A community-engaged approach to environmental health research: Process and lessons learned

Abstract

Background:

This study used a community-engaged approach to examine associations between environmental contaminants and health outcomes among residents of Yuma, Arizona. Our team conducted a process evaluation to assess scientific rigor and adherence to community engagement principles.

Objective:

Our evaluation focused on four dimensions of community-based participatory research: 1) context, 2) group dynamics, 3) intervention and research, and 4) outcomes.

Methods:

Interviews were conducted with key informants from community partner organizations. Thematic analysis was used to evaluate community partners’ experiences with our collaborative process.

Lessons Learned:

Community partners reported collaborating to establish research goals, recruit participants, collect data, plan analyses, and formulate dissemination strategies. Training needs, roles, and expectations of community partners varied based on available resources, prior research experience, and perceived research challenges.

Conclusions:

Leveraging community-engaged principles for studies of environmental contamination can expedite recruitment efforts and stimulate action to improve health outcomes.

INTRODUCTION

Our project focused on residents of Yuma County, Arizona, which is bounded by the Colorado River to the west and the U.S. border with Mexico to the south. The economy of the Yuma region is based on year-round intensive agriculture.¹ As a result, the Yuma region sustains a large agricultural labor force that includes migrant farmworkers from Mexico in addition to local residents. Yuma’s agricultural production is facilitated by the large-scale use of pesticides and other agrochemicals.²

Prolonged exposure to environmental contaminants is linked to detrimental health effects in humans including cancer, endocrine disruption, congenital malformations, and cognitive impairment.^3–6 Yuma residents, farmworkers in particular, are exposed year-round to a complex suite of agrochemicals and also face the consequences of being medically underserved with limited access to healthcare resources, further exacerbating the negative health outcomes associated with exposure to toxicants.⁷

In 2016, a team of investigators from Northern Arizona University (NAU) along with key stakeholders in Yuma County identified regional priorities for collaborative health research. A high priority was to examine the health impacts of environmental toxicants. We firmly believed that community-engaged research (CEnR) would be the best approach for this project because it focuses on relationships between researchers and communities and aims to achieve a balance between research and action.8 An essential part of conducting effective CEnR is to identify community and academic partners with shared goals who can work together to develop research questions and methods before beginning a project.9

Accordingly, NAU assembled an interdisciplinary team of researchers with expertise in anthropology, ecotoxicology, endocrinology, and public health in collaboration with community partners, including the Yuma Regional Medical Center (YRMC), the Regional Center for Border Health, Inc. (RCBH), and Campesinos Sin Fronteras (CSF). The YRMC is a general medical and surgical hospital in Yuma. The patient base is approximately 50% Hispanic, and the hospital predominantly serves rural, agricultural populations. The RCBH, Inc. operates health clinics in communities in Yuma County, including Somerton and San Luis, and primarily provides care to the underserved, predominantly Hispanic and rural population of South Yuma county, including both resident and cross-border agricultural workers. CSF is an advocacy organization with a mission to promote self-sustainability to farmworkers by providing and facilitating access to health care, social services, housing rehabilitation, counseling, education, and workforce development. The community partners have relevant expertise and access to subpopulations of interest: Yuma County residents and migrant farmworkers.

Following multiple discussions, a literature review, and an interdisciplinary process of matching local needs with NAU resources, a joint community/university project was constructed to examine exposures to environmental contaminants in the region. The NAU investigators and the leadership from community partners participated in a series of consultations about the purpose, design, and subsequent research protocols for this study. These consultations, conducted over approximately 18 months, were supported by formal community-initiated meetings through a research education grant from the Arizona Biomedical Research Centre, and through resources from RCBH as well as pro bono resources from both NAU Yuma’s branch campus and other community partners. This resulted in a community-endorsed primary aim focused on analysis of endocrine function and environmental contaminant concentrations in patients with documented thyroid problems which may be due, in part, to exposure to environmental contaminants.

