prohibiting people convicted of domestic violence from owning a gun has significantly reduced the number of homicides by guns.¹² In addition, improving the built environment and offering access to green space is associated with a reduced number of incidents of intimate partner violence and gun violence,^1,13
–15 and increasing alcohol taxes and reducing the number of alcohol outlets has reduced the number of traffic crash fatalities and multiple forms of violence.^16

–19 Some policies, such as those for transferring juveniles into the adult justice system,²⁰ actually lead to increased violence. Other policy measures may reduce violence rates indirectly by addressing risk factors for violent behavior, with these risk factors including living in an area where economic opportunities are limited and social and health services are lacking.¹

Networks of organizations are more effective than individual organizations at addressing complex public health problems, providing services, and using resources efficiently.^21,22 Evidence on the role of interorganizational networks in the development, adoption, and implementation of public health policy is growing.²³ For example, organizations collaborating on community-level alcohol policy influenced policy adoption in one community,²⁴ and a comparison of tobacco policy networks found that tobacco policy adoption had been successful in a community with a denser network (ie, many relationships among the network members) and unsuccessful in another community in the same state with a less dense network.²⁵ Finally, one study found that a network of community partners working on physical activity policy increased the amount of time that students had in physical education classes²⁶; another study found that partnerships among organizations from different sectors improved policy engagement in a network of organizations working to reduce cancer disparities.²⁷

The composition and structure of public health organizational networks may also influence performance. Specifically, dense networks are associated with successful local public health practice,^28
–30 and local health departments (LHDs) in networks that include diverse partners are better than LHDs without diverse partners at providing essential services to their constituents.^{28,29,31
–33}

Numerous calls for multisectoral partnerships to improve injury and violence prevention (IVP) programs and policy have been made.^2,34
–36 Current evidence for the influence of multisectoral partnerships on IVP is scarce. Core sectors involved in preventing injury and violence include police and criminal justice, education, hospitals and medical service providers, social services and child protection, alcohol regulators, local public health, transportation, justice, and nongovernmental organizations.^34,35 One study found that information-sharing partnerships among hospital emergency departments, police departments, and local government can reduce violence-related intentional injury by providing police with information that helps them focus prevention efforts in locations where violence is occurring.^37,38 However, despite calls for collaboration, efforts to develop partnerships among stakeholders have been hampered by issues of leadership, turf, and omission of key stakeholders.^39,40

As a first step in understanding IVP policy partnerships, we used a network approach to examine IVP policy networks in 15 large US cities. We set out to answer the following questions: (1) What types of organizations are involved in IVP policy work at the local level? (2) Which organization types are perceived as important to IVP policy work at the local level? and (3) How are organizations partnering to work on IVP policy?

Methods

Participants

At the time of the study in August 2014, 18 LHDs were part of the Big Cities Health Coalition (http://www.bigcitieshealth.org), a group of health departments serving 54 million people across large urban areas in the United States. We invited all 18 LHDs to participate, and 15 agreed to do so. Then we asked the 15 LHDs to identify their local partners working on 5 types of public health policy: tobacco control, obesity and chronic disease, IVP, infant mortality, and core local funding. To identify the local partners in each policy area, we asked the senior official responsible for policy (eg, chief of policy) at each of the 15 LHDs to identify up to 10 local partners for each policy area. We also asked senior officials to identify 2 partners who were considered key leaders in each policy area who could identify additional local partners. We emailed these key leaders and asked if they could identify up to 10 additional local policy partners in each policy area not already listed by the LHD policy official. From these sources, we compiled a full list of policy partners for each of the 15 jurisdictions. Descriptive statistics quantifying network size (ie, number of network members including the LHD and its policy partners), density, degree centralization (ie, the extent to which a network includes a single member or a small number of members with more ties than the others), and composition (ie, how many types of partners are in each network) of the networks in all 5 areas are compared elsewhere⁴¹; this study focused on the composition and structure of the IVP networks.

We sent an online survey to each organization identified as a member of a policy network in each jurisdiction. Drawing on prior interorganizational network research,^25,42,43 we first asked participants to identify the organizations with which they had had contact in the past 12 months. From this list, we asked each organization to identify partners with whom they had worked in the past 12 months on IVP policy by making the following request: “Please indicate the organizations you have worked with in the past 12 months and the policy areas on which you worked with them.” The question was followed by a list of organizations and, for each organization, a list of the 5 policy areas. We also asked about the importance of each policy partner with the following question: “How important is each of the following organizations to policy and advocacy related to IVP?” We provided the following response options: not important, slightly important, important, and very important. Finally, each participant classified his or her organization type as government agency, nonprofit, for profit, school/university, hospital/clinic, coalition organization, foundation, or voluntary/advocacy organization.

