Community-Based Participatory Research: A Capacity-Building Approach for Policy Advocacy Aimed at Eliminating Health Disparities

Abstract

There have been increasing calls for community–academic partnerships to enhance the capacity of partners to engage in policy advocacy aimed at eliminating health disparities. Community-based participatory research (CBPR) is a partnership approach that can facilitate capacity building and policy change through equitable engagement of diverse partners. Toward this end, the Detroit Community–Academic Urban Research Center, a long-standing CBPR partnership, has conducted a policy training project. We describe CBPR and its relevance to health disparities; the interface between CBPR, policy advocacy, and health disparities; the rationale for capacity building to foster policy advocacy; and the process and outcomes of our policy advocacy training. We discuss lessons learned and implications for CBPR and policy advocacy to eliminate health disparities.

Stressors in the social and physical environment are associated with poor health outcomes and contribute to the gaps in health status between rich and poor and between Whites and non-Whites.^1–7 There is growing recognition that community-based participatory research (CBPR) is a viable approach for addressing these health disparities,^8–11 and that such community–academic–practice partnerships can engage the participation of community members in public health advocacy to effect structural change in communities aimed at eliminating health disparities.^12–16 To have a broader and more sustained effect on health, models for influencing policy need to enhance the capacity of community residents and organizations to engage in the policy change process.^11,17,18 By equitably engaging diverse partners in all aspects of the policy process, CBPR can be particularly effective at facilitating capacity building for policy change among community residents.^9,13,14 However, there are limited examples in the literature of CBPR efforts involving capacity-building strategies aimed at enhancing the knowledge and skills of community members to successfully engage in the policy process.

The Detroit Community–Academic Urban Research Center (URC), a long-standing CBPR partnership,^19–21 has conducted the Neighborhoods Working in Partnership (NWP) project aimed at strengthening policy advocacy skills within local neighborhoods, extending community voices in policymaking, and affecting policies aimed at creating healthy neighborhoods. We describe CBPR and its relevance to addressing health disparities, discuss the interface between CBPR and policy advocacy, provide a rationale for why capacity building for policy advocacy is needed at the community level, describe the process and outcomes of the NWP, and discuss lessons learned and implications for CBPR and policy advocacy aimed at eliminating health disparities.

COMMUNITY-BASED PARTICIPATORY RESEARCH AND HEALTH DISPARITIES

CBPR is a partnership approach to research that equitably involves community members, practitioners, and academic researchers in all aspects of the process, enabling all partners to contribute their expertise and share responsibility and ownership.⁸ Its purpose is to enhance understanding of a given phenomenon and to integrate knowledge gained with actions to improve health in the communities involved.⁸ Among the core principles of CBPR are a commitment to build on community strengths and resources, to foster colearning and capacity building, and to balance research and action for mutual benefit of all partners.^8,9,22

Although the use of a CBPR approach is applicable for addressing many health concerns in different contexts, it has primarily been carried out in predominantly low-income communities and communities of color.^9,23 With its emphasis on power sharing and action as well as research, CBPR is an especially viable approach for addressing the understandable distrust of academic research that exists within communities of color,^9,23,24 and for examining and addressing the health disparities experienced by the communities involved.^9,11,25

COMMUNITY-BASED PARTICIPATORY RESEARCH AND POLICY ADVOCACY

Because most of the factors associated with health disparities are beyond any 1 individual's ability to control,^1–7 an advocacy approach that addresses social and economic policy is needed to achieve the structural changes required to eliminate health disparities.^6,13,14,26 CBPR is appropriate for engaging in policy advocacy efforts given community partners' knowledge of local contexts and their relationships with policymakers, and the fact that partnerships are uniquely positioned to identify problems, engage community members in research and interventions, disseminate research findings, and mobilize community members and organizations to advocate for policy change.^6,12,13,27 There are also challenges associated with the use of a CBPR approach for policy advocacy, including the length of time it can take, the difficulty in using a community-based approach to change state and national policies, and the difficulty of sustaining and monitoring action. These challenges make empirical studies of the use of CBPR for policy advocacy all the more important.