The overall community engaged design for the project was based on the guiding principles of partnership of the Community Campus Partnerships for Health¹⁰ and extensive experience in the successful formulation and conduct of CEnR projects on the part of both the university and community partners.^11–15 To enhance future partnering practices and achieve research goals, we collectively decided to adapt principles of community-based participatory research (CBPR) for collective reflection and evaluation.^16,17 Following Dolwick et al.,¹⁸ qualitative inquiry was used to provide community members with a platform to communicate their experiences with our collaborative process. Our research team has collaboratively developed models for culturally sensitive evaluation designs over the past 30 years and incorporated those models along with CBPR principles for this project.^19,20

METHODS

The NAU investigators conducted a systematic community-oriented process evaluation focused on four dimensions of the CBPR conceptual logic model21: 1) context, 2) group dynamics/equitable partnerships, 3) intervention and research, and 4) outcomes. To assess adherence to CEnR guidelines, telephone and in-person interviews were conducted with five key informants who represented the leadership at community partner organizations. Interview questions were based on the dimensions of the conceptual logic model of CBPR²¹ and included questions related to interest and relevance of the project topic, effectiveness of training, partner involvement in project development and implementation, potential for impact on local policy, and recommendations for future research.

Key informant interviews were transcribed, and thematic analysis was conducted by NAU researchers to evaluate the experiences of community partners with our collaborative process. Quotes were extracted from the text by assigning a priori codes based on dimensions of the CBPR conceptual model.¹⁷ Member checks were conducted with all interviewees to ensure the accuracy of qualitative results.

While all biological sample and survey data collection procedures were reviewed and approved by the NAU Institutional Review Board (Project number: 1167230–14), key informant interviews were seen as evaluative and thus deemed nonhuman subjects research.

LESSONS LEARNED

CBPR principles are valuable standards for quality control and process evaluation in a wide range of community engaged projects, even those that are not full-blown CBPR. Lessons learned from this project are organized by major constructs in the Wallerstein et al.²¹ model in terms of Contexts, Group Dynamics, Interventions, and Outcomes and bolded for emphasis. Quotes from the interviews, as related to these constructs, are provided in Table 1.

Table 1.

Dimensions of Community-Based Participatory Research (CBPR) and associated key informant quotes

