An Examination of the Complexity of Power and Race When a Predominately White Institution partners with Black Communities to Address Infant Mortality

Evolution of the LIHF Initiative

The power dynamics described below are best understood with some context about the Initiative’s origins and how it evolved over its roughly 10-year time span. LIHF grew out of the recognition of and sense of urgency to address the huge disparities in Black infant mortality in Wisconsin. The Dean of the University of Wisconsin School of Medicine and Public Health (UW SMPH) championed the Initiative and directed funds from the WPP endowment administered within the school toward this effort. These funds were endowed to the SMPH by an agreement that allowed the conversion of a nonprofit health insurance company (Blue Cross-Blue Shield) to a for-profit corporation (Kittle, 2018).

The WPP sponsored and managed LIHF, which included funding of community-based planning, the funding of 23 specific project grants, and local collaboratives in four communities (Beloit, Kenosha, Milwaukee, and Racine). During the planning phase, there were contracts with community agencies to convene local partners to plan and develop Community Action Plans (CAPs). WPP leadership emphasized that CAPs should focus on approaches that address social and structural determinants of health and health disparities in Black-white birth outcomes, an approach championed most notably by Dr. Michael Lu and other maternal and child health epidemiologists and providers (Lu et al., 2010). After the CAP planning phase, WPP funded and supported community collaboratives to convene with local partners to spread awareness of Black infant mortality, develop and implement strategies for communities to address the disparities in birth outcomes, and to identify and address policies, systems, and environments that were contributing to the disparity. The concept of collective impact (Kania & Kramer, 2011) was introduced to guide interactions between LIHF collaboratives and other community public health initiatives.

The project grants funded community organizations with the expectation that they would implement specific recommendations identified by the CAPs in consultation with an academic partner. The project grant recipients were expected to participate in their local community collaboratives but had no direct reporting relationship or responsibility to the collaboratives. Additionally, the WPP established a LIHF Program Office to provide technical assistance and monitor the LIHF collaboratives, individual evaluation plans, activities, and regional strategies. The LIHF model is further described in Frey et al. (2014); Cotton et al. (2019) provides a structural analysis and case study of the implementation of LIHF.

Power Dynamics of Race and Racism

This article focuses on how key participants viewed power dynamics of the initiative, comparing the perceptions of Black and non-Black stakeholders. The juxtaposition of Black voices to non-Black voices is central and Black stakeholder voices are centered since the initiative is focused on the health of the Black community. Most stakeholders posited that the unequal power dynamics pertaining to race and racism created challenges within the Initiative. More specifically, Black stakeholders in survey and interviews stated that non-Black stakeholders within this initiative failed to address the role that they played in racism, health disparities, and the power dynamics throughout the project.

Racial dynamics and positionality

Black stakeholders throughout this initiative encouraged the WPP and the LIHF Program Office to hire a Black program evaluator familiar with the racial and power dynamics between Madison and Milwaukee, Racine, and Kenosha and who had experience conducting culturally responsive evaluations (Hood et al., 2015). For most of the duration of the Initiatives, the evaluation team was white, with one Asian evaluator for a period of time. Taking a step to address this lack of representation, in April of 2018 the Population Health Institute hired Troy M. Williams, a Black program evaluator who had experience conducting research for community-based organizations but who was unfamiliar with the regionally unique power and racial dynamics that many of the stakeholders described.

All of the authors were evaluators working at an academic institute at a large research university located one to two hours’ drive from the LIFH communities. The LIHF lead evaluator (AG) is a mid-career white woman with experience evaluating community-based programs and researching racial and ethnic disparities in health. The LIHF Evaluation’s Principal Investigator (DPM), a white man, had over 40 years of experience evaluating community-based programs through both quantitative and qualitative methods. Sadly, Dr. Moberg passed away after this manuscript was submitted. A supporting author and evaluator, AWF, is a white woman with academic training in applied anthropology and public health and experience in community education settings.

Evaluation Design

This evaluation used a mixed-methods approach with an explanatory design that incorporated multiple stakeholder perspectives to better explain results from surveys and other quantitative data (Creswell & Plano Clark, 2007). A web-based survey of LIHF Stakeholders was conducted in the summer of 2018 to obtain broad structured feedback regarding the Initiative. This article draws on data from the quantitative and qualitative responses from the online stakeholder survey (n = 132) and follow up interviews with key informants (n = 25).

Sample and Data Collection

The online stakeholder survey was developed by the evaluation team in consultation with the LIHF evaluation working group. The working group included representatives from the LIHF Program Office (majority white, academic), the collaborative convening agencies (majority Black, non-profit organizations and community), and the funding agency (majority white). The working group reviewed and discussed the goals of the evaluation, multiple drafts of the survey, and early results and data interpretation. Stakeholders were identified by the working group and included collaborative members, public health professionals, health care professionals, community members, and others who had past and present involvement with the Initiative. Once the questions were finalized, the survey were sent to all the email addresses submitted (n = 461) via a web-based survey platform in late spring 2018 with several reminder/thank you emails.