There have been few studies in the literature on the impact of CBPR on public policy.¹³ Most recently, Minkler et al. have documented 10 diverse CBPR partnerships that have been effective at promoting policy or systems-level change.^{13,15,16,28–30} One of the key success factors identified across these case studies was partners' knowledge and skills to implement the steps in the policy process.¹³

RATIONALE FOR CAPACITY BUILDING TO FOSTER POLICY ADVOCACY

Although there are numerous frameworks that variously describe the components of the policy advocacy process, all include active community involvement and dialogue among diverse constituencies as critical for effecting policy change.^12,31–36 Benefits of such approaches include involving public interests and values in the decisionmaking process and educating community residents.^34,37,38

The various models of policy advocacy, as with CBPR, emphasize the active engagement of community members at the local level. Missing from these models, however, is a specific focus on providing training to enhance the capacity of community members to engage in the policy change process.^11,17,18,39 A number of specific skills have been identified as critical for successful policy advocacy, including the ability to identify and mobilize individuals and organizations, frame a message to engage different constituencies, and organize activities to gain media coverage.¹⁷ However, the literature includes few examples of how CBPR efforts have carried out strategies aimed at enhancing the capacity of the partners, which is the purpose of the NWP project.

METHODS

The methods described here include both a description of the design and implementation of the NWP intervention and the evaluation design and data collection and analysis methods used.

Intervention

The NWP is being conducted by the Detroit URC, a CBPR partnership established in 1995.⁴⁰ The URC involves collaboration among partners from 8 community-based organizations in Detroit, the local health department, an integrated care system, and an academic institution (see “Acknowledgments” for a list of partners). These organizations comprise the URC Board, which oversees all activities, including adherence to its CBPR principles, and development of new affiliated projects.¹⁹ Prior to the NWP, the work of the URC and affiliated projects was within neighborhoods on the east, southwest, and northwest sides of the city of Detroit. East and northwest Detroit are predominantly African American and southwest Detroit is over 60% Latino. The overarching goal of the URC is to examine and address the social environmental and physical environmental determinants of health disparities. The focus of URC-affiliated projects includes etiologic and intervention research on the social and environmental determinants of diabetes, cardiovascular disease, childhood asthma, and other chronic diseases.^41–46

In 2003, the URC developed a strategic plan that included 2 policy advocacy goals aimed at eliminating health disparities: (1) to enhance capacity at the organization, local, state, and national levels to affect policies and (2) to translate research findings to foster policy change. The NWP project emerged out of these priorities, and the Policy Subcommittee of the URC Board, comprising academic and community partners, has been responsible for overseeing the development, implementation, and evaluation of the NWP.

The NWP project's goals are to strengthen policy advocacy skills within local neighborhoods, to expand community voices in policymaking arenas, and ultimately to promote local, regional, state, and national policies that create healthy, safe, and supportive neighborhoods for children and families. The URC partnered with PolicyLink, a nonprofit organization based in Oakland, California, that is engaged in efforts to advance policies to achieve economic and social equity and to develop the policy training aspects of the NWP project. The NWP is funded by The Skillman Foundation and the University of Michigan and is part of the foundation's Good Neighborhoods Initiative (GNI). The 10-year, $100 million GNI program, which encourages the development of safe, healthy, and vibrant neighborhoods, is involved in 6 Detroit neighborhoods. A GNI community liaison in each of the 6 neighborhoods is responsible for engaging community residents in a planning and community-building process, which focuses on improving the lives of children and youths. The NWP is being carried out in these neighborhoods (in 3 of which the URC was previously involved) and in 1 other neighborhood in which the URC operates.

To achieve the NWP goals, we implemented a 3-step process: (1) a train-the-trainers program to strengthen the capacity of URC partners and other community leaders to train others in policy advocacy, (2) a series of 4 workshops to train neighborhood residents in policy advocacy, and (3) technical assistance to workshop participants in their subsequent policy change efforts.