CBPR DIMENSION	Key Informant Quotes	Partner Organization
Context
Community capacity and leadership	“The NAU partnership is very important and improves our organization’s capacity to engage in research.”	YRMC
Place: distance between partners	“The phlebotomist appreciated going to the clinic; It was far better than having to send staff to a different city for training.”	RCBH
Community priorities	“I thought that this might be something good for our community, this might be something good we can find out about how to better protect the health of our population…it resonated with our mission.”	CSF
Group Dynamics
Cultural humility	“Staff and care coordinators appreciated [that one researcher delivered the project introduction in] Spanish.”	RCBH
Dialogue/mutual learning	“Us, as a grassroots community organization, working with the university and other strong partners in the community is not something that happens every day. It gives us an opportunity to educate others around the table [about our organization]. We also had to study why we were doing what we were doing, it was new information for us to educate our community as well.”	CSF
Equitable power dynamics	“Sometimes community groups aren’t understood as well as clinics, universities, etc. We needed to be seen as equal partners and see what training is needed to bring us up to speed with other partners (no medical records, no clinical practice, etc.).”	CSF
Self and collective reflection	“I guess that was one of the biggest lessons learned on this project that working and engaging the community is not as easy as it seems, we may be able to reach out and get people to participate but to actually complete the whole study, we don’t have that infrastructure and it would be a lot harder.”	CSF
Partnership length of time	“We had been working with two of the researchers for years and knew them from involvement in the ABRC advisory board/steering committee.”	RCBH
Community reputation of Principal Investigator	“Since the beginning the relationship, the university researchers have been very genuine and I think we can see that… we appreciated time when we saw his lab and his motivation.”	CSF
Participatory decision-making	“The next phase, going forward, as we start to do statistical analysis of how the contaminant levels we find relate to various health outcomes, that’s where we’ll need to reassess who is doing what and meet back with community partners to decide how to move forward.”	NAU
Shared responsibility	“Division of labor fell out naturally, as the advocacy org for the farmworkers, it made sense that CSF would focus on recruitment and the hospitals would then work with patients to get samples and health workers and NAU would use their expertise on the analytical side.”	NAU
Budget sharing/resource distribution	“This was not a lot of funding involved for anyone (you or me) to do any of this. It would have been better to have one person assigned to this job. Not enough time to do this work on top of our other jobs. It was an additional burden.”	CSF
Trust: following through, keeping promises	“We feel the most important responsibility that we have is to those who participated and trusted us with their blood samples.”	CSF
Language: opinions valued	“Listening to each other and respecting each other, as equal partners and valuing the information that each brings makes a big difference. You guys know a lot about research, but you don’t know about my community, you don’t know about the things we live on a daily basis, the challenges, so that makes me an expert on my own issues.”	CSF
Intervention/Research
Partnership effectiveness/synergy	“We are very committed to continue collaborative work. It has a strong value to the community, but it’s not easy to do. We need to navigate the process better. Communication between community partners made things difficult. Roles needed to be defined better to make sure protocol was followed. Better understanding between [community partners] was needed.”	RCBH
Appropriate research design	“We originally thought that we were going to do the questionnaire and also the health screenings at the same time, and yes the logistics part of that, we thought that it was going to happen one way and it didn’t so that created a barrier with our participants because they had to go somewhere else, they had to find a time, they were limited on time…they dropped out because they couldn’t go for the lab work somewhere else so they had to recruit more people, more than we originally thought so we could have a complete study.”	CSF
Outcomes
Distal “health outcomes”	“We have to be careful how we bring this information into the community not to scare them. We have to educate them and find the best way to give this information back to them. It’s really a process. The best way to disseminate to the participants would be through us - we will play the role as cultural broker.”	CSF
Program sustainability	“It really became hard… because it can’t just be anybody, you have to go through the IRB, you have to be certified to do the work, there was a limitation because you had to only have people who were certified by IRB to be able to work with the population…”	CSF
Capacity to create desired community changes	“This project has the potential to improve the health of individuals in the community. We will be able to apply the findings of this research in our work at the hospital.” “Having information about what really is happening, and having information that is scientific, will empower us.”	YRMC CSF
Preparing academic partner for future collaborative work	“I recommend that there be more investment of time by principal researchers at the beginning of the work to promote understanding of the differences among the various study sites, making sure roles and responsibilities among partners were clearly understood, and how the data collection will be done. This will ensure more internal validity of findings.”	YRMC
Engaging government officials in order to affect policy	“I think finding out more about the results will give us a clue whether there is a problem that needs to be addressed. If there are any issues. Certainly we can have conversations with public works if something needs to be addressed. Surely we can all bring up a public forum.”	RCBH

Abbreviations. “CSF” = Campesinos Sin Fronteras, “NAU” = Northern Arizona University, “RCBH” = Regional Center for Border Health, Inc., “YRMC” = Yuma Regional Medical Center

Contexts

This dimension focuses on factors that influence partnerships, including the historical context of trust/mistrust between universities and communities, the salience of health issues to the community, and the capacity and readiness to engage in a project.¹⁷ In this case, an existing relationship between NAU researchers and community partners helped to pave the way as expressed by one of the community partners: “I had been working with [authors JB and RT] for years and served alongside them on the Arizona Biomedical Research Commission steering committee; I was happy to follow up with them regarding participation in this study.”

Developing rapport among partners and ensuring that the project focused on a health issue salient to the community was of utmost importance. When community partners were asked why they were interested in participating in the project, they responded by saying they had personal connections to the research topic (e.g., had a relative impacted by negative consequences of environmental toxicants) and thought it could affect their community: “I was interested in participating because of how the findings could inform us about the health and well-being of our community.” The research topic was viewed as a community priority by all three partner organizations.

