Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Replicable, Solution-Focused Approach to Cross-Sector Data Sharing for Evaluation of Community Violence Prevention Programming

A Replicable, Solution-Focused Approach to Cross-Sector Data Sharing for Evaluation of Community... Practice Full Report A Replicable, Solution-Focused Approach to Cross-Sector Data Sharing for Evaluation of Community Violence Prevention Programming Carlene A. Mayfield, PhD, MPH; Rachel Siegal, BS; Melvin Herring, PhD, MSW; Tracie Campbell, MS, CHES; Catie L. Clark, PhD; Jennifer Langhinrichsen-Rohling, PhD ABSTRACT Context: Community violence is a public health problem that erodes social infrastructure. Structural racism contributes to the disparate concentration of violence in communities of color. In Mecklenburg County, North Carolina, increasing trends in community violence show racial and geographic disparities that emphasize the need for cross-sector, data-driven approaches to program and policy change. Cross-sector collaborations are challenged by data sharing barriers that hinder implementation. Program: In response to community advocacy, Mecklenburg County Public Health (MCPH) launched a Community Violence Prevention Plan with evidence-based programming. The Cure Violence (CV) model, a public health approach to disrupting violence through equitable resource provision, network building, and changing norms, was implemented at the community level. The Health Alliance for Violence Intervention (HAVI) model, a hospital-based screening and case management inter- vention for victims of violence, was implemented at Carolinas Medical Center in Charlotte, the region’s only level I trauma center. Methods: A data collaborative was created to optimize evaluation of CV and HAVI programs including MCPH, the city of Charlotte, Atrium Health, Charlotte-Mecklenburg Schools, Johnson C. Smith University, and the University of North Carolina Charlotte. A comprehensive approach to facilitate data sharing was designed with a focus on engaging stakeholders and generating solutions to commonly reported barriers. Structured interviews were used to inform a solution-focused strategy. Results: Stakeholders reported perceptions of their organization’s barriers and facilitators to cross-sector data sharing. Com- mon technology, legal, and governance barriers were addressed through partnership with a local integrated data system. Solutions for trust and motivational challenges were built into ongoing collaborative processes. Discussion: Data silos inhibit the understanding of complex public health issues such as community violence, along with the design and evaluation of collective impact efforts. This approach can be replicated and scaled to support cross-sector collaborations seeking to influence social and health inequities stemming from structural racism. KEY WORDS: barriers and facilitators, community violence, data sharing, researcher-practitioner partnerships partnership and on this effort, and Octavia Ramsey from Johnson C. Smith Author Affiliation: Department of Community Health, Atrium Health, University for supporting the interview coding process. Charlotte, North Carolina (Dr Mayfield); Department of Psychological This work was supported, in part, by the Data Across Sectors for Health Sciences, University of North Carolina Charlotte, Charlotte, North Carolina (Ms (DASH) Mentor 3.0 Program. Siegal and Dr Langhinrichsen-Rohling); Department of Social Work, Johnson The authors have indicated they have no potential conflicts of interest to C. Smith University, Charlotte, North Carolina (Dr Herring); Office of Violence disclose. Prevention, Mecklenburg County Public Health, Charlotte, North Carolina (Ms Campbell); and Mecklenburg County Criminal Justice Services, Charlotte, Supplemental digital content is available for this article. Direct URL citations North Carolina (Dr Clark). appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (http://www.JPHMP.com). A number of people have made significant contributions to the creation and continuous success of the Community Violence Prevention Data This is an open-access article distributed under the terms of the Collaborative. The authors acknowledge Rebecca Hefner, Chief Data Officer Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 for the City of Charlotte, and Donna Smith, Epidemiologist for Mecklenburg (CCBY-NC-ND), where it is permissible to download and share the work County Public Health, as co-chairs of the collaborative, and Dr Lori Thomas, provided it is properly cited. The work cannot be changed in any way or used Executive Director of the Institute for Social Capital, for their expert guidance commercially without permission from the journal. and continuous support throughout the evaluation design process. The Correspondence: Carlene A. Mayfield, PhD, MPH, Department of Community authors also express gratitude to other collaborative members, Elyse Health, Atrium Health, 4135 S Stream Blvd, Charlotte, NC 28217 Hamilton-Childres, Carole McKernan, Haita Toure, Monica Nguyen, Robert (carlene.mayfield@atriumhealth.org). Broughton, Kaiti Mrak, Stacey Butler, Lindsay Messinger, and Kelly Moriarty, Copyright © 2022 The Authors. Published by Wolters Kluwer Health, Inc. along with their DASH Mentors Robert Gradeck from the University of DOI: 10.1097/PHH.0000000000001426 Pittsburgh and Erikka Gilliam from Drexel University, for their feedback and January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S43 S44 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing Context and evaluation. While the benefits of data shar- ing are well supported, the additional burden of Violence is a recognized public health problem technological and legal infrastructure required to with long-term consequences for individuals and adhere with federal regulations is prohibitive for communities. Homicide is the third leading cause of 18 many organizations. Integrated Data Systems (IDSs) death in the United States among 15- to 34-year-olds, represent a network of partnerships to link admin- and the leading cause of death among Black males istrative data across governmental and nonprofit in the same age group. The overall rate of homi- organizations providing social and health care ser- cide in the United States is also 7.5 times higher than vices within a community. By providing a legal in comparable high-income countries. Community and data governance framework, IDSs facilitate data violence occurs when interpersonal violence, the in- sharing between organizations to improve the under- tentional use of physical force against another person standing of complex community issues beyond typical resulting in injury or death, is perpetuated by individ- service silos. The Institute for Social Capital (ISC) is uals not intimately related to the victim. Common a registered nonprofit IDS in Mecklenburg County, examples include physical assault, property crimes, North Carolina, that serves as a signature research ini- group fights, and public shootings. Survivors of vi- tiative of the University of North Carolina Charlotte’s olence are at a greater risk for repeated victimization, Urban Institute. The ISC integrates data from more with an estimated 11% to 44% of hospital trauma than 45 partners across local agencies, sectors, and admissions related to violence occurring among pa- programs to increase the community’s capacity for tients with prior admissions. Long-term exposure to data-informed decision making and advance research violence also has profoundly harmful effects on child that deepens understanding of complex community development and emotional well-being and erodes the issues. 7,8 social and economic infrastructure of communities. The objective of this article is to describe a replica- Episodes of violence are interconnected by struc- ble strategy for cross-sector data sharing to support tural racism through the policies and practices that the evaluation of community violence prevention contribute to the disparate concentration of violence efforts using the technological and governance infras- in communities of color. Individual risk factors for tructure of the ISC. Our solution-focused partnership violence such as low family income, high unemploy- engagement approach was informed by structured ment, and lack of available resources, along with interviews with organizational stakeholders that in- witnessing and experiencing violence, are perpetu- cluded a discussion of barriers and facilitators to ated by mutually reinforcing systems of structural in- cross-sector data sharing. Results were used to design equity and community disinvestment. Historical poli- a responsive strategy to facilitate cross-sector data cies such as redlining, the practice of denying credit sharing and collective engagement in the evaluation for home loans based on the location of property of violence prevention activities. within a minority or disadvantaged neighborhood, have enduring effects on contemporary racialized Program residential segregation, and the inequitable distribu- tion of social and health risk factors for community Setting violence. Consequently, solutions for community violence require cross-sector collaborations that ad- Mecklenburg County, North Carolina, encompass- dress racial discrimination within housing, education, ing the city of Charlotte, is the largest and most employment, health care, and criminal justice sys- diverse area in the Carolinas. As a “minority major- tems through program and policy change. Often ity”county, more than half of Mecklenburg’s residents communities face the simultaneous adoption of mul- are people of color, with non-Hispanic White res- tiple prevention efforts by disparate agents of change, idents accounting for 46% of the population in 20 21 making the study of program-specific impacts both 2019. A widely circulated study by Chetty et al necessary and challenging. ranked Charlotte last among America’s 50 largest cities in economic mobility, highlighting a cycle of in- Public health decision making is becoming in- tergenerational poverty, poor educational attainment, creasingly reliant on complex data sharing and inte- and health disparities that disproportionately impact gration, creating additional barriers for cross-sector the communities’ residents of color. Data from the collaborations. Advances in health information county’s community health assessments show increas- technology (IT) and informatics sciences, including ing trends in community violence with notable racial the widespread adoption of electronic health records and geographic disparities. Between June 2018 and by community health sites and health care sys- June 2019, the city of Charlotte reported a 119% tems, have enhanced the capacity for surveillance January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S45 increase in homicides, along with a higher incidence Intervention (HAVI) framework, at Atrium Health, in areas with concentrated social and health risk fac- CMC Charlotte. tors including poverty, low educational attainment, high unemployment, and worse access to health care Methods relative to the larger county. Atrium Health is one of the largest integrated, non- Violence Prevention Data Collaborative profit health systems in the southeast region with 37 The Violence Prevention Data Collaborative (Data hospitals and more than 1350 care locations across Collaborative) was formed to leverage cross-sector re- North Carolina, South Carolina, and Georgia. In ad- sources for data sharing, analysis, reporting, and eval- dition to being the largest health care provider for uation in support of activities aligned with the Com- historically marginalized and indignant patients in the munity Violence Prevention Plan. Partners included local community, Atrium Health provides the region’s multiple Mecklenburg