The program activities of the NWP were guided by an experiential action learning model.^47–50 The model recognizes learning as a continuous cycle of knowledge acquisition, experience, observation and reflection, development of concepts and generalizations, and integration and application of these concepts and generalizations to real-life situations.⁵⁰ In accordance with this model, the aim of the NWP program activities was to create new understandings and skills through a combination of didactic methods, role plays, group discussions, problem-solving tasks, reflection, and feedback.

Train the trainers.

The URC Policy Subcommittee identified academic and community trainers who would be responsible for carrying out the project; these trainers worked in pairs in each of the 7 different neighborhoods. The trainers' roles included participating in training, developing curricula for the neighborhood-level workshops, cofacilitating workshops, and providing ongoing technical assistance. Six academic and 2 community trainers were members of the URC Board or an affiliated project steering committee. Seven additional community trainers were hired on the basis of criteria that included the following: experience in community organizing, skills in public speaking and group facilitation, familiarity with at least 1 of the neighborhoods involved, cultural competence, experience with policy-related activities, Detroit residency, and experience working with youths.

Community and academic trainers participated in a series of train-the-trainers workshops developed by PolicyLink staff in consultation with the policy subcommittee. The content of the 3 day-long training sessions, which focused on topics relevant for policy advocacy at the organizational, local, state, and national levels, included the following: an in-depth overview of policy and advocacy, building the skills needed to design an advocacy campaign, power mapping (a power analysis tool to map power and shape a campaign strategy), communicating with policymakers, critical elements used to advance campaigns (e.g., building alliances with key partners), and communication techniques and development of a communication strategy. Following the train-the-trainers series, the trainer teams worked closely with project staff to adapt the curriculum, both content and process, for the neighborhood-level training sessions.

Training recruitment and implementation at the neighborhood level.

To make more efficient use of staff time and community resources, adjacent GNI neighborhoods were grouped together and the trainings were conducted in a total of 4 different neighborhood venues at community- and faith-based organizations that were convenient and familiar to local residents. Between April and August 2008, a series of 3 half-day workshops were conducted in each of the 4 neighborhood groupings. Each series of workshops was carried out over a 6- to 8-week period, with all sessions conducted on Saturdays. A fourth, citywide workshop, which was held in October 2008, was open to all participants who had attended at least 1 of the previous workshops. With assistance from the GNI community liaisons, the URC trainers, staff, and community partners recruited adults and youths (aged 14–22 years) to attend the training series in their respective neighborhoods. Child care and a continental breakfast and lunch were provided at each session. Attendees received a $10 gift card at each of the trainings to acknowledge their participation and as an incentive to participate in subsequent trainings.

Teams of URC community and academic trainers facilitated the first 2 sessions in each series. Workshops combined short, didactic segments with interactive exercises and role playing. Each session started and ended with the youth and adult participants meeting together, but they were divided into separate groups for parts of the sessions. The first session included 3 topics: an introduction to policy and advocacy, an understanding of the difference between programs and policies, and the steps involved in developing a policy advocacy campaign. The second session covered 2 topics: a discussion of power (who has it and how to mobilize it using “power mapping”) and the development of talking points for communicating with policymakers. Case examples, based on priority issues from GNI action plans that had been developed in each of the 6 neighborhoods, were used in interactive exercises throughout the workshops.

The third session in each of the neighborhoods was conducted by staff from Michigan's Children, a statewide organization that works with multiple constituencies to provide an independent voice for children and their families. This session covered the Michigan legislative process (e.g., how a bill becomes law, who the major players are in state government) and how to communicate with policymakers and to influence and become involved in that process. The session included a discussion with invited state and local policymakers, giving participants an opportunity to talk with policymakers about issues on which they were working in their communities.

The fourth session, which was attended by residents from all 7 neighborhoods, was facilitated by all of the URC trainers. The content focused on selecting strategies to achieve policy change, using communications for policy change, and building alliances across neighborhoods. The trainers discussed the provision of ongoing technical assistance to participants as they engage in the policy advocacy process.