We calculated the organization-level response rate by dividing the number of organizations that responded to the survey in each of the 15 jurisdictions by the number invited in the jurisdiction. Response rates ranged from 23 of 36 (64%) to 11 of 11 (100%) (mean, 76%). Using the Pearson χ² test, we found a significant association between organization type and participation (P < .001), with standardized residuals indicating significantly fewer-than-expected for-profit (37.5%) and voluntary/advocacy (55.1%) organizations participating. Of the 319 participating organizations (including LHDs and their partners), 280 were represented by a single participant, 29 by 2 participants, and 10 by 3 to 6 participants. Additional details on data collection are available elsewhere.⁴¹ The Washington University in St Louis Institutional Review Board approved this study.

Data Management

In each IVP policy network, 2 network members were considered linked if 1 or both of the organizations indicated working together on IVP policy. To determine how often each organization in the network was perceived as important, we counted all nominations of important or very important. For those organizations not responding to the survey, we determined organization type by examining the organization’s website. In addition, we reviewed all government agencies in the network to identify the presence of specific types of governmental organizations considered important to IVP, including police and criminal justice, transportation, justice, and child protective services.

Analysis

We computed network descriptive statistics, such as density, transitivity, and diversity. Density and transitivity measure network connectivity. Density is computed by dividing the number of connections in a network by the total number of possible connections. Transitivity is a measurement of how often a “friend of a friend is also a friend.” In other words, if organizations A and B are connected and if organizations B and C are connected, the triad is considered “transitive” when A and C are also connected to form a triangle. Transitivity ranges from 0 (no triangles) to 1 (all possible triangles).

We measured diversity by using the index of dispersion and tie assortativity. The index of dispersion quantifies the extent to which characteristics are distributed among members of a group. In the Big Cities Health Coalition networks, we used the index of dispersion to quantify the extent to which network members were distributed across the 8 organization types. The index ranged from 0 (all members were the same organization type) to 1 (members were equally distributed across all organization types). Tie assortativity measures the extent to which ties connect organizations from 2 types, and it ranges from –1 to 1. A network in which organizations were connected to other organizations of only the same type would have an assortativity score of –1, whereas networks in which organizations were linked only to others of the different types would have an assortativity score of 1.

We also computed network member degree centrality for comparison with the importance nominations that the network member received. Degree centrality for a network member was defined as the number of connections that the member had. It could range from 0 to n – 1, where n was the number of network members. We compared the number of working relationships (degree) with the number of importance nominations that each organization had to determine whether organizations that were considered important were also those with the most working relationships. We used the Pearson χ² test to compare proportions across groups and analysis of variance to compare means across groups.

Finally, we used exponential random graph modeling to determine whether organization type was associated with tie formation in the network. Exponential random graph modeling is an inferential method with a form and interpretation similar to logistic regression. In exponential random graph modeling, the characteristics of network members and network structures predict the probability of tie formation between 2 network members, and a tie between 2 network members is the outcome.⁴⁴ We also included a term in each model coded as 1 for the LHD and 0 for all other organizations to control for tie formation by the LHD.

Results

An average of 21.4 (SD = 9.4) local organizations worked together on IVP policy per city (Table). The 15 networks—each including the LHD and its local policy partners as network members and the relationships among them (1 network per city)—ranged in size from 8 to 45 local organizations.

Table.

Structural characteristics of injury and violence prevention policy networks^a in 15 large US cities, 2014

Characteristic	Mean (SD)
Size, no. of organizations	21.4 (9.4)
Connectivity (ie, extent to which the network members are connected)
Density (ie, no. of ties out of total possible ties; range, 0-1)	0.23 (0.08)
Transitivity (ie, extent to which there are triangle formations of ties in the network; range, 0-1)	0.40 (0.13)
Diversity (ie, extent to which the network comprises different types of partners and ties among different types of partners)
Member type dispersion (ie, how much diversity there is in network member types; range, 0-1)	0.84 (0.08)
Tie assortativity (ie, extent to which ties are among members of different types; range, –1 to 1)	–0.05 (0.08)

^aInjury and violence prevention policy networks comprise local health departments from the Big Cities Health Coalition and its local injury and violence prevention policy partners, including foundations, government agencies, nonprofit organizations, schools and universities, and voluntary/advocacy organizations (n = 319).