County departments (Public only level I trauma center through the Carolinas Med- Health, Criminal Justice Services, Department of So- ical Center (CMC), Charlotte facility. Results from cial Services, and Community Support Services), city a recent (2019) publication show that a majority of of Charlotte (Innovation & Technology, Charlotte- patients admitted to CMC for violent trauma be- Mecklenburg Police Department), and institutional tween 2009 and 2015 were Black (64.9%) and male partners from health care (Atrium Health) and educa- (87.4%) and that 24.9% of patients without mor- tion systems (Charlotte-Mecklenburg Schools, John- tality were readmitted with a second violent injury son C. Smith University, and University of North within an average of 31.9 ± 21.0 months. Vio- Carolina Charlotte). Each partner is represented by at lent injury was also independently associated with least one organizational member. Members reviewed higher morbidity and mortality related to the in- and signed a nonlegally binding Memorandum of dex trauma and subsequent injuries among otherwise Understanding (MOU) that details the terms of healthy young adults. Disparities in violent injury are collaboration including governance and leadership, a symptom of long-standing systemic inequity in the collaborative charge and priorities, and membership larger community. Thus, equitable solutions require a expectations. comprehensive approach to intervention and sec- ondary prevention that is informed by the evaluation of cross-sector data. The Institute for Social Capital The ISC uses individual-level identifiable data so that information can be linked across different sources. Community Violence Prevention Plan Because of the sensitive nature of personally identi- Community stakeholders responded to these findings fiable information, the ISC has numerous policies and by mobilizing partnerships and aligning resources to procedures in place to protect the data including le- support a comprehensive approach to community gal, technical, procedural, and physical protections. violence prevention. In 2019, the city of Charlotte The ISC establishes data sharing agreements with each launched SAFE Charlotte, a comprehensive approach data depositor that are aligned with all federal and in- to improve police/community relations, reduce violent stitutional regulations. To obtain data from the ISC, crime, and create opportunities for social mobility. At requestors must complete a data license request that the same time, Mecklenburg County Public Health details the evaluation project, requested data fields, launched a Community Violence Prevention Plan, the methodology and analyses that will be used, and including the creation of the Office of Violence Pre- the dissemination plan. Data requests are reviewed vention and the implementation of 2 new programs by the ISC Data and Research Oversight Commit- using evidence-based models for community violence tee (DAROC), which oversees the use of data for prevention. First, and in partnership with the city, the ISC Board of Directors and includes university the Cure Violence (CV) violence interrupter model, researchers, members of the community, and a rep- a data-driven, research-based, community-centric ap- resentative from each data depositor. The agency that proach to violence prevention, was implemented at owns the data being requested (ie, the depositor) must the community level by a local nonprofit agency, approve the use of their data during the approval pro- Youth Advocate Program (YAP), with priority given cess. If a request is approved, the requestor will sign a to neighborhoods experiencing the highest rates of vi- Limited License Agreement (LLA) outlining how the olence. Second, a hospital-based screening and case data can and cannot be used. After the LLA is com- management intervention was implemented at the in- plete and requested data are integrated, the ISC and dividual level using the Health Alliance for Violence DAROC members will review the data set prior to S46 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing release to ensure it meets the terms of the LLA and The researchers used deductive thematic analysis, does not inadvertently identify any individual. Upon informed by the 6 Wiehe categories, to code the data approval, de-identified data are securely transferred to for barriers and solutions to cross-sector data shar- the requestor. ing. For example, participants’ reference to a lack of personnel to support data sharing was coded as an economic barrier. Following initial coding, the Stakeholder engagement researchers revised the definitions and included ad- We developed a solution-focused approach to engage ditional examples for each of the 6 Wiehe categories stakeholders in cross-sector data sharing to support to better capture the breadth of participant responses. evaluation of the new CV (community-based) and The most significant change was to expand the defini- HAVI (hospital-based) interventions. This process in- tion of motivational barriers to include both barriers cluded the design and execution of an integrated for individuals charged with data sharing and moti- data request though the ISC governance process. A vational barriers for the data holding organization. conceptual framework is summarized in the Figure. Throughout the initial coding and coding refinement, First, collaborative members representing the 2 uni- the coders discussed any disagreements until a con- versities (the researchers) conducted semistructured sensus was reached. However, given that stakeholders interviews with the other data sharing stakehold- specifically responded to queries about each of the 6 ers. The researchers developed the interview guide categories of barriers in the interview, coding disagree- using an existing data sharing barriers framework, ments were rare. Instead, discussion typically focused which describes 6 categories of barriers: technical, on revising the Wiehe framework to capture barriers motivational, economic, political, legal, and ethical. occurring at multiple levels (individual, collaborative, During the interview, the researchers asked a general organizational). question about participants’ barriers to data sharing: Following coding, participants provided feedback “From your perspective, what are the main barriers to on the identified themes in an extended Data Col- cross-sector data sharing in your organization?” The laborative meeting and as part of a member check. researchers then asked participants a set of 6 ques- No additional themes were added or removed fol- tions, each of which focused on a different barrier lowing the member check. Participants also discussed from the Wiehe framework (eg, Motivational bar- potential solutions to the identified barriers. Solutions riers to data sharing can include a lack of incentives that could be implemented immediately were identi- or opportunity cost disincentives or disagreement on fied by the Data Collaborative as next steps in the data data use. What do motivational barriers look like at sharing process. The stakeholder interview protocol your organization?). for this process was reviewed and approved by the FIGURE Conceptual Framework Abbreviations: DAROC, Data and Research Oversight Committee; LLA, Limited License Agreement. January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S47 institutional review board at University of North Car- Content Table 1, available at http://links.lww.com/ olina Charlotte (#21-0379). JPHMP/A857, and Supplemental Digital Content Second, a literature review was conducted to iden- Table 2, available at http://links.lww.com/JPHMP/ tify published studies from programs designed under A858) were presented to the Data Collaborative each of the evidence-based frameworks. Studies with in a series of three 90-minute meetings that oc- design and implementation similar to the Commu- curred bimonthly. Members brainstormed solutions nity Violence Prevention Plan were prioritized. For to data sharing barriers that were raised (outlined in HAVI-based programs, we included studies that were Table 1) and identified evaluation priorities for each located at level I trauma centers, excluded victims of of the evidence-based programs. For the HAVI pro- domestic violence, sexual assault, and child abuse, and gram, the Data Collaborative wanted to answer the utilized some form of case management and commu- following: (i) Did the program serve the intended pop- nity resource referral as the primary intervention. For ulation? (ii) What community resources/community CV-based programs, we included studies that were lo- partners were most frequently used by participants? cated in US cities, implemented interventions at the (iii) What impact did the program have on injury individual and community levels, and relied on some recidivism, clinical utilization, and repeat victimiza- form of outreach workers, violence interrupters, or a tion/perpetration? For the CV program, the Data combination of the two. Collaborative wanted to answer the following: (i) Third, we conducted a series of expert review meet- Did the program serve the intended population? ings with Data Collaborative members to synthesize (ii) Who referred clients to the program? (iii) Was results from the previous steps (stakeholder interviews the program successfully implemented? (iv) What and the literature review) to define evaluation ques- resources/community partners were clients most fre- tions and align metrics, data sources, and elds fi needed quently referred to and what resources/community for data sharing. Specifically, results from the stake- partners did clients most frequently use? Metrics and holder interviews were used to inform the process of data sources required to answer these questions were achieving data sharing and data dissemination while identified and are summarized in Table 2. the literature review was used to inform the metrics, and subsequent data fields, identified for data sharing. Discussion We designed a solution-focused approach to achieve Results cross-sector data sharing that was informed by A total of 10 members representing 8 distinct orga- key stakeholders and built on the legal and gover- nizations were interviewed, comprising 100% par- nance infrastructure of the ISC, a local IDS. The ticipation from the Data Collaborative members. stakeholder interviews contributed several important Preliminary analyses identified a range of barriers to learning objectives for the Data Collaborative. First, data sharing across technical, legal, ethical, and moti- even organizations with sustainable financing that vational domains. Participants expressed inconsistent are motivated to engage in data sharing may ex- perspectives on the benefits of cross-sector data shar- perience different levels of organizational readiness ing, along with potential concerns about the legal and for data sharing. Some of these organizations re- governance protections available through partnership ported a more informal governance structure, which with the ISC. Some participants felt that their individ- may impact their readiness. In response to this ual and organizational roles in the Data Collaborative finding, members of the collaborative at organiza- were unclear and that the current work lacked a tions with more formal governance structures are strong rationale for sharing sensitive information supporting these organizations’ governance structure and assuming risk. Motivation for data sharing was development process. Second, individuals and orga- aligned with individual members’ position within nizations engaging in cross-sector data sharing often their organization, in addition to their perception of have previous data sharing experiences that include broader structural and leadership support. Finally, experiences of ongoing