Evaluation

The purpose of this evaluation was to assess the process, outcomes, and impact of the NWP program. We used a participatory and formative approach that involved URC Board members, policy trainers, and the NWP evaluation team in a collaborative process that included the development of measurement instruments, data collection procedures, and the interpretation and application of findings.^51–53 The participatory approach recognizes that the partners' active involvement in the evaluation enhances the relevance and increases the usefulness of the results.^52,54 As a formative approach, results were shared with the URC Board and policy trainers on an ongoing basis, providing opportunities for interpretation and application of the findings to improve the program.^55,56

The process evaluation answered questions related to project implementation and to understanding how and why various aspects of the project were or were not effective. The outcome evaluation addressed questions related to the program's effect on participants' knowledge, perceived self-efficacy and collective efficacy, behavioral intentions, and behaviors related to policy advocacy. The impact evaluation addressed the extent to which knowledge and skills gained by participants were applied to policy change efforts. In this article, we describe the data collected and the results for the process and outcome evaluation. The impact evaluation data, which were collected through group interviews approximately 6 to 8 months after the last training, are presently being analyzed and will be reported in a separate article.

Data collection.

Pre- and postseries questionnaires and workshop evaluation questionnaires were developed on the basis of a review of the literature, with input from the trainers and URC Board members. All questionnaires were self-administered. We tried to keep questionnaires near a sixth-grade reading level, using Microsoft Word 2007 reading grade-level assessment tool (Microsoft Corporation, Redmond, WA). All questionnaires were translated and back-translated to and from Spanish, and were piloted tested with adults and youths who represented the focal neighborhoods.

At the workshop registration, adults were asked to read and sign the informed consent form. Youths were asked to sign the youth assent form, and their parents were asked to sign the parental consent form. All postworkshop questionnaires were administered during the last 15 minutes of each workshop.

Pre- and postseries questionnaires were developed to assess the extent to which curricular materials and the training activities increased the capacity of participants to carry out policy-related work. Following standard survey protocol, participants completed a preassessment questionnaire prior to the start of the workshop. Closed-ended questions assessed content covered in the training, such as distinguishing between policies and programs, defining advocacy, and developing talking points. The preseries assessment questionnaire was administered immediately before the first session that the participant attended. The postseries questionnaire was administered immediately after the third and fourth workshop sessions.

A workshop evaluation questionnaire was administered at the end of each session. Closed-ended items were used to assess participants' satisfaction with the workshop, the extent to which participants found the training useful, and their intentions to use the material learned in the training.⁵⁷ The questionnaire also assessed the extent to which participants felt they had the knowledge and power necessary to advocate for policy change, and the extent to which people in Detroit had the power to change policies in their neighborhoods (i.e., perceived self-efficacy and collective efficacy).⁵⁸ Each of these items was measured on a 5-point Likert scale (strongly disagree to strongly agree), and responses were dichotomized to “disagree or neutral” or “agree.” Additional closed-ended questions asked whether participants had worked for policy change in the last 6 months, and if so, the types of policy advocacy activities they had done (e.g., attended a hearing, participated on a policymaking board). Open-ended questions asked what participants felt was most important and least important about the training and how the training could be improved. The questionnaire administered at the fourth workshop session also asked in what ways, if any, participation in the training had made a difference in their community. Questions to obtain demographic information (e.g., sex, age) were included in the workshop evaluation questionnaire of the first session that the participant attended.

Ninety-five of 115 participants in session 1 (83%), 95 of 135 participants in session 2 (70%), 100 of 116 participants in session 3 (86%), and 52 of 65 participants in session 4 (80%) completed the workshop evaluation questionnaire. Of the 111 participants who attended 2 or more sessions, 98 completed the questionnaires at 2 or more points in time (88%). Data from the first and last session attended were used to analyze change over time in participants' behavioral intentions, policy change behaviors, and perceived self-efficacy and collective efficacy. Of 228 total workshop participants, 165 completed the preassessment questionnaire (72%), and 48 of those also completed the postassessment questionnaire (21%).

Data analysis.

All descriptive quantitative analysis was conducted with SPSS version 13.0 for Windows (SPSS Inc, Chicago, IL). Analysis of change over time was conducted with SAS version 9.1 (SAS Institute, Cary, NC).

