A Community Engagement Method to Design Patient Engagement Materials for Cardiovascular Health

Study Setting

The EvidenceNOW Southwest Cooperative is a collaborative effort between Colorado and New Mexico and covers the diverse geographic and cultural regions across both states. It is one of 7 regional cooperatives funded by the Agency for Healthcare Research and Quality to help small and medium-size primary care practices across the country use the latest evidence to improve cardiovascular health. The project was approved by both the Colorado Multiple Institutional Review Board and the University of New Mexico Human Research Protections Office.

EvidenceNOW Southwest Intervention

Practices within both the standard and enhanced arms received the following: ongoing practice transformation support from a trained practice facilitator, support from a clinical health information technology advisor, connection to a regional health extension agent (regional health connector in Colorado, health extension rural officer or HERO in New Mexico), and opportunity to attend biannual collaborative learning sessions in which knowledge is shared across practice teams (also attended by practice facilitators, regional health connectors/health extension regional officers, and clinical health information technology advisors).

Enhanced-arm practices received additional support to increase patient, family, and community engagement within their practice transformation efforts through the use of BCT materials and support to engage patients and families in practice transformation efforts. Because CVD guidelines are relatively inaccessible to the general population owing to medical complexity and jargon, the intention of the BCTs was to allow communities to create materials to provide CVD education in a way that may be more relevant to enhanced-arm communities and practices. Throughout the project, printed BCT materials from all 4 BCTs were available for order at no cost to all enhanced practices (regardless of their geographic proximity to the BCT). Processes and outcomes related to utilization of BCT materials are being tracked and will be reported in a later publication after the completion of the study.

EvidenceNOW Southwest Study Design

Because of the intent to implement the community engagement method (ie, BCT) at the community and regional level, randomization at the level of the primary care practice was not feasible. There was concern about the strong potential for contamination between study arms if standard intervention practices were located in regions from which BCT participants would be recruited and their products would be promoted. Additionally, because the timeline of the study required beginning BCT activities before actual practice recruitment and enrollment, it was necessary to select regions for the BCTs that would include only those practices from the enhanced intervention study arm. We therefore defined geographic regions (ie, cluster) as the unit of randomization. Finally, regional differences across 2 states in underlying population characteristics were considerable, and simple randomization could result in unbalanced study arms, requiring approaches for achieving balanced study arms in terms of the underlying populations and resources.

Covariate Constrained Randomization

Because EvidenceNOW Southwest is a cooperative study across 2 states, and local knowledge about geographic regions was key to informing design decisions, randomization was carried out separately in each state.

In Colorado, we used data from the most recent US Census, Health Landscapes, the Cold Spotting Colorado project,²⁰ and local knowledge about geography and proximity to health care services to define 26 regions consisting of 1 or more counties that are roughly equivalent to the Colorado Health Statistics Regions. These were split into 2 strata for randomization: (1) Front Range and Eastern Plains and (2) mountains and Western Slope. Within each stratum, we used covariate constrained randomization^21,22 to achieve balanced study arms with respect to population and health resource characteristics. Variables included total population of the region, average county population, percentage at or below poverty level, median income, percentage white, percentage Hispanic, number of primary care physicians per 10,000 population, number of physician assistants/nurse practitioners per 10,000 population, unemployment rate, and number of uninsured adults aged 18 to 64 years old. The goal was to define clusters as groups of counties that were relatively homogeneous and large enough to facilitate implementation of BCTs without undue risk of contamination into the standard intervention counties, as well as to achieve balanced study arms on underlying population characteristics.

In New Mexico, using data from the US Census Bureau and the US Department of Labor’s Bureau of Labor Statistics, we divided the state’s 33 counties into 16 regions based upon population size, geography, and proximity to health care services. Using knowledge of local communities and regional differences within the state, we paired the regions into 8 groups and randomized regions within each group, resulting in an even number of regions (8 each) per study arm.