The survey mostly consisted of questions that used Likert-type scales but also contained open-ended questions included to gain context related to the closed-ended questions and additional topics (e.g., “What were LIHF’s most significant successes?”). Survey topics included access to healthcare; collaboration; LIHF events; community engagement; funding of the Initiative; and racism and representation in LIHF’s leadership, community collaboratives, and priorities. After the survey closed, 132 responses were included in further analyses.

Intensive qualitative interviews were conducted in person for a separate purposively selected sample (n = 25) of key stakeholders. A list of potential interviewees was developed by the evaluation team and the LIHF Program Office, and the final selection of interviewees was made by the evaluation team to protect the identity of respondents. Interview participants were categorized into five groups based on their role in the grant: seven university affiliates, eight program staff, five project grant recipients, six community leaders, and two collaborative members. There was some overlap between categories; some individuals had multiple roles sequentially or simultaneously in the Initiative, so the 25 interviewees worked in 27 different roles throughout the Initiative. The respondents represented all LIHF communities and persons whose roles cut across the four communities: two Beloit, five Kenosha, nine Milwaukee, two Racine, and seven who worked across all sites. Of the 25 interviewees, 19 identified as Black and six identified as non-Black.

The evaluation team developed the interview guide after reviewing preliminary quantitative and qualitative survey results (e.g., open-ended responses to “What would you consider to be the strongest qualities of the initiative?”). The interviewers used a semi-structured guide to ensure that all desired content was covered while remaining open to the lived experience and thoughts of the participants (DiCicco-Bloom & Crabtree, 2006). Recruitment for interviews emphasized that this evaluation was trying to understand how the Initiative was implemented and that participants’ views of the program would assist with the future planning of the Initiative. This evaluation used a culturally-responsive approach and the interviews were primarily arranged and conducted by an evaluator on our team that identifies as Black and has community-based organization experience (TMW). Interviews took place between July and October of 2018, were recorded with the verbal consent of the interviewees, and were transcribed for analysis by a professional transcriptionist. Evaluators communicated with participants via email to arrange meetings in locations that were convenient for interviewees, then drove to sites (e.g., coffee shops, jobs, hotel lobbies, offices) throughout the state to accommodate the participants. These procedures allowed for a faithful representation of the stakeholder’s opinions reproducible in their own words and we uphold the stakeholders’ voice by using extensive direct quotations to illustrate findings. Interviewees who were not employed through the Initiative were compensated for their time.

Data Analysis

The analysis of the survey and interview responses was a lengthy and complex process. Evaluators ensured that the LIHF evaluation working group was active in the analysis to improve the quality of the evaluation and illuminate the process for all parties involved. Quantitative analyses include comparisons of Black and non-Black respondents to survey items to highlight differences between the Black community and non-Black members of this Initiative who were considered to have more numerical and political power within the structure and success of the Initiative. Chi-square tests for independence and a two-sample t-test were used to identify survey items with significant (p < .05) differences between Black and non-Black respondents. Note that both Black and non-Black respondents represented academic and community-based stakeholders.

Analysis for open-ended survey responses and interviews was conducted using thematic analysis, a qualitative method that identifies, analyzes, and reports patterns throughout the data (Braun & Clarke, 2006). Open-ended survey responses and interview transcripts were imported into and analyzed with MAXQDA 2018.1, a software package designed to manage the organization and analysis of textual data. The survey and interview data were coded by two evaluators separately (TMW, AWF) and confirmed by a third (DPM). The evaluators read each open-ended response and categorized emerging themes. The four-person evaluation team met weekly to discuss themes and analysis and reviewed at least 10% of transcripts collectively to create a reliable, collaborative, iterative process of data analysis. Early results were reviewed and further refined by the LIHF evaluation working group.

Quantitative Survey

Table 1 shows results for select individual survey items. Overall, 91% of respondents thought that Black infant mortality was a very or extremely significant issue in their community. For most processes or components of LIHF such as overall leadership, membership in their collaborative, and leadership of their collaborative, just over half of respondents perceived that Black participants were very or extremely equitably or meaningfully involved. However, about two-thirds of respondents did not think that LIHF incorporated the assets and strengths of the Black community into its strategies or actions. There were few significant differences in perceptions between Black and non-Black respondents. Black respondents were more likely than non-Black respondents to perceive Black infant mortality as an “extremely significant” issue but over half of Black respondents saw Black involvement in leadership or decision-making of the collaborative to be not at all, a little, or somewhat involved, a significantly lower perception of Black leadership than perceived by non-Black respondents. There were no significant differences in perceptions on items addressing Black involvement in the overall leadership of the Initiative, equitable and meaningful membership of the collaboratives, developing community action plans, involvement in community project grants, planning collaborative work, and strengths of LIHF as a leader in efforts to improve Black infant mortality.