We calculated frequencies and percentages for the demographics of workshop participants, participants' assessment of the usefulness of the training, behavioral intentions, policy advocacy behaviors, and self-efficacy and collective efficacy related to policy advocacy. These calculations were made for each neighborhood, across neighborhoods for each session, and across all neighborhoods and sessions. Means for each of the 38 knowledge items were calculated across all study participants who completed either a pre- or a postseries questionnaire (n = 165). We examined means for each knowledge item for the subset of participants who completed both the pre- and postseries questionnaire (n = 48), and separately for adults (n = 27) and youths (n = 17). Finally, we created an index showing the percentage of all possible knowledge items answered correctly for all participants (n = 165) and for the subset who completed both pre- and postseries questionnaires (n = 48).

We first analyzed open-ended questions by session and neighborhood and then analyzed them across neighborhood, following standard procedures for the analysis of qualitative data.⁵⁵ All responses to the open-ended questions in the workshop evaluation were first typed into a Microsoft Word document and then grouped by conceptually related themes. We gave each group of themed responses a heading that captured the meaning of all responses in the group, using verbatim language from at least 1 of the comments. Any comment that did not fit into a themed group was included in a miscellaneous category. Four members of the evaluation team met to discuss and agree on all groups and heading names. We then combined the themed groups from each neighborhood into 1 document for each session across neighborhoods, using a similar grouping technique. Finally, these themed groups for each session were combined to create a summary document that included responses across all neighborhoods and sessions.

RESULTS

As shown in Table 1, 228 people participated in at least 1 session of the NWP training: 107 attended 1 session, 59 attended 2, 41 attended 3, and 21 attended all 4 sessions. Women represented 63% of workshop participants, 52% of participants were adults, and 48% were youths. To help us understand the characteristics of those who attended sessions, we compared participants who attended 2 or more sessions with those who attended only 1 session, by sex and age. The 2 groups did not differ significantly by sex. Participants who were aged 18 to 39 years were significantly less likely to attend more than 1 session than were participants who were aged younger than 18 years or older than 39 years.

TABLE 1.

Number of Sessions Attended and Demographic Characteristics of Participants: Neighborhoods Working in Partnership Workshop Series, Detroit, MI, 2008

	Participants, No. or %
No. of sessions attended
1 only	107
2 only	59
3 only	41
All 4	21
≥ 1	228
Sex (n = 227), %
Female	63
Male	37
Age (n = 228), y, %
≥ 18	52
< 18	48

Participants' Assessment of Training

The responses to the question “What was most important about the training?” were overwhelmingly positive, with numerous participants responding that “everything” was most important. Themes identified through the analysis of qualitative data across all of the training sessions, organized by the content and the process of the workshops, are presented in the box on the next page. Responses indicated that the most important aspects of the content were learning about the advocacy process and how to make a difference and learning how to communicate with policymakers. The most important aspects of the process included the small-group activities, networking, and speaking with policymakers.

At the end of the fourth, citywide session, participants were asked, “In what ways, if any, has your participation in the policy training made a difference in your involvement in your community?” As shown in the box on the next page, themes identified here included learning that they had power and the information, tools, and skills needed to organize and bring about change in their community.

Assessment of Training's Usefulness and Behavioral Intentions

As presented in Table 2, the responses to the questions related to the usefulness of the training were quite positive, with a high percentage of participants across the 4 training sessions agreeing or strongly agreeing that the training was useful (94%), that the ideas discussed would work in their neighborhood (86%), and that they would be able to use what they had learned to bring about change (87%). Responses to questions related to behavioral intentions were also quite high (Table 2), with participants agreeing or strongly agreeing that they planned to do more work to change policies (87%), that they planned to work with others who attended the training to bring about change in their neighborhood (88%), and that they planned to work with other neighborhoods to advocate for policy changes (80%).

TABLE 2.