Types of Organizations Involved in IVP Policy Work at the Local Level

All 15 networks included government, nonprofit, and voluntary/advocacy organizations; 14 included coalitions (ie, a group of organizations working together for 1 purpose) and schools/universities; 12 included hospitals and clinics; 6 included foundations; and 7 included for-profit organizations. Nonprofit organizations constituted the largest proportion of organizations (32.4% of each network on average), followed by government agencies (26.3% of each network); for-profit organizations represented an average of 5.7% of each of the 7 networks in which they were included (Figure 1a). Of government agencies identified as important to IVP, 5 networks included police, 5 included justice, 4 included transportation, and 2 included child protective services.

Figure 1.

Network composition and perceived importance, by organization type, for members of 15 injury and violence prevention policy networks in 15 large US cities, 2014. a, Mean percentage of each network by organization type. b, Mean percentage of organizations of each type in a network that were rated as important by the other organizations in the network. The 15 injury and violence prevention policy networks comprised local health departments from the Big Cities Health Coalition and its local injury and violence prevention policy partners. Data were collected from network members (n = 319) through an online survey conducted from August to November 2014. Error bars show 95% CIs.

Overall, dispersion of organization type was high, ranging from 0.71 to 0.94, with an average of 0.84 (SD = 0.08) on a scale from 0 (all from the same sector) to 1 (equal number from all sectors). Although dispersion of organization type was high, the other measure of diversity—assortativity of ties—was low (Table), indicating that organizations were forming ties with other organizations of the same type more often than with organizations of another type.

Organization Types Perceived as Important to IVP Policy Work Locally

Organizations received a mean of 7.0 importance nominations (SD = 3.7). Government agencies were perceived as important by the most partners; a mean of 42.8% of other organizations in a network perceived them as important or very important, followed by for-profit organizations (25.8%) and voluntary/advocacy organizations (24.3%; Figure 1b). We found no significant association between the mean number of importance nominations and organization type (analysis of variance, F_97,79 = –1.33, P = .25). However, we found a mismatch between importance nominations and working relationships for some organizations. The difference between the number of working relationships and the number of importance nominations ranged from –11 to 19 (median, −2), indicating that organizations tended to have an average of 2 more importance nominations than working relationships. However, the distribution of the difference between the number of working relationships and the number of importance nominations had a large range and was right skewed. So, some organizations were perceived as important by many but were not connected through working relationships, whereas others were highly connected through working relationships but not perceived as important.

Figure 2 shows a policy network of near-average size. A few nodes are large, indicating many working relationships, but the white circles inside the nodes are much smaller, indicating few importance nominations. The network includes 10 nonprofit organizations that range in size from small to large. Three of the 4 government agencies were more connected (as shown by larger size) than most of the policy partners in this network; the 2 school/university nodes were different sizes; and the voluntary/advocacy agencies had the fewest total ties (n = 3) of any organization type. The only foundation in the network was perceived as important by few others in the network despite being connected to numerous organizations, whereas the 2 voluntary/advocacy organizations had nearly as many importance nominations as working relationships.

Figure 2.

Example of an injury and violence prevention policy network comprising a local health department (LHD) from the Big Cities Health Coalition and its local injury and violence prevention policy partners, United States, 2014. Each node (circle) represents an organization in the network. Node size indicates the number of working relationships; the larger the node, the greater the number of working relationships. The white circle inside each node indicates the number of times that the organization was rated by others in the network as important or very important; the larger the white circle, the greater the number of importance nominations. Lines indicate a working relationship between 2 organizations. Data collected from network members through an online survey conducted from August to November 2014.

How Organizations Partnered to Work on IVP Policy

We examined the likelihood of forming ties with others in the network by perceived importance and organization type. Perceived importance was significantly associated with tie formation in 10 of the 15 networks (Figure 3a), with odds ratios ranging from 1.07 (95% CI, 1.02-1.13) to 2.35 (95% CI, 1.68-3.28). That is, for each importance rating that an organization received, the likelihood of forming a tie increased by 7% to 135%. After controlling for importance nominations, we found that organization type was significantly associated with tie formation in 10 networks (Figure 3b). In the 10 networks where organization type was associated with tie formation, for-profit organizations were no more likely or less likely than government organizations to form partnerships, and foundations were more likely than government agencies to form ties in 1 network. Coalitions, hospitals/clinics, nonprofit organizations, schools/universities, and voluntary/advocacy organizations were more likely to form ties in some networks and less likely to form ties in other networks when compared with government agencies.