disempowerment that impact preliminary findings suggested that individuals’ and their current motivation to engage in data sharing. organizations’ different positions (eg, in relation to To address this barrier, we conducted organization-led other collaborative members or in relation to other or- presentations to the collaborative about the organi- ganizations), and individuals’ ability to leverage those zation’s data and data governance practices, previous positions, had the potential to impact data sharing experiences of data sharing, and ethical and legal efforts. frameworks. These presentations provided an avenue Stakeholder interview results and key studies for members to share their experiences and existing from the literature review (see Supplemental Digital concerns. S48 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing TABLE 1 Stakeholder Interview Results and Solutions to Common Data Sharing Barriers Category and Definition Quote Solutions Technical “One challenge is that we have organizations Initiate discussions with ISC and Data Data are not collected, preserved, using different tools and keeping their data Collaborative or available in a usable format; in different environments. Most of our Use 1:1 and group discussions to identify needs technical solutions are not existing technical resources are directed and (re)allocate resources available; a lack of metadata and toward keeping our environment secure; we standards might have to reallocate resources to share data effectively.” Motivational “So I don’t know that the missions changed. Define VPDC mission Mission/Values & Organizational Um, I just, I guess I wasn’t privy to all the Connect evaluation strategy design back to Buy-In things that are being done. And the things I mission Lack of incentives or opportunity was privy to, to me didn’t seem on mission. Leverage Violence Prevention leadership team cost disincentives, disagreement At the time.” to build organizational support on data use Motivational “I don’t have final decision power, but I have Leverage Violence Prevention leadership team Individual Position the ability to influence um higher level to support VPDC members Lack of incentives or opportunity leadership in terms of making Leverage VPDC members’ influence to support cost disincentives, disagreement recommendations about what we should other members on data use prioritize. And I have the ability to like escalate, specific projects up to leadership in a way that is very facilitative for uh those approval processes.” Economic “I think the main one is resources is that we Initiate discussions with ISC and Data Possible economic damage and/or don’t have a dedicated data position. So, um, Collaborative lack of resources for data sharing when we, when we share data, it involves Use 1:1 and group discussions to identify needs being a little piece of a bunch of people’s and (re)allocate resources jobs. And it’sjustait’s not, it would be ideal Leverage Violence Prevention leadership team if we had a dedicated kind of research and to increase resources in the long term data position.” Political “But when, particularly in the current climate, Initiate discussions with ISC and Data Lack of trust and guidelines, when we see data get used in a way that is Collaborative restrictive policies misleading.” Use evaluation strategy design process to collaboratively define evaluation questions and metrics Legal “If there were a data request that required Initiate discussions with ISC and Data Ownership and copyright, individually identifying information, that Collaborative protection would be a limitation. Especially I mean, it Use evaluation strategy design process to would go beyond PHI restrictions.” collaboratively define evaluation questions and metrics, build consensus on risk and reward Ethical “And so we wouldn’t want to, you know, be Initiate discussions with ISC and Data Lack of proportionality and flagging them as like, oh, [we] are Collaborative reciprocity supervising all of these dangerous people?” Use evaluation strategy design process to collaboratively define evaluation questions and metrics, build consensus on risk and reward Abbreviations: ISC, Institute for Social Capital; PHI, protected health information; VPDC, Violence Prevention Data Collaborative. a 25 Categories and definitions adapted from the Wiehe et al solution-based framework for data sharing partnerships. Indicates a change to the original definition. Although the Wiehe framework identifies motiva- the collaborative, and in particular organizations. At tion as one of 6 equivalent barriers to data sharing, the same time, some respondents struggled to under- stakeholders described motivation as a barrier that stand and articulate how the mission and goals of the underlies the other barriers. Lack of motivation, in Data Collaborative aligned with their organization’s conjunction with concerns about sharing sensitive purpose for participation in the Community Violence data collected from vulnerable populations, mani- Prevention Plan. Lacking a clear mission and concrete fested at the level of individual IT stakeholders, within goals is common in cross-sector partnerships and January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S49 TABLE 2 Data Sharing Metrics and Sources HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Demographics Process Age, race/ethnicity, length of Atrium Health Age, race/ethnicity, geographic location, time in program Cure Violence/YAP stay, EMS transport EHR/Data Warehouse Time in program is defined as the number of weeks the client continues to meet with the outreach worker Implementation Process #/% patients meeting selection Atrium Health #/% clients meeting selection criteria screened for program Cure Violence/YAP criteria screened for the EHR/Data Warehouse Selection criteria: Individuals aged 16-24 y, previous exposure program Case notes to gun violence, live in intervention area #/% eligible patients enrolled in #/% eligible clients enrolled in the program the program # VI mediations # people involved in mediation Retention Process Successful case management Atrium Health Successful client management defined as appropriate Cure Violence/YAP defined as follow-up for 30, 60, Case notes closed-loop referral in response to client-identified need 90 d binary (yes/no) #/% successful out of eligible, enrolled clients #/% successful out of eligible, enrolled populations Partnership Process # of CBO partners that own and Atrium Health # of city/county partners using CRH Cure Violence/YAP engagement manage their program listings CRH # of CBO partners that own and manage their program listings CRH on the CRH on the CRH # of CBOs and partners accepting # of partners accepting closed-loop referrals closed-loop referrals # partners (co-) hosting CVG community events Program alignment Process #/% of eligible patients referred Atrium Health #/% clients referred to from HAVI Cure Violence/YAP to CVH Cure Violence Atrium Health CRH CRH (continues) S50 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing fi TABLE 2 Data Sharing Metrics and Sources (Continued) HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Primary outcome Impact Injury recidivism: Hospital Atrium Health Reduced exposure to gun-related violence Cure Violence/YAP readmission for Charlotte-Mecklenburg (perpetration/victimization) for Cure Violence clients 3 mo CMPD trauma-related injury 6- and Police Department post–program enrollment MCPH 12 mo post–program Defined as: enrollment Summation of Reduced repeat (a) Violent gun crime victim person records victimization/perpetration (b) Violent gun crime arrest records (c) Lesser violent gun crime victim person records (d) Lesser violent gun crime arrest records [CMPD + Cure Violence/YAP] Reduced gun-related violence at the community level 12- and 18 mo post–site implementation Defined as: (1) Summation of (a) Violent gun crime counts [CMPD] (b) Lesser violent gun crime counts [CMPD] (2) Ratio of shots red calls for service vs assaults with a deadly weapon [CMPD] (3) Critical Incident Reports involving guns [MCPH] Reduced non–gun-related violence at the community level 12 and 18 mo post–site implementation Defined as: (1) Summation of (a) Violent crime counts that do not include gun crime [CMPD] (b) Lesser violent crime counts that do not include gun crime [CMPD] (2a) Ratio of violent or lesser violent arrests done in Charlotte of individuals who live within vs outside of the community [CMPD] (2b) Ratio of violent or lesser violent victims in Charlotte who live within vs outside of the community [CMPD] (3) Critical Incident Reports that do not involve guns [MCPH] (continues) January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S51 TABLE 2 Data Sharing Metrics and Sources (Continued) HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Secondary Impact Clinical utilization: Changes in ED Atrium Health Reduced exposure to violence for Cure Violence clients’ social Cure Violence/YAP outcome and inpatient visits 6 and MCPH network 18 mo post–site implementation CMPD 12 mo post–program City of Charlotte (1) For individuals related to Cure Violence clients, summation of MCPH enrollment (compared with 6 CRH (a) Violent crime victim person records and 12 months prior) by visit (b) Violent crime arrest records type (overall vs avoidable/ (c) Lesser violent crime victim person records unavoidable) (d) Lesser violent crime arrest records [CMPD + Cure # of CRH connections (by Violence/YAP] category, food, housing, (2) Critical Incident Reports for individuals related to Cure transit, health, etc) Violence clients [MCPH + Cure Violence/YAP] # of electronic referrals (by Change in community norms toward violence 18 mo post–site category, food, housing, implementation transit, health, etc) Pre/posttest survey on attitudes toward gun violence [Cure # of closed-loop electronic Violence/YAP] referrals (by category, food, # of Cure Violence clients’ exposure to non–gun-related housing, transit, health, etc) violence Summation of (a) Violent crime that do not include guns victim person records (b) Violent crime that do not include guns arrest records (c) Lesser violent crime that do not include guns victim person records (d) Lesser violent crime that do not include guns arrest records [CMPD + Cure Violence/YAP] Abbreviations: CBO, community-based organization; CMPD, Charlotte-Mecklenburg Police Department; CRH, community resource hub; ED, emergency department; EHR, electronic health record; EMS, emergency medical services; HAVI, Health Alliance for Violence Intervention; MCPH, Mecklenburg County Department of Public Health; YAP, Youth Advocate Program. Italic indicates organizational data sharing partner. S52 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing can result in a perceived high-risk, low-reward sce- experiences of structural racism within these systems nario that reduces motivation and stagnates the data (eg, the Tuskegee syphilis studies, forced hysterec- 25 31 sharing process. To address this motivational bar- tomies on Black women ). An implication of this rier, we revisited the purpose of our current work work is that individuals and organizations engaging within the context of the goals of the collaborative in cross-sector data sharing to address health dispari- and the larger cross-sector partnership-building pro- ties have a responsibility to ensure ethical protections cess. We also asked members to identify how their when using these particularly important and sensi- organization’s mission, goals, and data align with the tive data. Enacting this responsibility may include but collaborative. is not limited to the following: (1) engaging com- We adjusted the Wiehe framework to propose that munities represented in the data in data analysis, motivation for data sharing can exist for both the