Participants' Assessment of Neighborhoods Working in Partnership Workshop Series and Participants' Behavioral Intentions: Detroit, MI, 2008

Questionnaire Item	Agree or Strongly Agree, %	Disagree, Strongly Disagree, or Neutral, %	No. Across 4 Sessions
Usefulness
I found the Neighborhoods Working in Partnership training useful.	93.6	12.2	376
The ideas discussed in the training will work in my neighborhood.	85.9	14.1	375
I will be able to use what I learned here to bring about change in my neighborhood.	87.3	12.7	378
Behavioral intentions
Because of what I learned here, I plan to do more work to change policies.	87.3	12.7	377
I plan to work with others who attended the training to bring about change in our neighborhood.	88.4	11.6	388
I plan to work with other neighborhoods to advocate for policy changes that we all care about.a	79.5	20.5	39

^aThis item was included only at session 4.

Change in Perceptions of Self-Efficacy and Collective Efficacy

Table 3 presents and compares mean scores on 3 efficacy questions answered at the first (preassessment) and last (postassessment) session attended by individuals who participated in 2 or more workshop sessions (n = 95). Participants' perceptions of self-efficacy, as expressed by the statement “I know how to work for policy change,” increased from a mean of 3.95 at the preassessment session to 4.23 at the postassessment session (1 = disagree strongly, 5 = agree strongly; P = .01). Participants further reported a high level of collective efficacy in the ability, by working with others, to change policies that affected their neighborhood, with no significant change between the preassessment session (4.37) and postassessment session (4.48; P = .2). Finally, participants reported a moderate level of collective efficacy and no significant change in response to the statement that people in Detroit did not have enough power to change policies in their neighborhood (2.39 at preassessment and 2.41 at postassessment; P = .94).

TABLE 3.

Mean Scores of Participants' Responses to Questions Regarding Efficacy: Neighborhoods Working in Partnership Workshop Series, Detroit, MI, 2008

Questionnaire Item	Preseries Score	Postseries Score	P
I know how to work for policy change.	3.95	4.23	.01
Working with others, I can change policies that affect my neighborhood.	4.37	4.48	.2
I feel that people in Detroit do not have enough power to change policies in their neighborhood.	2.39	2.41	.94

Note. The response range was 1 to 5 (1 = strongly disagree; 2 = disagree; 3 = neutral; 4 = agree; 5 = strongly agree). The total number of respondents was 95.

Participants' Knowledge

Results for the knowledge items showed relatively high levels of knowledge on the items assessed through the pre- and postseries questionnaires, with some differences between those who completed both pre- and postseries knowledge assessments and those who completed just 1 of the assessments. Three quarters of those who completed both the pre- and postseries knowledge questionnaires correctly answered 80% or more of the 38 items at both points in time. In contrast, among those who completed only the postseries assessment (n = 72), only 49% correctly responded to 80% or more of the 38 knowledge items. There were no discernable patterns of change over time in individual items included in the knowledge assessments for those who completed both the pre- and postseries questionnaire (results not shown).

Themes Identified From Participants' Responses to Open-Ended Questions: Neighborhoods Working in Partnership Workshop Series, Detroit, MI, 2008

What was most important to you about today's training?

Content

• Learning about advocacy process: how to make a difference

• “Power mapping”: identifying supporters and opponents

• Learning how to communicate to policymakers

• Learning how to make things happen by working together

Process

• Small-group activities

• Networking: bringing youths and adults together to bring about change

• Giving everyone a chance to talk and be heard

• Opportunity to speak with policymakers

In what ways, if any, has your participation in the policy training made a difference in your involvement with your community? Please describe.

• Showed that I have power and can make a change in my community

• Gained more information, more tools to get what is needed to move forward

• Feel able to help organize a successful group for positive change

• Helped to network and share stories and reunite with old friends

Change in Participants' Behaviors

Of the 91 participants who attended at least 2 or more training sessions and completed a question about their policy change activities, 58 (64%) indicated at the end of their first session that they had not worked for policy change in the past 6 months. By the end of their final session, 23 (40%) of those who originally indicated they had not worked for policy change reported that they had worked for policy change in the previous 6 months. This represents a substantial increase in the number of participants working for policy change since the beginning of their participation in the NWP (for χ² test for independence, P < .001; odds ratio = 6.84; 95% confidence interval [CI] = 2.45, 19.16).