EvidenceNOW Southwest BCT Communities

Once all regions of both states had been assigned to enhanced or standard intervention arms, the regions for the BCTs were identified. Four BCTs were conducted, 2 in each state, with regions selected to represent diverse populations. The limited number of BCTs conducted was due to resource limitations to conduct the BCT and produce the resulting materials. Each region in which a BCT occurred had been randomized to the enhanced study arm.

BCTs were conducted in Hobbs, New Mexico, the South Valley region of Albuquerque, northeast Colorado, and the Denver metropolitan area. These regions differed in ethnic diversity, industry, population, and rural-urban mix. Individual community descriptions can be found in Table 1.

BCT Process

The goal of the EvidenceNOW Southwest BCT was to take evidence-based CVD prevention strategies and translate that content into messaging and materials that are relevant and understandable for community members. Through participation in BCT, community members become better prepared to discuss CVD within their communities, and participants developed unique materials for practices and other community agencies to use for patient/community engagement.^14,15 BCT participants also included practice clinicians and staff, as well as local public health officers, to ensure that the messages and materials would be relevant for use by the practices and local public health.

Each BCT began with an all-day kickoff retreat. For one-half of the retreat, a medical expert presented the current evidence and guidelines for the prevention of CVD, specifically focusing on the ABCS (aspirin use, blood pressure control, cholesterol management, smoking cessation) guidelines. Although each BCT had a different medical expert, presenters used the same set of informational slides across all presentations. After the presentation, the remaining retreat time was spent brainstorming answers to (1) what does our community need to know about CVD prevention, and (2) what actions do we want people to take around CVD prevention?

After the kickoff retreat, each group had a series of 4 to 6 conference calls interspersed with 3 to 4 in-person meetings over a 6- to 9-month period during which they continued to refine key messages and products. Through these follow-up sessions, each group created a set of locally relevant actionable messages and materials.

Recruitment of BCT Participants

Within each state, participants were recruited through use of different organizations, based on which organizations would have better ties with the community. As a result, in New Mexico, with its well-established health extension system, the health extension rural officers recruited participants. In Colorado, with its long history of community-engaged research, practice-based research networks recruited participants. Participants were selected based on their availability to participate for the duration of the BCT, their ability to engage and be present, their interest in the topic, and the absence of a hidden agenda.

BCT Partners

Community members with varying amounts of experience with CVD were included in the BCT. They had differing skill sets and expertise, as shown in Table 3.

BCT Key Messages and Materials

Each BCT was tasked with translating evidence-based cardiovascular risk-reduction strategies into messaging and materials relevant to their local communities to be used by practices for patient engagement. The resulting materials varied considerably from one BCT to another.

The Hobbs, New Mexico, BCT included messaging about spirituality and universality of risk regardless of background, producing 3 products, including a heart-shaped fan that could be taken to football games (Table 1). In the South Valley region of New Mexico, messages aimed to elicit feelings around the importance of and responsibility to Latino communities and family. Prevention was a key message, particularly in the context of family. They produced 4 products including a grocery bag with “Cuídalo” printed on it (translation: “Take care of it”). It was culturally important to the participants that this bag be in Spanglish, a combination of Spanish and English. The northeast Colorado BCT emphasized early intervention but focused heavily on numbers (eg, individual CVD risk score, cholesterol levels) and putting those numbers into an easily interpretable context. The group created 2 products, including a heart chart, a personal tracking tool for recording blood pressure, cholesterol, body mass index, and 10-year risk of CVD similar to pediatric growth curves to provide comparison to normal age-related risk over time. The metropolitan Denver BCT also sought to raise awareness and empower patients to act now to lower their risk. They produced 3 products, including a risk factor checklist with questions prompting patients to think about their own CVD risk factors. Table 1 displays a complete description of key messages and products.

Practice Recruitment

After the BCTs were completed, a total of 211 primary care practices were recruited and enrolled from Colorado (n = 158) and New Mexico (n = 53) across the diverse geographic regions of the 2 states. BCT materials were available only to practices in the enhanced intervention arm until study completion.