Participation in a day-long, on-site training before the commencement of recruitment activities ensured systematic data collection and protocol adherence across sites. Our training began with a discussion about the goals of the research and recruitment procedures and proceeded to hands-on experience with the delivery and secure storage of informed consent forms, administration of questionnaires, and methods for collecting and storing biological samples. Community partners voiced that the in-person training was valuable. For example, one stated, “We appreciated the day-long training and staff and care coordinators appreciated that it was in Spanish.” Furthermore, community partners appreciated that the training was held in their community as opposed to at the university. For example, one stated, “Having the phlebotomist practice in the clinic made sure there were no challenges during recruitment.”

Although over 20 partner staff members received training and certificates in human subjects research ethics and compliance, it would be beneficial to have more staff trained and certified at the beginning of the project to allow for more flexible data collection. Training staff increased the capacity for the community partners to conduct future research; however, limiting the number of staff who were trained posed an obstacle for community partners during data collection because only those staff certified in human subjects research were allowed to recruit participants, as exemplified by this quote: “It really became hard … there was a limitation because you had to only have people who were certified by IRB [institutional review board] to be able to work with the population.”

We also learned that academic partners needed to understand that existing infrastructure and capacity to conduct research varied across community partners. One day of training may have been enough for some who already had related experience, but more preparation was needed for partners with less research experience. For instance, some field staff were familiar with the use of electronic tablets for data collection, making it easier for them to adapt the questionnaire process to their workflow, whereas others said they could have benefited from more training with this technology.

Group Dynamics

Group dynamics and equitable partnerships relate to relationships, the partnering processes, and the importance of structural agreements among partners to assure community benefit. The history of a strong partnership before entering into this study facilitated the building of trust (Table 1).

Shared responsibility was necessary and while the division of labor seemed to fall out naturally based upon expertise and capacity of each organization (e.g., partners in clinical settings conducted the biological sample collection), community partners expressed some frustration with the lack of clarity around roles and responsibilities between organizations. For example, one stated, “I recommend that there be more investment of time by principal researchers at the beginning of the work to promote understanding of the differences among the various study sites, making sure roles and responsibilities among partners were clearly understood, and how the data collection will be done.” While there was some frustration with role definition, community partners did express feeling comfortable communicating with other members of the study team. For instance, when asked “Did you feel you were able to voice your opinion about the project in front of researchers, administrators, and co-workers?” all key informants responded “yes.”

Numerous respondents noted the importance of flexibility on the part of all partners. One example was the need for one of the organizations to open a lab for the biological sampling in another community with new hours to facilitate the work schedule of participants. The need for an equitable partnership was also voiced and heard; however, some challenges remained. With modest funding for this project, fair resource distribution was a concern. Community partners saw the value of the work, but at times, budgetary restrictions made it difficult to execute the work. For example, one partner stated, “This was not a lot of funding involved for anyone to do any of this. There was not enough time to do this work on top of our other jobs.”

Concerns over the impact of power disparities among partners need to be assessed in relation to the maturity and complexity of on-going partnerships, not simply focused on single project contexts and configurations, especially where the projects are embedded in longer term goals and relationships. We found it valuable to assess and, if necessary, to address contextual issues between different partner organizations in addition to between those organizations and the researchers.

Interventions

This dimension focuses on the extent to which community partners have a voice in how their cultural norms and knowledge are integrated into the design of interventions, methods, and instruments²¹ and the extent of bi-directional translation, implementation, and dissemination so that research findings are used.

To ensure study materials were culturally sensitive, staff from the community partner organizations with prior research experience used their local knowledge to help design and implement recruitment scripts and survey questions. In addition to assisting with the development of the study materials, community partners translated all written materials (e.g., recruitment posters, consent forms, and questionnaires) into Spanish to accommodate non-English-speaking participants.

The knowledge of local project personnel was also particularly effective in the recruitment and retention of participants. Over the course of a one-year period, we recruited, consented, and enrolled 323 adults currently residing in Yuma County (male = 68, female = 255). Most participants identified as Hispanic or Latino (n = 286 [86%]). We conducted outreach and data collection in Spanish for those participants who requested it. Of the 323 participants, 147 (46%) opted to complete the questionnaire in Spanish.