DISCUSSION

Participants reported that numerous aspects of both the content and the process of the workshops were important to them. These included learning about steps and strategies for conducting a successful policy advocacy campaign and the experiential learning activities and opportunities to network with other people in their neighborhoods (including youths and adults) and policymakers. They also indicated that they thought the training was very useful and applicable for bringing about policy change in their neighborhoods, and that they had high intentions of applying the training to work toward policy change. This latter finding is important given that behavioral intention is widely considered to be an immediate antecedent of behavior,⁵⁷ and there is also research evidence that behavioral intention may engender change in behavior.⁵⁹

There were several important findings regarding change in self-efficacy and collective efficacy, knowledge, and behavior. A high proportion of participants started the program with a sense that by working together they could change policies affecting their own neighborhood, which perhaps explains why there was not a significant change from the first to the last session. Over the course of the workshop series, the greatest change in perceptions of efficacy was at the individual level of knowing how to work for policy change, even though the rating was also fairly high at the start of the program. Both self-efficacy, the belief in one's ability to succeed in a specific situation, and collective efficacy, the shared belief in a group's ability to take action and solve problems,⁵⁸ play key roles in engaging people in policy change efforts to improve public health.^60,61

There were no significant changes over time in the individual knowledge items included in the pre- and postseries questionnaires. However, 75% of individuals who completed the knowledge items at both points in time correctly answered 80% or more of the knowledge items. By contrast, only 49% of individuals who completed only the postseries assessment correctly responded to 80% or more of the knowledge items. It may be that given the high percentage of persons who answered the knowledge questions correctly during the preassessment, there was little room for an increase in knowledge levels over time. Regarding behavior change, of those who indicated on the workshop evaluation at their first session that they had not worked for policy change in the preceding 6 months, 40% indicated at their final session that they had worked for policy change. This suggests a considerable increase in involvement in policy-related activities following participation in the training.

Limitations

There are several limitations of the evaluation presented here. First, although we recruited 228 neighborhood residents to the training, almost half of whom were youths, only 21 participants were able to attend all 4 sessions. In addition, there was a considerable gap between the number of participants who attended the trainings and the number who also completed the pre- and postassessment questionnaires. Furthermore, although the knowledge questions were generated by the evaluation team with assistance from the trainers and URC Board members and were pilot tested with Detroit-based residents, for institutional review board purposes the questions had to be developed before the curriculum for all 4 sessions was finalized. Some of the questions were therefore less relevant for capturing change in knowledge, and several questions were eliminated from these analyses because the content was not covered in the training. The knowledge questions were not tested for reliability. Despite our efforts to phrase questions at a sixth-grade literacy level, the overall questionnaire was rated at an eighth-grade level because of some terms that had to be included (e.g., policy advocacy). Finally, because knowledge alone is not a good predictor of behavior change,⁶² assessment of change in knowledge may not be the most appropriate assessment approach for our ultimate interest in behavioral changes. Our original evaluation design therefore included questions on behavioral intentions and efficacy, and we are presently analyzing group interviews conducted after the end of the training sequence to further assess the impact of the training on participants' involvement in policy advocacy efforts.

Lessons Learned and Implications for Practice

The following recommendations are based on our lessons learned from conducting the NWP, with the understanding that CBPR partnerships need to tailor their efforts to the local community context.

Design the process of training to incorporate participants' diverse learning needs.

The NWP used an experiential learning model that emphasizes providing participants with opportunities to engage with others to apply knowledge learned, to practice that knowledge to enhance skill development, and to reflect on the process.^48,49 As presented in the box on the previous page, the use of small-group activities and the opportunity for everyone to talk and be heard were identified as 2 of the most important aspects of the training. When such an approach is used, it is important to give participants an adequate amount of time to interact with others in small-group activities while neither compromising the essential content nor rushing through the training. In addition, if youths and adults are trained together, it is important to provide adequate time for youths to work on activities without adults being present. This enables the trainers to present materials and activities in age-appropriate ways, and to provide youths the time and space to share their own voices. At the same time, it is important to provide youths and adults an opportunity to meet together to inform and learn from each other.