We also examined items related to successes of the initiative in achieving its primary stated goals for the 10-year program (Table 2). The overall perceptions of success in reducing Black infant mortality and disparities in mortality and in improving the health of Black mothers and infants were modest and did not differ between Black and non-Black respondents. While not shown in the table, there was no significant difference in perceptions of success on any of the areas’ recommended lifecourse topics and social determinants of health areas for reducing Black infant mortality (Lu et al., 2010), and no significant differences in responses regarding success of collaborative processes. Black respondents did provide a significantly (p < .05) more positive assessment of the collaboratives’ capacity to address policy, systems, and environmental change than did non-Black respondents. On a scale of 1 = strongly agree to 7 = strongly disagree for agreement with the statement, “The Collaborative has the capacity to address policy, systems and environmental change approaches to improve Black birth outcomes,” Black respondents had a mean of 2.9 (SD 1.1) and non-Black respondents a mean of 3.7 (SD 1.9); p = .029, overall mean 3.4 (SD1.7).

Qualitative Survey Responses and Intensive Interviews

Power dynamics around race and racism were not sufficiently addressed

Despite efforts made by the funding agency, program office, and convening agency staff to be inclusive, qualitative data from the summative evaluation revealed that members of the Initiative who were not affiliated with the university felt that their experience was not as valued or appreciated. Unequal power dynamics pertaining to race and racism between community members and academic partners surfaced in survey and qualitative interviews. Many of the community-based interviewees stated that as LIHF began to gain momentum, they noticed that the Black community partners were not involved in the decision-making process. They stated that decisions would be made in meetings that did not include community members or partners. After the meetings, university representatives would ask community partners their opinions on decisions that have already been made. They stated that these types of actions added to the already fragile relationship between all parties involved.

White-led institutions directing Black-focused initiatives is problematic

Several of the non-Black academic partners also recognized this tension and expressed their limitations of operating within the university structure to create the systemic change that was needed. Although these non-Black academic partners believed that they were not intentionally withholding power from Black residents of southeastern Wisconsin, Black stakeholders did not believe them. While white academics stated that they were complying with the regulations of the funders, oversight committees, and the university’s Board of Regents, they were perceived by the community stakeholders to be in control of the Initiative. Therefore, when decisions were made, community members were offended that they were not included in the conversations. The intentions of white academics and administrators could have been positive. Still, the history of anti-Black racism in southeastern Wisconsin and white-led institutions planning on behalf of Black communities contributed to how Black stakeholders experienced the power dynamics throughout the Initiative (Harris, 2012; Jones, 2009; Kendi, 2016; Race to Equity Project Team, 2013). In the face of complex power dynamics, academics must advocate for community presence in decision-making.

LIHF was envisioned as a community-academic partnership; however, the concept, funding, and control rested primarily with the university funding agency. It was the WPP’s initiative; its vision, approach, funding process and leadership drove implementation of strategies for reducing Black infant mortality in southeastern Wisconsin. The funding mechanism resembled the federal model of a “cooperative agreement” rather than an outright grant. However, this mechanism led to resistance from community leaders and may not have been well understood.

The End of LIHF

While the focus of this paper is on power dynamics, it is important to note that based on the Pregnancy Risk Assessment Monitoring System and vital statistics data, Black infant mortality disparities and new mother experience was disappointingly constant after 10 years of LIHF (Moberg et al, 2019). While there was no demonstrated change, there was near universal feedback that the Initiative should continue as disparities remained and the issue is critically important. Shortly after the summative evaluation of LIHF was complete, funding for the Initiative was ended. The funder has continued to invest in Black infant mortality through individual project grants aimed at improving maternal and infant outcomes, such as expanding doula services, mother- and father-specific collectives, and building capacity for community-based organizations to collaborate. However, the community collaborative structures described above were undone.

How do we support new dynamics?

Future efforts should have a more hierarchical approach with a centralized priority setting, narrower focus, transparent reporting, and accountability processes would yield a more integrated Initiative with a better chance of success between academic and community partners. Such a structure should continue to give voice to both the “grass tops” and “grass roots” (e.g., Christens and Inzeo, 2015) in the priority communities.

Based on feedback from our interviews, we suggest that university and community partners create steering committees. We believe that if initiatives are guided by diverse staff with content expertise, dissemination and implementation experience, and leadership respected in both the community and the university, it will show both university and community partners how to work together to address pressing issues. These steering committees should be composed of people from the impacted zip codes, offer free childcare and meals during meeting times, honorariums after terms, and be socioeconomically diverse.

Our review of the literature on collaboratives and coalitions found evidence that externally structured community collaboratives significantly influence population health outcomes primarily when they serve as vehicles to advocate for and assure the implementation of evidence-based policies and strategies (Brown et al., 2014). When Black stakeholders and organizations partner with academic institutions, these partnerships must develop organically. All parties can voice concerns, offer suggestions, and engage in healthy debates that enhance community health. All parties involved should have processes to check personal and organizational egos in these partnerships and be reminded that this work is not about the legacy or the reputation of these partners but the lives lost and impacted because of health disparities.