interpretation, and dissemination; (2) providing his- individual and the organization and different but torical context with these data in all data products; complementary solutions are needed to address a lack and (3) creating structures that allow the communities of motivation at these two levels. Notably, the pro- represented by these data to decide how these data are cess of conducting interviews with each stakeholder used. and sharing findings back with the Data Collaborative Our stakeholder engagement process was specifi- was one approach through which individual motiva- cally designed to leverage both the collective expertise tion was addressed and enhanced. We used the power of the group and the unique perspectives of the of the Data Collaborative to leverage cooperation and individual Data Collaborative members. Similar to enhance motivation at the organizational level. We participatory research approaches, stakeholder en- also generated organization-specific project activities gagement should promote active collaboration and and results to bolster organizational trust and partici- participation, foster co-learning, and encourage mu- pation. Early wins that are relevant and can be dissem- tual capacity-building. Through iterative cycles of inated form another motivational strategy. Although information gathering (interviews and literature re- the local IDS was identified as a solution to cross- view), synthesis, and collaborative solution-driven sector data sharing, our initial results also revealed discussion, each partner contributed its institutional that individuals’ understanding of and comfort with knowledge and expertise, resulting in a robust evalu- the local IDS varied and contributed to hesitancy with ation strategy with stronger partnership buy-in. Data data sharing through this mechanism. In response to Collaborative members participated as both intervie- these concerns, we hosted the ISC at an interactive wees who contributed the data and as experts who Data Collaborative meeting and we facilitated meet reviewed and designed solutions to barriers identi- and greet exchanges between the ISC and individual fied in the data. This approach allowed members organizations outside of the larger meeting to address to anonymously share concerns and questions about organization-specific concerns. We also asked organi- cross-sector data sharing that may have remained zations that were already sharing data with the ISC unshared in group settings or when members only to describe their experience. As IDSs continue to be operated as experts. Thus, an intentional stakeholder used to support cross-sector data sharing efforts, the engagement process as described here fosters trust and partnership-building process between IDSs and the buy-in through a collective navigation of barriers and data partners should be a central component of that motivations. work. Data sharing moves at the speed of trust, and Experts have called the implications of cross-sector this trust must be built among the cross-sector part- data sharing “staggering” and a solution to “un- ners as well as between each organization and the IDS. doubtedly the biggest detriment to program growth Even with legal protections offered by the local and the least [client]-centered approach” to violence IDS, there is still a risk of identification and the po- prevention service delivery. Data silos inhibit the un- tential to contribute to existing stigmas about the derstanding of complex public health issues such as communities represented by the data being shared. community violence, as well as the design and evalu- This risk is particularly salient for communities of ation of solutions, and reduce the individual quality color, which, because of the intertwining relationship of care for those struggling with a variety of support between poverty and racism in the United States, are needs. As technology continues to advance and inte- disproportionately represented in data related to ex- grated platforms become the standard across health posure to violence and other health disparities. Data care and community settings, increasing the capac- Collaborative members were acutely aware of the ity to share data and join disparate data sources sometimes tenuous relationship between the individu- will broaden the scope of work and collective im- als they work with and the systems being represented, pact across sectors. While our work will be used to in part, because of individuals’ and communities’ improve community violence prevention in the local January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S53 13. Benns M, Ruther M, Nash N, Bozeman M, Harbrecht B, Miller K. Implications for Policy & Practice The impact of historical racism on modern gun violence: redlining in the city of Louisville, KY. Injury. 2020;51(10):2192-2198. ■ Data collaboratives can create a meaningful foundation for 14. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and cross-sector collaborations to address complex public health interventions. Lancet North Am Ed. 2017;389(10077):1453-1463. challenges and improve collective impact across sectors. 15. Thacker SB, Qualters JR, Lee LM; Centers for Disease Control and Prevention. Public health surveillance in the United States: ■ IDSs offer valuable solutions to common legal and technol- evolution and challenges. MMWR Surveill Summ. 2012;61(suppl): ogy barriers to cross-sector data sharing and address gaps 3-9. 16. Birkhead GS, Klompas M, Shah NR. Uses of electronic health in data capture across community organizations providing records for public health surveillance to advance public health. social and health care services. Annu Rev Public Health. 2015;36:345-359. 17. Fischer RL, Richter FG, Anthony E, Lalich N, Coulton C. Leverag- ■ A stakeholder engagement approach represents a significant ing administrative data to better serve children and families. Public way to foster individual and organizational motivation for Adm Rev. 2019;79(5):675-683. cross-sector data sharing, both of which are impacted by 18. Van Panhuis WG, Paul P, Emerson C, et al. A systematic review of barriers to data sharing in public health. BMC Public Health.2014; trust, equity, and power dynamics across systems. 14(1):1-9. 19. Culhane D, Fantuzzo J, Hill M, Burnett TC. Maximizing the use of integrated data systems: understanding the challenges and ad- vancing solutions. Ann Am Acad Pol Soc Sci. 2018;675(1):221- community, it can also be replicated and scaled across 20. US Census Bureau. Quick facts Mecklenburg County, other cross-sector collaborations seeking to influence North Carolina. https://www.census.gov/quickfacts/fact/table/ social and health inequities stemming from structural mecklenburgcountynorthcarolina/PST045219#qf-headnote-a. Accessed April 2, 2021. racism. 21. Chetty R, Hendren N, Kline P, Saez E. Where is the land of oppor- tunity? The geography of intergenerational mobility in the United States. QJEcon. 2014;129(4):1553-1623. References 22. Mecklenburg County Government. Mecklenburg County 1. American Public Health Association. Violence Is a Public Health community health assessment. https://www.mecknc.gov/ Issue: Public Health Is Essential to Understanding and Treating HealthDepartment/HealthStatistics/Pages/Community-Health- Assessment.aspx. Accessed April 2, 2021. Violence in the U.S. Washington, DC: American Public Health 23. Cure Violence Global. http://cureviolence.org. Accessed June 14, Association; 2018. APHA Policy No. 20185. 2. Centers for Disease Control and Prevention. WISQARS. Lead- 24. The Health Alliance for Violence Intervention. https://www.thehavi. ing causes of death visualization tool. https://wisqars-viz.cdc.gov: org. Accessed June 14, 2021. 8006/lcd/home. Accessed June 14, 2021. 25. Wiehe SE, Rosenman MB, Chartash D, et al. A solutions-based 3. Grinshteyn E, Hemenway D. Violent death rates in the US com- approach to building data-sharing partnerships. EGEMS (Wash DC). pared to those of the other high-income countries, 2015. Prev Med. 2018;6(1):20. 2019;123:20-26. 26. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res 4. Rutherford A, Zwi AB, Grove NJ, Butchart A. Violence: a glossary. Psychol. 2006;3(2):77-101. J Epidemiol Community Health. 2007;61(8):676-680. 27. Schmit C, Kelly K, Bernstein J. Cross sector data sharing: necessity, 5. The National Child Traumatic Stress Network. Community vio- challenge, and hope. JLaw MedEthics. 2019;47(2)(suppl):83-86. lence. https://www.nctsn.org/what-is-child-trauma/trauma-types/ 28. Siegal R, Langhinrichsen-Rohling J, Herring MH. Advancing the community-violence#:∼:text=Common%20types%20of%20 framework for community-academic partnerships: the importance community%20violence,%2C%20spontaneous%20or%20 of equity, power and historical context, and the role of trauma- terrorist%20attacks. Accessed April 2, 2021. informed relationship-building. Poster presented at: Society for 6. Kao AM, Schlosser KA, Arnold MR, et al. Trauma recidivism and Community Research and Action; June 2021; online. mortality following violent injuries in young adults. JSurgRes. 29. Drahota AM, Meza RD, Brikho B, et al. Community-academic part- 2019;237:140-147. nerships: a systematic review of the state of the literature and 7. Gorman-Smith D, Tolan P. The role of exposure to community vi- recommendations for future research. Milbank Q. 2016;94(1):163- olence and developmental problems among inner-city youth. Dev Psychopathol. 1998;10(1):101-116. 30. Alsan M, Wanamaker M. Tuskegee and the health of Black men. Q 8. Fowler PJ, Tompsett CJ, Braciszewski JM, Jacques-Tiura AJ, Baltes BB. Community violence: a meta-analysis on the effect of expo- JEcon. 2018;133(1):407-455. sure and mental health outcomes of children and adolescents. Dev 31. Prather C, Fuller TR, Jeffries WL 4th, et al. Racism, African Ameri- Psychopathol. 2009;21(1):227-259. can women, and their sexual and reproductive health: a review of 9. Bailey ZD, Feldman JM, Bassett MT. How structural racism historical and contemporary evidence and implications for health works—racist policies as a root cause of US racial health inequities. equity. Health Equity. 2018;2(1):249-259. N Engl J Med. 2021;384:768-773. 32. Israel BA, Schulz AJ, Parker EA, Becker AB. Review of community- 10. Sampson RJ, Lauritsen JL. Violent victimization and offending: based research: assessing partnership approaches to improve individual-, situational-, and community-level risk factors. In: Un- public health. Ann Rev Public Health. 1998;19(1):173-202. 33. Aboutanos MB, Jordan A, Goldberg S, Foster R, Garland S. Bridg- derstanding and Preventing Violence. Volume 3: Social Influences . ing the gap: hospital community-based youth violence prevention Washington, DC: The National Academies Press; 1994:1-114. program—pitfalls and lessons learned. Curr Trauma Rep. 2017;3(2): 11. Farrington DP. Early prediction of violent and non-violent youthful 79-88. offending. Eur J Criminal Policy Res. 1997;5(2):51-66. 34. O’Neil S, Hoe E, Ward E, Goyal R. Data Across Sectors for Health 12. Woods LL. The Federal Home Loan Bank Board, redlining, and the Initiative: Systems Alignment to Enhance Cross-Sector Data Shar- national proliferation of racial lending discrimination, 1921-1950. J ing. Princeton, NJ: Mathematica; 2020. Urban History. 2012;38(6):1036-1059. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Public Health Management and Practice Wolters Kluwer Health

A Replicable, Solution-Focused Approach to Cross-Sector Data Sharing for Evaluation of Community Violence Prevention Programming

Download PDF
11 pages

Loading next page...
 
/lp/wolters-kluwer-health/a-replicable-solution-focused-approach-to-cross-sector-data-sharing-ohbkCvYwCB

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Wolters Kluwer Health
Copyright
© 2022 The Authors. Published by Wolters Kluwer Health, Inc.