Acknowledge that competing demands create challenges for consistent participation and plan trainings accordingly.

Even though we provided child care, food, and gift cards as incentives for attendance, there were not many participants who were able to attend all 4 workshops (Table 1). Because of the competing demands on community members' lives, it is probably unrealistic to expect participants to be able to spend multiple half-days in such training, and this may be particularly challenging on Saturdays during the summer months. The finding that participants aged 18 to 39 years were less likely than were participants of other ages to attend more than 1 session lends further insight, as participants in this age range are likely to have jobs, children, and other family responsibilities that constrain their ability to participate. One approach to accommodating this situation is to review content at the beginning of each session, to ensure that all participants are exposed to materials covered in previous sessions. At the same time, it is important that the material not be too redundant for participants who were able to attend previous sessions. Another approach is to package learning modules within a shorter time frame that is manageable within the local community context, while striving not to compromise the content that needs to be covered. We recommend the latter approach for increasing the engagement of community residents in policy advocacy: cover core content in 2 or 3 sessions and then combine this core content with ongoing technical assistance to participants engaged in a specific policy advocacy campaign.

Conducting participatory and formative evaluation, feeding back, and applying results to improve the program.

As described in the Evaluation section, we were able to analyze the evaluation data in a timely manner and feed back the results to the trainers and members of the URC Board to incorporate the results into the design of subsequent trainings. This was helpful between neighborhoods to improve the training quality for the next neighborhood and within neighborhoods to prepare for the next training in the series. As has been suggested by others, we recommend the use of such a participatory and formative evaluation, using both quantitative and qualitative methods, to enhance the quality of the program as it is being conducted.^51–55

Use partnership process, which is critical for the success of capacity building and policy change.

As discussed in the Intervention section, the NWP arose out of an existing CBPR partnership, the Detroit URC, whose members share a vision about their role in increasing the capacity of neighborhood residents to engage in policy change. Through this endeavor, the partnership has been able to solidify its existing relationships and extend beyond them to develop relationships with new organizations and neighborhoods. Doing this work through a partnership approach has strengthened the expertise and capacity of the URC to expand its efforts into the policymaking arena, and has increased its visibility and reach to other neighborhoods in Detroit. The long history of positive relationships that exists between the community and academic partners, and the credibility that diverse partners bring, have greatly contributed to these accomplishments. No one partner organization alone would have been able to create the synergy required to accomplish similar goals. We therefore recommend, as have others, the use of a CBPR partnership approach for efforts aimed at enhancing capacity for policy change.^12–14

Conclusions

The NWP has addressed many approaches to policy change that have relevance for community members and neighborhoods interested in affecting policy.^63–65 Of critical importance is the use of information obtained through community assessments and research findings to inform policy development based on issues of concern to the communities involved. Such assessments and research findings also enhance the credibility and strategic focus of community policy initiatives with policymakers and the larger community.^{14 ,66} Furthermore, a CBPR partnership approach to policy change needs to incorporate features that assist the participating residents, organizations, and neighborhoods to achieve goals such as building the capacity of the organization and neighborhood, solving relevant problems, and fostering a sense of community.¹⁴ Our evaluation provides evidence of participants' increased learning about how to engage in policy change, high levels of behavioral intentions, and achieved changes in self-efficacy and policy-related behaviors.

In the next phase of the NWP, we will continue to provide training to neighborhood residents and academic partners, as well as technical support to community groups engaged in specific policy advocacy campaigns. We will also place greater emphasis on the translation of community assessment and research findings to affect policy change. The use of a CBPR approach to enhance the capacity of neighborhood residents to engage in policy advocacy is a necessary, but not sufficient, step toward addressing the broad-scale changes needed to eliminate health disparities. As stated by one of the workshop participants, what was most important about the NWP training was “knowing that I have power that will positively impact my neighborhood and that now I know how to use it.”