Figure 3.

Odds ratios of tie formation among members of 15 injury and violence prevention policy networks, with each network comprising local health departments from the Big Cities Health Coalition and its local injury and violence prevention policy partners, United States, 2014. Error bars indicate 95% CIs. Data collected from network members through an online survey (n = 319) conducted from August to November 2014. a, Odds of forming a working relationship by importance nomination for each jurisdiction. b, Odds of forming a working relationship with others in the network, by organization type, with government organizations as the reference group.

Discussion

Preventing and responding to injury and violence cannot be accomplished by any single organization or sector.⁴⁵ The networks that we examined were moderately dense and had highly diverse organizational types, which is promising for public health given that (1) dense networks are associated with successful local public health practice^28
–30 and (2) LHDs in networks with diverse partners are better than LHDs without diverse partners at providing essential services to their constituents.^{28,29,31
–33} In addition to broad structural and compositional characteristics, such as density and diversity, specific partner types within certain sectors can influence performance. Police and criminal justice, education, hospitals/medical service providers, social services and child protection, alcohol regulators, local public health, transportation, justice, and nongovernmental organizations are important sectors for IVP.^34,35 All networks included nongovernmental organizations and LHDs. The function of alcohol regulation is typically conducted by LHDs, which are in all networks. Most networks included partners from education and hospitals/medical service providers. However, few networks included partners from police and criminal justice, child protective services, or transportation. The lack of police and criminal justice collaborators could be related to barriers identified in previous research—for example, the perception of certain stakeholders that law enforcement focuses on punishment rather than on prevention and might dominate collaborative efforts.³⁹

We do not know if stakeholders in our study were unaware of evidence on the importance of certain sectors (eg, police, transportation) to IVP, were aware but unsuccessful in engaging these sectors, or faced barriers to collaboration.³⁹ Future work might focus on developing and disseminating (1) strategies for increasing awareness of sectors to include in IVP networks and (2) strategies to understand and overcome barriers to improve engagement of important agencies that are not currently collaborators. For example, for-profit organizations are important partners in public health^46,47 that are not always well integrated into collaborative efforts.^29,47 In our study, for-profit organizations were included in fewer than half of the networks, but they were equally as likely as government organizations to form ties and were considered important or very important by an average of 25.8% of their partners. Identifying strategies that engage for-profit organizations might help other jurisdictions improve their networks. Likewise, more in-depth examination of IVP networks in which police and criminal justice are involved and are perceived as important by their partners could identify strategies for adding police and criminal justice partners. Finally, multiple studies have identified social connections across organizations^48,49 and linkages to the same boards or associations^50,51 as facilitators of partnerships, suggesting that partnering might be encouraged by engaging in indirect or less formal connections with important but missing partners and partner types.

Limitations

This study had several limitations. First, our results may not be generalizable to networks that are not in large urban areas. Specifically, findings from urban areas may not translate well to rural jurisdictions. The organizational networks of LHDs in rural areas tend to be less diverse than those in urban settings,³² and rural health departments are less active than urban health departments in creating policies.^52
–54 The study may also have been limited by our network delineation process: participants may have been unaware of additional local partners. In addition, many organizations were represented by 1 participant, who may not have been aware of all the connections maintained by their organization.

Conclusion

To our knowledge, this study is the first to map local IVP policy networks, providing new insight into the IVP policy process. Our results provide evidence on the density and diversity of existing IVP networks, areas in need of improvement in the network composition, and a better understanding of how collaboration influences IVP policy and programming. Future research should examine how the composition and structure of local IVP policy networks influence the development, implementation, and success of local IVP policy.

Footnotes

Acknowledgments

We would like to acknowledge the contribution of Shannon Carrillo, who was instrumental in the initial data collection and management for this project.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by the de Beaumont Foundation and was conducted in partnership with the National Association of County and City Health Officials.

References

Sumner

Mercy

Dahlberg

. Violence in the United States: status, challenges, and opportunities. JAMA. 2015;314(5):478–488.

Haegerich

Dahlberg

Simon

. Prevention of injury and violence in the USA. Lancet. 2014;384(9937):64–74.

Mouzakes

Koltai

Kuhar

. The impact of airbags and seat belts on the incidence and severity of maxillofacial injuries in automobile accidents in New York State. Arch Otolaryngol Head Neck Surg. 2001;127(10):1189–1193.