ISSN
1078-4659
eISSN
1550-5022
DOI
10.1097/phh.0000000000001426
Publisher site
See Article on Publisher Site

Abstract

Practice Full Report A Replicable, Solution-Focused Approach to Cross-Sector Data Sharing for Evaluation of Community Violence Prevention Programming Carlene A. Mayfield, PhD, MPH; Rachel Siegal, BS; Melvin Herring, PhD, MSW; Tracie Campbell, MS, CHES; Catie L. Clark, PhD; Jennifer Langhinrichsen-Rohling, PhD ABSTRACT Context: Community violence is a public health problem that erodes social infrastructure. Structural racism contributes to the disparate concentration of violence in communities of color. In Mecklenburg County, North Carolina, increasing trends in community violence show racial and geographic disparities that emphasize the need for cross-sector, data-driven approaches to program and policy change. Cross-sector collaborations are challenged by data sharing barriers that hinder implementation. Program: In response to community advocacy, Mecklenburg County Public Health (MCPH) launched a Community Violence Prevention Plan with evidence-based programming. The Cure Violence (CV) model, a public health approach to disrupting violence through equitable resource provision, network building, and changing norms, was implemented at the community level. The Health Alliance for Violence Intervention (HAVI) model, a hospital-based screening and case management inter- vention for victims of violence, was implemented at Carolinas Medical Center in Charlotte, the region’s only level I trauma center. Methods: A data collaborative was created to optimize evaluation of CV and HAVI programs including MCPH, the city of Charlotte, Atrium Health, Charlotte-Mecklenburg Schools, Johnson C. Smith University, and the University of North Carolina Charlotte. A comprehensive approach to facilitate data sharing was designed with a focus on engaging stakeholders and generating solutions to commonly reported barriers. Structured interviews were used to inform a solution-focused strategy. Results: Stakeholders reported perceptions of their organization’s barriers and facilitators to cross-sector data sharing. Com- mon technology, legal, and governance barriers were addressed through partnership with a local integrated data system. Solutions for trust and motivational challenges were built into ongoing collaborative processes. Discussion: Data silos inhibit the understanding of complex public health issues such as community violence, along with the design and evaluation of collective impact efforts. This approach can be replicated and scaled to support cross-sector collaborations seeking to influence social and health inequities stemming from structural racism. KEY WORDS: barriers and facilitators, community violence, data sharing, researcher-practitioner partnerships partnership and on this effort, and Octavia Ramsey from Johnson C. Smith Author Affiliation: Department of Community Health, Atrium Health, University for supporting the interview coding process. Charlotte, North Carolina (Dr Mayfield); Department of Psychological This work was supported, in part, by the Data Across Sectors for Health Sciences, University of North Carolina Charlotte, Charlotte, North Carolina (Ms (DASH) Mentor 3.0 Program. Siegal and Dr Langhinrichsen-Rohling); Department of Social Work, Johnson The authors have indicated they have no potential conflicts of interest to C. Smith University, Charlotte, North Carolina (Dr Herring); Office of Violence disclose. Prevention, Mecklenburg County Public Health, Charlotte, North Carolina (Ms Campbell); and Mecklenburg County Criminal Justice Services, Charlotte, Supplemental digital content is available for this article. Direct URL citations North Carolina (Dr Clark). appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (http://www.JPHMP.com). A number of people have made significant contributions to the creation and continuous success of the Community Violence Prevention Data This is an open-access article distributed under the terms of the Collaborative. The authors acknowledge Rebecca Hefner, Chief Data Officer Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 for the City of Charlotte, and Donna Smith, Epidemiologist for Mecklenburg (CCBY-NC-ND), where it is permissible to download and share the work County Public Health, as co-chairs of the collaborative, and Dr Lori Thomas, provided it is properly cited. The work cannot be changed in any way or used Executive Director of the Institute for Social Capital, for their expert guidance commercially without permission from the journal. and continuous support throughout the evaluation design process. The Correspondence: Carlene A. Mayfield, PhD, MPH, Department of Community authors also express gratitude to other collaborative members, Elyse Health, Atrium Health, 4135 S Stream Blvd, Charlotte, NC 28217 Hamilton-Childres, Carole McKernan, Haita Toure, Monica Nguyen, Robert (carlene.mayfield@atriumhealth.org). Broughton, Kaiti Mrak, Stacey Butler, Lindsay Messinger, and Kelly Moriarty, Copyright © 2022 The Authors. Published by Wolters Kluwer Health, Inc. along with their DASH Mentors Robert Gradeck from the University of DOI: 10.1097/PHH.0000000000001426 Pittsburgh and Erikka Gilliam from Drexel University, for their feedback and January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S43 S44 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing Context and evaluation. While the benefits of data shar- ing are well supported, the additional burden of Violence is a recognized public health problem technological and legal infrastructure required to with long-term consequences for individuals and adhere with federal regulations is prohibitive for communities. Homicide is the third leading cause of 18 many organizations. Integrated Data Systems (IDSs) death in the United States among 15- to 34-year-olds, represent a network of partnerships to link admin- and the leading cause of death among Black males istrative data across governmental and nonprofit in the same age group. The overall rate of homi- organizations providing social and health care ser- cide in the United States is also 7.5 times higher than vices within a community. By providing a legal in comparable high-income countries. Community and data governance framework, IDSs facilitate data violence occurs when interpersonal violence, the in- sharing between organizations to improve the under- tentional use of physical force against another person standing of complex community issues beyond typical resulting in injury or death, is perpetuated by individ- service silos. The Institute for Social Capital (ISC) is uals not intimately related to the victim. Common a registered nonprofit IDS in Mecklenburg County, examples include physical assault, property crimes, North Carolina, that serves as a signature research ini- group fights, and public shootings. Survivors of vi- tiative of the University of North Carolina Charlotte’s olence are at a greater risk for repeated victimization, Urban Institute. The ISC integrates data from more with an estimated 11% to 44% of hospital trauma than 45 partners across local agencies, sectors, and admissions related to violence occurring among pa- programs to increase the community’s capacity for tients with prior admissions. Long-term exposure to data-informed decision making and advance research violence also has profoundly harmful effects on child that deepens understanding of complex community development and emotional well-being and erodes the issues. 7,8 social and economic infrastructure of communities. The objective of this article is to describe a replica- Episodes of violence are interconnected by struc- ble strategy for cross-sector data sharing to support tural racism through the policies and practices that the evaluation of community violence prevention contribute to the disparate concentration of violence efforts using the technological and governance infras- in communities of color. Individual risk factors for tructure of the ISC. Our solution-focused partnership violence such as low family income, high unemploy- engagement approach was informed by structured ment, and lack of available resources, along with interviews with organizational stakeholders that in- witnessing and experiencing violence, are perpetu- cluded a discussion of barriers and facilitators to ated by mutually reinforcing systems of structural in- cross-sector data sharing. Results were used to design equity and community disinvestment. Historical poli- a responsive strategy to facilitate cross-sector data cies such as redlining, the practice of denying credit sharing and collective engagement in the evaluation for home loans based on the location of property of violence prevention activities. within a minority or disadvantaged neighborhood, have enduring effects on contemporary racialized Program residential segregation, and the inequitable distribu- tion of social and health risk factors for community Setting violence. Consequently, solutions for community violence require cross-sector collaborations that ad- Mecklenburg County, North Carolina, encompass- dress racial discrimination within housing, education, ing the city of Charlotte, is the largest and most employment, health care, and criminal justice sys- diverse area in the Carolinas. As a “minority major- tems through program and policy change. Often ity”county, more than half of Mecklenburg’s residents communities face the simultaneous adoption of mul- are people of color, with non-Hispanic White res- tiple prevention efforts by disparate agents of change, idents accounting for 46% of the population in 20 21 making the study of program-specific impacts both 2019. A widely circulated study by Chetty et al necessary and challenging. ranked Charlotte last among America’s 50 largest cities in economic mobility, highlighting a cycle of in- Public health decision making is becoming in- tergenerational poverty, poor educational attainment, creasingly reliant on complex data sharing and inte- and health disparities that disproportionately impact gration, creating additional barriers for cross-sector the communities’ residents of color. Data from the collaborations. Advances in health information county’s community health assessments show increas- technology (IT) and informatics sciences, including ing trends in community violence with notable racial the widespread adoption of electronic health records and geographic disparities. Between June 2018 and by community health sites and health care sys- June 2019, the city of Charlotte reported a 119% tems, have enhanced the capacity for surveillance January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S45 increase in homicides, along with a higher incidence Intervention (HAVI) framework, at Atrium Health, in areas with concentrated social and health risk fac- CMC Charlotte. tors including poverty, low educational attainment, high unemployment, and worse access to health care Methods relative to the larger county. Atrium Health is one of the largest integrated, non- Violence Prevention Data Collaborative profit health systems in the southeast region with 37 The Violence Prevention Data Collaborative (Data hospitals and more than 1350 care locations across Collaborative) was formed to leverage cross-sector re- North Carolina, South Carolina, and Georgia. In ad- sources for data sharing, analysis, reporting, and eval- dition to being the largest health care provider for uation in support of activities aligned with the Com- historically marginalized and indignant patients in the munity Violence Prevention Plan. Partners included local community, Atrium Health provides the region’s multiple Mecklenburg County departments (Public only level I trauma center