In line with a CEnR approach,²² community partners led all data collection activities using methods amenable to their organizational cultures. For example, while the medical centers (YRMC and RCBH) were able to use patient registries to locate potential participants, CSF opted to screen and recruit participants through telephone outreach. Informed consent and survey administration were conducted across all three community partner sites; however, the responsibility of collecting biological samples was assumed solely by the medical centers which had existing infrastructure and licensed personnel to support sample collection.

Frequent communication between all partners was required throughout the data collection and analysis process to ensure balanced samples across participant groups and effective dissemination of results. In addition to progress reports, weekly communication between partners allowed for several midstream protocol deviations in order to better accommodate participants’ schedules. Following data collection and analysis, multiple in-person and virtual meetings were held between community partners and NAU investigators to review results, coordinate outreach to participants with elevated toxic metal concentrations, and plan for dissemination to the broader community.

Training researchers to be more careful during sample collection or offering alternative sample collection options to align with cultural values of participants could reduce issues of missing data. Although overall data collection methods were considered culturally congruent, some participants expressed concerns about having a hair sample collected. As stated by one community partner, “People were having too much hair cut… We have to take into consideration what the participants are not wanting to do. Is there any other way to get these levels of metals?” Participants who viewed this sample collection method as too invasive chose to opt out of hair collection.

Outcomes

Outcomes range from intermediate systems (e.g., policy and capacity changes) to improved health and social justice.

Community partners felt that the project had the potential to improve the health of individuals in the community, but they expressed frustration about the length of time it took to analyze and disseminate results. Both the service requirements of laboratory equipment and amendments to the institutional review board prolonged the dissemination process; however, all participants with potentially elevated toxic metal concentrations were contacted by their primary care physicians to arrange for follow-up diagnostic testing. Community partners also commented on how the project increased research capacity across sites. For example, over 20 community-based staff were trained and certified for the first time in human subjects research (i.e., CITI training).

A great deal of discussion also occurred regarding best approaches to disseminate results back to the community so as not to alarm people about environmental contaminants and possible exposure. The focus was on how to best engage community members and public officials about the importance of environmental contaminants and the ways to mitigate possible negative health outcomes.

CONCLUSIONS

Leveraging community-engaged approaches in studies of environmental contamination can increase the quality of research and stimulate action that can lead to improved community health. Leadership by community partners in the recruitment process resulted in a highly successful recruitment rate for an otherwise difficult to reach population. With trust, communication, and a shared vision, involvement by community partners in the research process can expedite and enhance participant recruitment, increase study quality, and lead to future collaborative efforts.

The results also suggest a number of arenas in which research capacity might continue to be strengthened, for example, the capacity of research institutions to return results in a timely manner, to present results to the community in a manner that is both informative and provides community members with specific actions that can be taken, and to respond to information about exposures in a manner that is productive. Developing a joint dissemination plan before the onset of the study may also help to overcome some of these issues.

However, overall, similar to other studies, we found that using a CEnR approach improves recruitment and retention efforts,⁸ increases compliance with public health initiatives,²² enhances translation of research findings for community and policy change,²³ and increases the likelihood that the project results are meaningful at the local level.²⁴ Using interdisciplinary research teams can also lead to a more comprehensive understanding of the complex social and chemical/biological issues that lead to health disparities and enhance the translation of findings back to the community.^25,26 For example, the ecotoxicologist, endocrinologist, and community partners in our study all worked together to explain the impact that elevated metal concentrations could have on thyroid function in a way that resonated with participants. Embracing a collaborative model of scientific research and aligning research questions across disciplines can lead to results that are more meaningful to both the participants and the investigators.^22,24,26

Using CEnR principles for on-going quality control measures for community engaged research partnerships may ensure that those partnerships are sustained over time, enhanced through adherence to the basic principles combined with problem solving process evaluation strategies, and lead to the production of both community driven and scientifically based dissemination efforts. This condition enables long term (rather than single project-oriented) collaborations to grow, evolve over time, and be sustainable for both sides of the partnership.¹⁹