Brown

. Costs and benefits of enforcing housing policies to prevent childhood lead poisoning. Med Decis Making. 2002;22(6):482–492.

Murphy

Baker

Leo-Summers

. Integration of fall prevention into state policy in Connecticut. Gerontologist. 2013;53(3):508–515.

Loftin

McDowall

Wiersema

. Effects of restrictive licensing of handguns on homicide and suicide in the District of Columbia. N Engl J Med. 1991;325(23):1615–1620.

Chapman

Alpers

Agho

. Australia’s 1996 gun law reforms: faster falls in firearm deaths, firearm suicides, and a decade without mass shootings. Inj Prev. 2006;12(6):365–372.

Webster

Starnes

. Reexamining the association between child access prevention gun laws and unintentional shooting deaths of children. Pediatrics. 2000;106(6):1466–1469.

Santaella-Tenorio

Cerda

Villaveces

. What do we know about the association between firearm legislation and firearm-related injuries? Epidemiol Rev. 2016;38(1):140–157.

10.

Kalesan

Mobily

Keiser

. Firearm legislation and firearm mortality in the USA: a cross-sectional, state-level study. Lancet. 2016;387(10030):1847–1855.

11.

Webster

Crifasi

Vernick

. Effects of the repeal of Missouri’s handgun purchaser licensing law on homicides. J Urban Health. 2014;91(2):293–302.

12.

Raissian

. Hold your fire: did the 1996 federal Gun Control Act expansion reduce domestic homicides? J Policy Anal Manag. 2016;35(1):67–93.

13.

Kuo

Sullivan

. Aggression and violence in the inner city: effects of environment via mental fatigue. Environ Behav. 2001;33(4):543–571.

14.

Branas

Cheney

MacDonald

. A difference-in-differences analysis of health, safety, and greening vacant urban space. Am J Epidemiol. 2011;174(11):1296–1306.

15.

Beyer

Wallis

Hamberger

. Neighborhood environment and intimate partner violence: a systematic review. Trauma Violence Abuse. 2015;16(1):16–47.

16.

Wagenaar

Tobler

Komro

. Effects of alcohol tax and price policies on morbidity and mortality: a systematic review. Am J Public Health. 2010;100(11):2270–2278.

17.

Chaloupka

Grossman

Saffer

. The effects of price on alcohol consumption and alcohol-related problems. Alcohol Res Health. 2002;26(1):22–34.

18.

Grossman

Markowitz

. Alcohol Regulation and Violence on College Campuses. Cambridge, MA: National Bureau of Economic Research; 1999.

19.

Sloan

Reilly

Schenzler

. Effects of prices, civil and criminal sanctions, and law enforcement on alcohol-related mortality. J Stud Alcohol. 1994;55(4):454–465.

20.

McGowan

Hahn

Liberman

. Effects on violence of laws and policies facilitating the transfer of juveniles from the juvenile justice system to the adult justice system: a systematic review. Am J Prev Med. 2007;32(suppl 4):S7–S28.

21.

Provan

Kenis

. Modes of network governance: structure, management, and effectiveness. J Public Admin Res Theory. 2008;18(2):229–252.

22.

Provan

Milward

. Do networks really work? A framework for evaluating public-sector organizational networks. Public Admin Rev. 2001;61(4):414–423.

23.

Roussos

Fawcett

. A review of collaborative partnerships as a strategy for improving community health. Annu Rev Public Health. 2000;21:369–402.

24.

Holder

Reynolds

. Application of local policy to prevent alcohol problems: experiences from a community trial. Addiction. 1997;92(suppl 2):S285–S292.

25.

Moreland-Russell

Carothers

. An examination of two policy networks involved in advancing smokefree policy initiatives. Int J Environ Res Public Health. 2015;12(9):11117–11131.

26.

Harris

Richter

Paine-Andrews

. Community partnerships: review of selected models and evaluation of two case studies. J Nutr Educ. 1997;29(4):189–195.

27.

Ramanadhan

Salhi

Achille

. Addressing cancer disparities via community network mobilization and intersectoral partnerships: a social network analysis. Plos One. 2012;7(2):e32130.

28.

Lovelace

. External collaboration and performance: North Carolina local public health departments, 1996. Public Health Rep. 2000;115(4):350–357.

29.

Mays

Scutchfield

Bhandari

. Understanding the organization of public health delivery systems: an empirical typology. Milbank Q. 2010;88(1):81–111.