through the Carolinas Med- Health, Criminal Justice Services, Department of So- ical Center (CMC), Charlotte facility. Results from cial Services, and Community Support Services), city a recent (2019) publication show that a majority of of Charlotte (Innovation & Technology, Charlotte- patients admitted to CMC for violent trauma be- Mecklenburg Police Department), and institutional tween 2009 and 2015 were Black (64.9%) and male partners from health care (Atrium Health) and educa- (87.4%) and that 24.9% of patients without mor- tion systems (Charlotte-Mecklenburg Schools, John- tality were readmitted with a second violent injury son C. Smith University, and University of North within an average of 31.9 ± 21.0 months. Vio- Carolina Charlotte). Each partner is represented by at lent injury was also independently associated with least one organizational member. Members reviewed higher morbidity and mortality related to the in- and signed a nonlegally binding Memorandum of dex trauma and subsequent injuries among otherwise Understanding (MOU) that details the terms of healthy young adults. Disparities in violent injury are collaboration including governance and leadership, a symptom of long-standing systemic inequity in the collaborative charge and priorities, and membership larger community. Thus, equitable solutions require a expectations. comprehensive approach to intervention and sec- ondary prevention that is informed by the evaluation of cross-sector data. The Institute for Social Capital The ISC uses individual-level identifiable data so that information can be linked across different sources. Community Violence Prevention Plan Because of the sensitive nature of personally identi- Community stakeholders responded to these findings fiable information, the ISC has numerous policies and by mobilizing partnerships and aligning resources to procedures in place to protect the data including le- support a comprehensive approach to community gal, technical, procedural, and physical protections. violence prevention. In 2019, the city of Charlotte The ISC establishes data sharing agreements with each launched SAFE Charlotte, a comprehensive approach data depositor that are aligned with all federal and in- to improve police/community relations, reduce violent stitutional regulations. To obtain data from the ISC, crime, and create opportunities for social mobility. At requestors must complete a data license request that the same time, Mecklenburg County Public Health details the evaluation project, requested data fields, launched a Community Violence Prevention Plan, the methodology and analyses that will be used, and including the creation of the Office of Violence Pre- the dissemination plan. Data requests are reviewed vention and the implementation of 2 new programs by the ISC Data and Research Oversight Commit- using evidence-based models for community violence tee (DAROC), which oversees the use of data for prevention. First, and in partnership with the city, the ISC Board of Directors and includes university the Cure Violence (CV) violence interrupter model, researchers, members of the community, and a rep- a data-driven, research-based, community-centric ap- resentative from each data depositor. The agency that proach to violence prevention, was implemented at owns the data being requested (ie, the depositor) must the community level by a local nonprofit agency, approve the use of their data during the approval pro- Youth Advocate Program (YAP), with priority given cess. If a request is approved, the requestor will sign a to neighborhoods experiencing the highest rates of vi- Limited License Agreement (LLA) outlining how the olence. Second, a hospital-based screening and case data can and cannot be used. After the LLA is com- management intervention was implemented at the in- plete and requested data are integrated, the ISC and dividual level using the Health Alliance for Violence DAROC members will review the data set prior to S46 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing release to ensure it meets the terms of the LLA and The researchers used deductive thematic analysis, does not inadvertently identify any individual. Upon informed by the 6 Wiehe categories, to code the data approval, de-identified data are securely transferred to for barriers and solutions to cross-sector data shar- the requestor. ing. For example, participants’ reference to a lack of personnel to support data sharing was coded as an economic barrier. Following initial coding, the Stakeholder engagement researchers revised the definitions and included ad- We developed a solution-focused approach to engage ditional examples for each of the 6 Wiehe categories stakeholders in cross-sector data sharing to support to better capture the breadth of participant responses. evaluation of the new CV (community-based) and The most significant change was to expand the defini- HAVI (hospital-based) interventions. This process in- tion of motivational barriers to include both barriers cluded the design and execution of an integrated for individuals charged with data sharing and moti- data request though the ISC governance process. A vational barriers for the data holding organization. conceptual framework is summarized in the Figure. Throughout the initial coding and coding refinement, First, collaborative members representing the 2 uni- the coders discussed any disagreements until a con- versities (the researchers) conducted semistructured sensus was reached. However, given that stakeholders interviews with the other data sharing stakehold- specifically responded to queries about each of the 6 ers. The researchers developed the interview guide categories of barriers in the interview, coding disagree- using an existing data sharing barriers framework, ments were rare. Instead, discussion typically focused which describes 6 categories of barriers: technical, on revising the Wiehe framework to capture barriers motivational, economic, political, legal, and ethical. occurring at multiple levels (individual, collaborative, During the interview, the researchers asked a general organizational). question about participants’ barriers to data sharing: Following coding, participants provided feedback “From your perspective, what are the main barriers to on the identified themes in an extended Data Col- cross-sector data sharing in your organization?” The laborative meeting and as part of a member check. researchers then asked participants a set of 6 ques- No additional themes were added or removed fol- tions, each of which focused on a different barrier lowing the member check. Participants also discussed from the Wiehe framework (eg, Motivational bar- potential solutions to the identified barriers. Solutions riers to data sharing can include a lack of incentives that could be implemented immediately were identi- or opportunity cost disincentives or disagreement on fied by the Data Collaborative as next steps in the data data use. What do motivational barriers look like at sharing process. The stakeholder interview protocol your organization?). for this process was reviewed and approved by the FIGURE Conceptual Framework Abbreviations: DAROC, Data and Research Oversight Committee; LLA, Limited License Agreement. January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S47 institutional review board at University of North Car- Content Table 1, available at http://links.lww.com/ olina Charlotte (#21-0379). JPHMP/A857, and Supplemental Digital Content Second, a literature review was conducted to iden- Table 2, available at http://links.lww.com/JPHMP/ tify published studies from programs designed under A858) were presented to the Data Collaborative each of the evidence-based frameworks. Studies with in a series of three 90-minute meetings that oc- design and implementation similar to the Commu- curred bimonthly. Members brainstormed solutions nity Violence Prevention Plan were prioritized. For to data sharing barriers that were raised (outlined in HAVI-based programs, we included studies that were Table 1) and identified evaluation priorities for each located at level I trauma centers, excluded victims of of the evidence-based programs. For the HAVI pro- domestic violence, sexual assault, and child abuse, and gram, the Data Collaborative wanted to answer the utilized some form of case management and commu- following: (i) Did the program serve the intended pop- nity resource referral as the primary intervention. For ulation? (ii) What community resources/community CV-based programs, we included studies that were lo- partners were most frequently used by participants? cated in US cities, implemented interventions at the (iii) What impact did the program have on injury individual and community levels, and relied on some recidivism, clinical utilization, and repeat victimiza- form of outreach workers, violence interrupters, or a tion/perpetration? For the CV program, the Data combination of the two. Collaborative wanted to answer the following: (i) Third, we conducted a series of expert review meet- Did the program serve the intended population? ings with Data Collaborative members to synthesize (ii) Who referred clients to the program? (iii) Was results from the previous steps (stakeholder interviews the program successfully implemented? (iv) What and the literature review) to define evaluation ques- resources/community partners were clients most fre- tions and align metrics, data sources, and elds fi needed quently referred to and what resources/community for data sharing. Specifically, results from the stake- partners did clients most frequently use? Metrics and holder interviews were used to inform the process of data sources required to answer these questions were achieving data sharing and data dissemination while identified and are summarized in Table 2. the literature review was used to inform the metrics, and subsequent data fields, identified for data sharing. Discussion We designed a solution-focused approach to achieve Results cross-sector data sharing that was informed by A total of 10 members representing 8 distinct orga- key stakeholders and built on the legal and gover- nizations were interviewed, comprising 100% par- nance infrastructure of the ISC, a local IDS. The ticipation from the Data Collaborative members. stakeholder interviews contributed several important Preliminary analyses identified a range of barriers to learning objectives for the Data Collaborative. First, data sharing across technical, legal, ethical, and moti- even organizations with sustainable financing that vational domains. Participants expressed inconsistent are motivated to engage in data sharing may ex- perspectives on the benefits of cross-sector data shar- perience different levels of organizational readiness ing, along with potential concerns about the legal and for data sharing. Some of these organizations re- governance protections available through partnership ported a more informal governance structure, which with the ISC. Some participants felt that their individ- may impact their readiness. In response to this ual and organizational roles in the Data Collaborative finding, members of the collaborative at organiza- were unclear and that the current work lacked a tions with more formal governance structures are strong rationale for sharing sensitive information supporting these organizations’ governance structure and assuming risk. Motivation for data sharing was development process. Second, individuals and orga- aligned with individual members’ position within nizations engaging in cross-sector data sharing often their organization, in addition to their perception of have previous data sharing experiences that include broader structural and