Project HEART was the first phase (2005-2008) of a CBPR initiative and included 1) 3 academic partners: the University of Texas at El Paso, the University of Texas at Houston Health Science Center School of Public Health, El Paso Regional Campus, and El Paso Community College; 2) a community clinic, Centro San Vicente, that provides services to residents of El Paso who do not have adequate health insurance coverage; 3) promotores from a network of partner organizations in El Paso; and 4) a community advisory council. Approval was obtained from both institutional review boards of the University of Texas at El Paso and the University of Texas at Houston Health Science Center School of Public Health.

During the first year of the project (September 2005-August 2006), a variety of CBPR strategies were used for Project HEART to engage different constituencies and partners in a dialogue about how to decrease the prevalence of CVD risk factors in Hispanics (most of Mexican origin) who live in El Paso, Texas. They included a promotora community forum, focus groups, and a community advisory council. Information about these activities can be obtained from the first author (HB).

Promotores from the network of partner organizations were trained for 1 week with the Su Corazón, Su Vida curriculum. This curriculum has been used in several of the Salud Para Su Corazón promotora community interventions (13-16). Twenty promotores participated in approximately 16 to 18 hours of training to complete Su Corazón, Su Vida lessons taught by a lead promotora from Centro San Vicente clinic. From the 20 promotores, 3 were hired by the clinic to participate in the 4-month project.

Recruitment of participants for the 4-month intervention started in the fall of 2006. Ten US Census tracts in the 79915 zip code (area of the Lower Valley of El Paso selected by the community advisory council) were randomly assigned to either the experimental or the control group. The Lower Valley area is characterized by a larger percentage of residents of Hispanic descent (94%) and lower mean educational attainment than the rest of El Paso. Only approximately 5% of residents have an education beyond high school, the median annual family income is approximately $26,000, and approximately 26% of families live below the federal poverty threshold (17).

Blocks within each census tract were assigned to recruiters (graduate students and employees of Centro San Vicente clinic), who were unaware whether the census tract was assigned to the experimental or control group. Using a screening instrument designed for the project (Appendix), recruiters visited 3,959 households to determine whether their inhabitants qualified for the study. To qualify for the study, participants had to be aged 30 to 75 years and have at least 1 self-reported risk factor for CVD (smoking, overweight or obese, diabetes, hypertension, or high cholesterol). Exclusion criteria were pregnancy, having a history of CVD, or not planning to stay in El Paso for the remainder of the study. For people who met the inclusion criteria, the study was explained and informed consent was sought. Only 1 participant per household was selected. The intervention phase took place from the fall of 2006 through the spring of 2007.

The experimental group was assigned to a series of 8 health classes using the Su Corazón, Su Vida curriculum conducted by promotores. Each class lasted approximately 2 hours and was delivered every week for 2 months. Participants were then followed for 2 months. Follow-up consisted of 3 telephone calls and a small group session at Centro San Vicente clinic guided by the promotores to discuss changes made as a result of the Su Corazón, Su Vida classes and to encourage further changes. The control participants were given the basic educational materials from the curriculum in person at the time of the baseline assessment. No promotora involvement was provided to the control group.

Height, weight, waist circumference, blood pressure, hemoglobin A1c (HbA1c), and lipids were measured for all participants at baseline and 4 months after the intervention. Body mass index, metabolic syndrome, and Framingham 10-year CVD risk factor scores were also calculated by using standard protocols (18,19). Prehypertension and hypertension prevalence estimates were calculated according to accepted standards from the blood pressure data obtained from participants (20). The questionnaire used validated behavioral measures from previous Salud Para Su Corazón promotora interventions (13-16). The questions assessed 1) perceived susceptibility (feeling that CVD risk factors put them at risk; 3 questions), 2) perceived severity (feeling that CVD is a serious condition; 4 questions), 3) perceived benefits (benefits of behaviors that will help them control CVD risk factors; 7 questions), 4) self-efficacy (confidence in their ability to perform certain behaviors to control CVD risk factors; 6 questions). Responses were measured on a Likert-type scale of 1 to 4 (strongly disagree to strongly agree or not at all confident to very confident for self-efficacy). Demographic characteristics and medication use also were assessed.

Finally, the My Habits Scale previously tested in several Salud Para Su Corazón models (13-16) was used to assess participants' heart-healthy behaviors associated with salt consumption (9 items), cholesterol and fat consumption (7 items), and weight control (7 items). These subscales have shown acceptable reliabilities (Cronbach α coefficients>0.70) in similar promotora studies (13,15).

Differences between baseline demographic variables for the experimental and control groups were tested by using χ² tests for frequencies or independent-samples t test for continuous variables. Unadjusted values for between-group differences before and after the intervention were obtained by using paired-samples t tests. Between-groups analysis of covariance was used to evaluate postintervention results, adjusting for baseline values and covariates that were significant between groups at baseline. An intent-to-treat analysis was used for all comparisons.

At baseline, more than 70% of participants were female, and the mean age was 54 years (Table 1). More than 90% of participants spoke Spanish or were bilingual, and approximately 40% reported having no health insurance. More than half of the participants were born in Mexico (53%). Participants had lived in the United States an average of 39 years. Almost 50% of both experimental and control groups reported a family history of CVD, and more than 65% reported a family history of diabetes. Place of birth, years lived in the United States, language, educational attainment, and self-reported financial status were different between experimental and control groups at baseline. However, except for blood pressure, the groups did not differ on clinical indicators at baseline (data not shown). Fifty-one percent of participants in the experimental group attended all 8 classes. Sixty percent attended at least 4 classes.

For the experimental group, significant decreases were observed at 4-month follow-up for weight, low-density lipoprotein (LDL) cholesterol, total cholesterol, non–high-density lipoprotein (HDL) cholesterol, systolic blood pressure, and diastolic blood pressure (Table 2). For the control group, significant changes were observed for waist circumference (increased), HbA1c (increased), and systolic and diastolic blood pressure (decreased). For both groups, the risk for metabolic syndrome, based on mean number of risk factors, increased slightly. The 10-year risk for CVD using the Framingham score was 5% lower for both groups. For the clinical indicators that were self-reported or calculated, only diastolic hypertension showed a significant decrease from baseline in the experimental group, whereas significant decreases were observed in systolic and diastolic hypertension in the control group.

Only 1 clinical indicator, diastolic blood pressure, was significantly different between the 2 groups after controlling for baseline values and confounders (Table 3). Significant improvements were seen in self-reported behaviors such as weight-control practices, salt intake, and cholesterol and fat intake for the experimental group. Perceived susceptibility to CVD and perceived benefits of behaviors that will help participants control their CVD risk factors were significantly higher in the experimental group (Table 4).