30.

Merrill

Keeling

Carley

. A comparative study of 11 local health department organizational networks. J Public Health Manag Pract. 2010;16(6):564–576.

31.

Mays

Scutchfield

. Improving public health system performance through multiorganizational partnerships. Prev Chronic Dis. 2010;7(6):A116.

32.

Beatty

Harris

Barnes

. The role of inter-organizational partnerships in health services provision among rural, suburban, and urban local health departments. J Rural Health. 2010;26(3):248–258.

33.

Varda

Chandra

Stern

. Core dimensions of connectivity in public health collaboratives. J Public Health Manag Pract. 2008;14(5):E1–E7.

34.

World Health Organization. Preventing Youth Violence: An Overview of the Evidence. Geneva, Switzerland: World Health Organization; 2015. http://apps.who.int/iris/bitstream/10665/181008/1/9789241509251_eng.pdf?ua=1. Accessed May 7, 2016.

35.

Koné

Zurick

Patterson

. Injury and violence prevention policy: celebrating our successes, protecting our future. J Safety Res. 2012;43(4):265–270.

36.

Bonnie

Fulco

Liverman

, eds. Reducing the Burden of Injury: Advancing Prevention and Treatment. Washington, DC: National Academies Press; 1999.

37.

Florence

Shepherd

Brennan

. Effectiveness of anonymised information sharing and use in health service, police, and local government partnership for preventing violence related injury: experimental study and time series analysis. BMJ. 2011;342:d3313.

38.

Horan

Mallonee

. Injury surveillance. Epidemiol Rev. 2003;25:24–42.

39.

Giacomazzi

Smithey

. Community policing and family violence against women: lessons learned from a multiagency collaborative. Police Q. 2001;4(1):99–122.

40.

Friedman

Reynolds

Quan

. Measuring changes in interagency collaboration: an examination of the Bridgeport Safe Start Initiative. Eval Program Plann. 2007;30(3):294–306.

41.

Harris

Leider

Carothers

. Multisector health policy networks in 15 large US cities. J Public Health Manag Pract. 2016;22(6):520–528.

42.

Harris

Clements

. Using social network analysis to understand Missouri’s system of public health emergency planners. Public Health Rep. 2007;122(4):488–498.

43.

Harris

Luke

Burke

. Seeing the forest and the trees: using network analysis to develop an organizational blueprint of state tobacco control systems. Soc Sci Med. 2008;67(11):1669–1678.

44.

Harris

. An Introduction to Exponential Random Graph Modeling. Thousand Oaks, CA: Sage Publications Inc; 2014.

45.

Fisher

, ed. Violence Against Women and Family Violence: Developments in Research, Practice, and Policy. Washington, DC: National Institute of Justice; 2004.

46.

Buehler

Whitney

Berkelman

. Business and public health collaboration for emergency preparedness in Georgia: a case study. BMC Public Health. 2006;6:285.

47.

Harris

Roche

Estlund

. Partnering to create a more livable city: the Livable St Louis Network. J Public Health Manag Pract. 2014;20(4):384–391.

48.

Shaw

. Successful collaboration between the nonprofit and public sectors. Nonprofit Manag Leadersh. 2003;14(1):107–120.

49.

Letaifa

Rabeau

. Too close to collaborate? How geographic proximity could impede entrepreneurship and innovation. J Bus Res. 2013;66(10):2071–2078.

50.

Brownson

Parra

Dauti

. Assembling the puzzle for promoting physical activity in Brazil: a social network analysis. J Phys Act Health. 2010;7(suppl 2):S242–S252.

51.

Guo

Acar

. Understanding collaboration among nonprofit organizations: combining resource dependency, institutional, and network perspectives. Nonprofit Voluntary Sector Q. 2005;34(3):340–361.

52.

Harris

Mueller

. Policy activity and policy adoption in rural, suburban, and urban local health departments. J Public Health Manag Pract. 2013;19(2):E1–E8.

53.

Hearne

Castrucci

Leider

. The future of urban health: needs, barriers, opportunities, and policy advancement at large urban health departments. J Public Health Manag Pract. 2015;21(suppl 1):S4–S13.

54.

National Association of County and City Health Officials. 2013 national profile of local health departments. http://archived.naccho.org/topics/infrastructure/profile/upload/2013-National-Profile-of-Local-Health-Departments-report.pdf. Published January 2014. Accessed July 10, 2016.