leadership support. Finally, experiences of ongoing disempowerment that impact preliminary findings suggested that individuals’ and their current motivation to engage in data sharing. organizations’ different positions (eg, in relation to To address this barrier, we conducted organization-led other collaborative members or in relation to other or- presentations to the collaborative about the organi- ganizations), and individuals’ ability to leverage those zation’s data and data governance practices, previous positions, had the potential to impact data sharing experiences of data sharing, and ethical and legal efforts. frameworks. These presentations provided an avenue Stakeholder interview results and key studies for members to share their experiences and existing from the literature review (see Supplemental Digital concerns. S48 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing TABLE 1 Stakeholder Interview Results and Solutions to Common Data Sharing Barriers Category and Definition Quote Solutions Technical “One challenge is that we have organizations Initiate discussions with ISC and Data Data are not collected, preserved, using different tools and keeping their data Collaborative or available in a usable format; in different environments. Most of our Use 1:1 and group discussions to identify needs technical solutions are not existing technical resources are directed and (re)allocate resources available; a lack of metadata and toward keeping our environment secure; we standards might have to reallocate resources to share data effectively.” Motivational “So I don’t know that the missions changed. Define VPDC mission Mission/Values & Organizational Um, I just, I guess I wasn’t privy to all the Connect evaluation strategy design back to Buy-In things that are being done. And the things I mission Lack of incentives or opportunity was privy to, to me didn’t seem on mission. Leverage Violence Prevention leadership team cost disincentives, disagreement At the time.” to build organizational support on data use Motivational “I don’t have final decision power, but I have Leverage Violence Prevention leadership team Individual Position the ability to influence um higher level to support VPDC members Lack of incentives or opportunity leadership in terms of making Leverage VPDC members’ influence to support cost disincentives, disagreement recommendations about what we should other members on data use prioritize. And I have the ability to like escalate, specific projects up to leadership in a way that is very facilitative for uh those approval processes.” Economic “I think the main one is resources is that we Initiate discussions with ISC and Data Possible economic damage and/or don’t have a dedicated data position. So, um, Collaborative lack of resources for data sharing when we, when we share data, it involves Use 1:1 and group discussions to identify needs being a little piece of a bunch of people’s and (re)allocate resources jobs. And it’sjustait’s not, it would be ideal Leverage Violence Prevention leadership team if we had a dedicated kind of research and to increase resources in the long term data position.” Political “But when, particularly in the current climate, Initiate discussions with ISC and Data Lack of trust and guidelines, when we see data get used in a way that is Collaborative restrictive policies misleading.” Use evaluation strategy design process to collaboratively define evaluation questions and metrics Legal “If there were a data request that required Initiate discussions with ISC and Data Ownership and copyright, individually identifying information, that Collaborative protection would be a limitation. Especially I mean, it Use evaluation strategy design process to would go beyond PHI restrictions.” collaboratively define evaluation questions and metrics, build consensus on risk and reward Ethical “And so we wouldn’t want to, you know, be Initiate discussions with ISC and Data Lack of proportionality and flagging them as like, oh, [we] are Collaborative reciprocity supervising all of these dangerous people?” Use evaluation strategy design process to collaboratively define evaluation questions and metrics, build consensus on risk and reward Abbreviations: ISC, Institute for Social Capital; PHI, protected health information; VPDC, Violence Prevention Data Collaborative. a 25 Categories and definitions adapted from the Wiehe et al solution-based framework for data sharing partnerships. Indicates a change to the original definition. Although the Wiehe framework identifies motiva- the collaborative, and in particular organizations. At tion as one of 6 equivalent barriers to data sharing, the same time, some respondents struggled to under- stakeholders described motivation as a barrier that stand and articulate how the mission and goals of the underlies the other barriers. Lack of motivation, in Data Collaborative aligned with their organization’s conjunction with concerns about sharing sensitive purpose for participation in the Community Violence data collected from vulnerable populations, mani- Prevention Plan. Lacking a clear mission and concrete fested at the level of individual IT stakeholders, within goals is common in cross-sector partnerships and January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S49 TABLE 2 Data Sharing Metrics and Sources HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Demographics Process Age, race/ethnicity, length of Atrium Health Age, race/ethnicity, geographic location, time in program Cure Violence/YAP stay, EMS transport EHR/Data Warehouse Time in program is defined as the number of weeks the client continues to meet with the outreach worker Implementation Process #/% patients meeting selection Atrium Health #/% clients meeting selection criteria screened for program Cure Violence/YAP criteria screened for the EHR/Data Warehouse Selection criteria: Individuals aged 16-24 y, previous exposure program Case notes to gun violence, live in intervention area #/% eligible patients enrolled in #/% eligible clients enrolled in the program the program # VI mediations # people involved in mediation Retention Process Successful case management Atrium Health Successful client management defined as appropriate Cure Violence/YAP defined as follow-up for 30, 60, Case notes closed-loop referral in response to client-identified need 90 d binary (yes/no) #/% successful out of eligible, enrolled clients #/% successful out of eligible, enrolled populations Partnership Process # of CBO partners that own and Atrium Health # of city/county partners using CRH Cure Violence/YAP engagement manage their program listings CRH # of CBO partners that own and manage their program listings CRH on the CRH on the CRH # of CBOs and partners accepting # of partners accepting closed-loop referrals closed-loop referrals # partners (co-) hosting CVG community events Program alignment Process #/% of eligible patients referred Atrium Health #/% clients referred to from HAVI Cure Violence/YAP to CVH Cure Violence Atrium Health CRH CRH (continues) S50 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing fi TABLE 2 Data Sharing Metrics and Sources (Continued) HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Primary outcome Impact Injury recidivism: Hospital Atrium Health Reduced exposure to gun-related violence Cure Violence/YAP readmission for Charlotte-Mecklenburg (perpetration/victimization) for Cure Violence clients 3 mo CMPD trauma-related injury 6- and Police Department post–program enrollment MCPH 12 mo post–program Defined as: enrollment Summation of Reduced repeat (a) Violent gun crime victim person records victimization/perpetration (b) Violent gun crime arrest records (c) Lesser violent gun crime victim person records (d) Lesser violent gun crime arrest records [CMPD + Cure Violence/YAP] Reduced gun-related violence at the community level 12- and 18 mo post–site implementation Defined as: (1) Summation of (a) Violent gun crime counts [CMPD] (b) Lesser violent gun crime counts [CMPD] (2) Ratio of shots red calls for service vs assaults with a deadly weapon [CMPD] (3) Critical Incident Reports involving guns [MCPH] Reduced non–gun-related violence at the community level 12 and 18 mo post–site implementation Defined as: (1) Summation of (a) Violent crime counts that do not include gun crime [CMPD] (b) Lesser violent crime counts that do not include gun crime [CMPD] (2a) Ratio of violent or lesser violent arrests done in Charlotte of individuals who live within vs outside of the community [CMPD] (2b) Ratio of violent or lesser violent victims in Charlotte who live within vs outside of the community [CMPD] (3) Critical Incident Reports that do not involve guns [MCPH] (continues) January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S51 TABLE 2 Data Sharing Metrics and Sources (Continued) HAVI Model CVH Model Construct Type Metric Data Source Metric Data Source Secondary Impact Clinical utilization: Changes in ED Atrium Health Reduced exposure to violence for Cure Violence clients’ social Cure Violence/YAP outcome and inpatient visits 6 and MCPH network 18 mo post–site implementation CMPD 12 mo post–program City of Charlotte (1) For individuals related to Cure Violence clients, summation of MCPH enrollment (compared with 6 CRH (a) Violent crime victim person records and 12 months prior) by visit (b) Violent crime arrest records type (overall vs avoidable/ (c) Lesser violent crime victim person records unavoidable) (d) Lesser violent crime arrest records [CMPD + Cure # of CRH connections (by Violence/YAP] category, food, housing, (2) Critical Incident Reports for individuals related to Cure transit, health, etc) Violence clients [MCPH + Cure Violence/YAP] # of electronic referrals (by Change in community norms toward violence 18 mo post–site category, food, housing, implementation transit, health, etc) Pre/posttest survey on attitudes toward gun violence [Cure # of closed-loop electronic Violence/YAP] referrals (by category, food, # of Cure Violence clients’ exposure to non–gun-related housing, transit, health, etc) violence Summation of (a) Violent crime that do not include guns victim person records (b) Violent crime that do not include guns arrest records (c) Lesser violent crime that do not include guns victim person records (d) Lesser violent crime that do not include guns arrest records [CMPD + Cure Violence/YAP] Abbreviations: CBO, community-based organization; CMPD, Charlotte-Mecklenburg Police Department; CRH, community resource hub; ED, emergency department; EHR, electronic health record; EMS, emergency medical services; HAVI, Health Alliance for Violence Intervention; MCPH, Mecklenburg County Department of Public Health; YAP, Youth Advocate Program. Italic indicates organizational data sharing partner. S52 Mayfield, et al • 28(1 Supp), S43–S53 A Replicable Approach to Cross-Sector Data Sharing can result in a perceived high-risk, low-reward sce- experiences of structural racism within these systems nario that reduces motivation and stagnates the data (eg, the Tuskegee syphilis studies, forced hysterec- 25 31 sharing process. To address this motivational bar- tomies on Black women ). An implication of this rier, we revisited the purpose of our current work work is that individuals and organizations engaging within the context of the goals of the collaborative in cross-sector data sharing to address health dispari- and the larger cross-sector partnership-building pro- ties have a responsibility to ensure ethical protections cess. We also asked members to identify how their when using these particularly important and sensi- organization’s mission, goals, and data align with the tive data. Enacting this responsibility may include but collaborative. is not limited to the following: (1) engaging com- We adjusted the Wiehe framework to propose that munities represented in the data in data analysis, motivation for data sharing can exist for both the interpretation, and dissemination; (2) providing his- individual and the organization and different but torical context with these data in all data products; complementary solutions are needed to address a lack and (3) creating structures that allow the communities of motivation at these two levels. Notably, the pro- represented by these data to decide how these data are cess of conducting interviews with each stakeholder used. and sharing findings back with the Data Collaborative Our stakeholder engagement process was specifi- was one approach through which individual motiva- cally designed to leverage both the collective expertise tion was addressed and enhanced. We used the power of the group and the unique perspectives of the of the Data Collaborative to leverage cooperation and individual Data Collaborative members. Similar to enhance motivation at the organizational level. We participatory research approaches, stakeholder en- also generated organization-specific project activities gagement should promote active collaboration and and results to bolster organizational trust and partici- participation, foster co-learning, and encourage mu- pation. Early wins that are relevant and can be dissem- tual capacity-building. Through iterative cycles of inated form another motivational strategy. Although information gathering (interviews and literature re- the local IDS was identified as a solution to cross- view), synthesis, and collaborative solution-driven sector data sharing, our initial results also revealed discussion, each partner contributed its institutional that individuals’ understanding of and comfort with knowledge and expertise, resulting in a robust evalu- the local IDS varied and contributed to hesitancy with ation strategy with stronger partnership buy-in. Data data sharing through this mechanism. In response to Collaborative members participated as both intervie- these concerns, we hosted the ISC at an interactive wees who contributed the data and as experts who Data Collaborative meeting and we facilitated meet reviewed and designed solutions to barriers identi- and greet exchanges between the ISC and individual fied in the data. This approach allowed members organizations outside of the larger meeting to address to anonymously share concerns and questions about organization-specific concerns. We also asked organi- cross-sector data sharing that may have remained zations that were already sharing data with the ISC unshared in group settings or when members only to describe their experience. As IDSs continue to be operated as experts. Thus, an intentional stakeholder used to support cross-sector data sharing efforts, the engagement process as described here fosters trust and partnership-building process between IDSs and the buy-in through a collective navigation of barriers and data partners should be a central component of that motivations. work. Data sharing moves at the speed of trust, and Experts have called the implications of cross-sector this trust must be built among the cross-sector part- data sharing “staggering” and a solution to “un- ners as well as between each organization and the IDS. doubtedly the biggest detriment to program growth Even with legal protections offered by the local and the least [client]-centered approach” to violence IDS, there is still a risk of identification and the po- prevention service delivery. Data silos inhibit the un- tential to contribute to existing stigmas about the derstanding of complex public health issues such as communities represented by the data being shared. community violence, as well as the design and evalu- This risk is particularly salient for communities of ation of solutions, and reduce the individual quality color, which, because of the intertwining relationship of care for those struggling with a variety of support between poverty and racism in the United States, are needs. As technology continues to advance and inte- disproportionately represented in data related to ex- grated platforms become the standard across health posure to violence and other health disparities. Data care and community settings, increasing the capac- Collaborative members were acutely aware of the ity to share data and join disparate data sources sometimes tenuous relationship between the individu- will broaden the scope of work and collective im- als they work with and the systems being represented, pact across sectors. While our work will be used to in part, because of individuals’ and communities’ improve community violence prevention in the local January/February 2022 • Volume 28, Number 1 Supp www.JPHMP.com S53 13. Benns M, Ruther M, Nash N, Bozeman M, Harbrecht B, Miller K. Implications for Policy & Practice The impact of historical racism on modern gun violence: redlining in the city of Louisville, KY. Injury. 2020;51(10):2192-2198. ■ Data collaboratives can create a meaningful foundation for 14. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, Bassett MT. Structural racism and health inequities in the USA: evidence and cross-sector collaborations to address complex public health interventions. Lancet North Am Ed. 2017;389(10077):1453-1463. challenges and improve collective impact across sectors. 15. Thacker SB, Qualters JR, Lee LM; Centers for Disease Control and Prevention. Public health surveillance in the United States: ■ IDSs offer valuable solutions to common legal and technol- evolution and challenges. MMWR Surveill Summ. 2012;61(suppl): ogy barriers to cross-sector data sharing and address gaps 3-9. 16. Birkhead GS, Klompas M, Shah NR. Uses of electronic health in data capture across community organizations providing records for public health surveillance to advance public health. social and health care services. Annu Rev Public Health. 2015;36:345-359. 17. Fischer RL, Richter FG, Anthony E, Lalich N, Coulton C. Leverag- ■ A stakeholder engagement approach represents a significant ing administrative data to better serve children and families. Public way to foster individual and organizational motivation for Adm Rev. 2019;79(5):675-683. cross-sector data sharing, both of which are impacted by 18. Van Panhuis WG, Paul P, Emerson C, et al. A systematic review of barriers to data sharing in public health. BMC Public Health.2014; trust, equity, and power dynamics across systems. 14(1):1-9. 19. Culhane D, Fantuzzo J, Hill M, Burnett TC. Maximizing the use of integrated data systems: understanding the challenges and ad- vancing solutions. Ann Am Acad Pol Soc Sci. 2018;675(1):221- community, it can also be replicated and scaled across 20. US Census Bureau. Quick facts Mecklenburg County, other cross-sector collaborations seeking to influence North Carolina. https://www.census.gov/quickfacts/fact/table/ social and health inequities stemming from structural mecklenburgcountynorthcarolina/PST045219#qf-headnote-a. Accessed April 2, 2021. racism. 21. Chetty R, Hendren N, Kline P, Saez E. Where is the land of oppor- tunity? The geography of intergenerational mobility in the United States. QJEcon. 2014;129(4):1553-1623. References 22. Mecklenburg County Government. Mecklenburg County 1. American Public Health Association. Violence Is a Public Health community health assessment. https://www.mecknc.gov/ Issue: Public Health Is Essential to Understanding and Treating HealthDepartment/HealthStatistics/Pages/Community-Health- Assessment.aspx. Accessed April 2, 2021. Violence in the U.S. Washington, DC: American Public Health 23. Cure Violence Global. http://cureviolence.org. Accessed June 14, Association; 2018. APHA Policy No. 20185. 2. Centers for Disease Control and Prevention. WISQARS. Lead- 24. The Health Alliance for Violence Intervention. https://www.thehavi. ing causes of death visualization tool. https://wisqars-viz.cdc.gov: org. Accessed June 14, 2021. 8006/lcd/home. Accessed June 14, 2021. 25. Wiehe SE, Rosenman MB, Chartash D, et al. A solutions-based 3. Grinshteyn E, Hemenway D. Violent death rates in the US com- approach to building data-sharing partnerships. EGEMS (Wash DC). pared to those of the other high-income countries, 2015. Prev Med. 2018;6(1):20. 2019;123:20-26. 26. Braun V, Clarke V. Using thematic analysis in psychology. Qual Res 4. Rutherford A, Zwi AB, Grove NJ, Butchart A. Violence: a glossary. Psychol. 2006;3(2):77-101. J Epidemiol Community Health. 2007;61(8):676-680. 27. Schmit C, Kelly K, Bernstein J. Cross sector data sharing: necessity, 5. The National Child Traumatic Stress Network. Community vio- challenge, and hope. JLaw MedEthics. 2019;47(2)(suppl):83-86. lence. https://www.nctsn.org/what-is-child-trauma/trauma-types/ 28. Siegal R, Langhinrichsen-Rohling J, Herring MH. Advancing the community-violence#:∼:text=Common%20types%20of%20 framework for community-academic partnerships: the importance community%20violence,%2C%20spontaneous%20or%20 of equity, power and historical context, and the role of trauma- terrorist%20attacks. Accessed April 2, 2021. informed relationship-building. Poster presented at: Society for 6. Kao AM, Schlosser KA, Arnold MR, et al. Trauma recidivism and Community Research and Action; June 2021; online. mortality following violent injuries in young adults. JSurgRes. 29. Drahota AM, Meza RD, Brikho B, et al. Community-academic part- 2019;237:140-147. nerships: a systematic review of the state of the literature and 7. Gorman-Smith D, Tolan P. The role of exposure to community vi- recommendations for future research. Milbank Q. 2016;94(1):163- olence and developmental problems among inner-city youth. Dev Psychopathol. 1998;10(1):101-116. 30. Alsan M, Wanamaker M. Tuskegee and the health of Black men. Q 8. Fowler PJ, Tompsett CJ, Braciszewski JM, Jacques-Tiura AJ, Baltes BB. Community violence: a meta-analysis on the effect of expo- JEcon. 2018;133(1):407-455. sure and mental health outcomes of children and adolescents. Dev 31. Prather C, Fuller TR, Jeffries WL 4th, et al. Racism, African Ameri- Psychopathol. 2009;21(1):227-259. can women, and their sexual and reproductive health: a review of 9. Bailey ZD, Feldman JM, Bassett MT. How structural racism historical and contemporary evidence and implications for health works—racist policies as a root cause of US racial health inequities. equity. Health Equity. 2018;2(1):249-259. N Engl J Med. 2021;384:768-773. 32. Israel BA, Schulz AJ, Parker EA, Becker AB. Review of community- 10. Sampson RJ, Lauritsen JL. Violent victimization and offending: based research: assessing partnership approaches to improve individual-, situational-, and community-level risk factors. In: Un- public health. Ann Rev Public Health. 1998;19(1):173-202. 33. Aboutanos MB, Jordan A, Goldberg S, Foster R, Garland S. Bridg- derstanding and Preventing Violence. Volume 3: Social Influences . ing the gap: hospital community-based youth violence prevention Washington, DC: The National Academies Press; 1994:1-114. program—pitfalls and lessons learned. Curr Trauma Rep. 2017;3(2): 11. Farrington DP. Early prediction of violent and non-violent youthful 79-88. offending. Eur J Criminal Policy Res. 1997;5(2):51-66. 34. O’Neil S, Hoe E, Ward E, Goyal R. Data Across Sectors for Health 12. Woods LL. The Federal Home Loan Bank Board, redlining, and the Initiative: Systems Alignment to Enhance Cross-Sector Data Shar- national proliferation of racial lending discrimination, 1921-1950. J ing. Princeton, NJ: Mathematica; 2020. Urban History. 2012;38(6):1036-1059.

Journal

Journal of Public Health Management and Practice – Wolters Kluwer Health

Published: Jan 1, 2022

There are no references for this article.