Introduction

Approximately 560 federally recognized American Indian and Alaska Native (AI/AN) communities and tribes share a history of conquest, displacement (often several times), and cultural trauma (Cobb, Wingo, & Edwards, 2008). Both rural and urban AI/AN populations experience greater health disparities, as well as lower education levels, poorer housing conditions, and lower socioeconomic status, than other U.S. racial/ethnic groups (Indian Health Service [IHS], 1999; U.S. Census Bureau, 2003). The Strong Heart Study, a longitudinal observational study, has shown that many AI communities bear a heavy burden of cardiovascular disease (CVD); for example, the incidence of coronary heart disease in AI men and women is twice as high as in the rest of the U.S. population (Howard et al., 1999). Moreover, among U.S. population groups, AI populations have the highest rates of three CVD risk factors: cigarette smoking, obesity, and diabetes (Lee et al., 1998; Lee et al., 2002). These risk factors are potentially modifiable through effective interventions.

Teufel-Shone, Fitzgerald, Teufel-Shone and Gamber (2009) systematically reviewed physical activity interventions (most of which also included dietary intervention components) implemented in AI/AN and Native Hawaiian populations. Only five of the 27 interventions with evaluation components reported significant changes in health and fitness measures: one reported significant weight loss; three reported measureable increases in physical activity; and two reported significant improvements in participants’ knowledge. Forty percent of the programs lasted 5 years or longer, with program sustainability linked to locally trained personnel, local leadership, and stable funding. Physical activity interventions impacted not only individual health, but also community resources. Teufel-Shone et al. identified the following significant gaps: (1) a paucity of programs focusing on urban AI/AN populations and (2) a lack of culturally acceptable and scientifically sound evaluation methods that can be implemented by local personnel to assess the health and social impact of the interventions. Teufel-Shone et al. ultimately recommended that future research (1) enhance the cultural and social relevance of the interventions, (2) integrate the interventions into the tribal infrastructure, and (3) develop sustainable evaluation plans.

One of the studies included in Teufel-Shone et al.’s systematic review was the Pathways Study, which was initiated after data from the Strong Heart Study showed significant health disparities in overweight and obesity (Caballero et al., 2003). Pathways was a cluster-randomized controlled trial in 41 schools, targeting AI children from third to fifth grade. The intervention had four components: (1) change in dietary intake, (2) increase in physical activity, (3) a classroom curriculum focused on healthy eating and lifestyle, and (4) a family-involvement program. The intervention resulted in a significant reduction in the percentage of energy from fat in the intervention schools, but not in percentage body fat. Children in intervention schools self-reported more physical activity, but accelerometer data showed similar activity levels for children in both intervention and control schools. Several components of knowledge, attitudes, and behaviors were also positively and significantly changed by the intervention.

In the area of smoking cessation, six intervention studies have been published in the literature. Research on smoking cessation in special populations has been reviewed by Doolan and Froelicher (2006) and by Lawrence, Graber, Mills, Meissner, and Warnecke (2003). The former found no published randomized controlled trials in AI/AN populations from 2000 to 2005, and the latter found only three. To date, no AI-oriented program for pan-tribal cessation of tobacco abuse has been developed and shown to be effective (Daley et al., 2006).

The exact reasons for the obesity and diabetes epidemic in AI/AN communities (and in other colonized indigenous populations) remain elusive (Lee et al., 2002). Some researchers have posited the existence of a “thrifty genotype” (e.g., Neel, 1962; Dowse & Zimmet, 1993; Swinburn, 1996), but others have challenged the existence of thrifty genes (e.g., Baschetti, 1998; Bradley, 1992; O’Dea, 1992; Ozanne & Hales, 1998). However, one important factor in the obesity and diabetes epidemic appears to be the adoption of unhealthy Western diets and lifestyles at the expense of more healthy, traditional Native diets and lifestyles (O’Dea, 1992). Traditional diets vary by tribe and region, but they are generally healthy and include plant products (e.g., corn, climbing beans [but not pinto beans], and squash, collectively known as “The Three Sisters”), gathered foods (e.g., wild rice, roots, berries, nuts), and lean game meats (e.g., bison, deer, turkey, fish). The traditional lifestyle is very active, and includes hunting, gathering, fishing, canoeing, traditional dances, and games (e.g., Native North Americans were the first to play hockey and lacrosse).

Tobacco is a sacred plant in many Indian cultures, and it is used in prayer and in other customs (Hodge, 2002; Struthers & Hodge, 2004). For example, the Six Nations of the Iroquois burn tobacco to open ceremonies and other events; the Diné (Navajo) consider tobacco one of their four most sacred plants, along with corn, beans, and squash (Daley et al., 2006). It was the Europeans who introduced habitual use of tobacco and mass-produced cigarettes. Smoking cigarettes is considered to be inconsistent with adhering to AI/AN culture and is known as abusing tobacco. Permissive attitudes towards smoking and smoking behavior now exist in Indian Country (Hodge & Casken, 1999).

Request for Applications

To address these health disparities, a Request for Applications (RFA) for cooperative agreements solicited applications to conduct 5-year studies in AI/AN communities. The purpose of these studies was to test the effectiveness of behavioral interventions developed to promote the adoption of healthy lifestyles and/or modification of behaviors related to CVD risk (e.g., healthy diet, regular physical activity, and smoking cessation). The targets of these behavioral interventions were biological CVD risk factors, such as hypertension, dyslipidemia, obesity, glucose intolerance, and diabetes (National Institutes of Health, 2004, 2005). An individual applicant could select one or more risky behaviors upon which to intervene, but applicants were not required to intervene upon all behavioral risks.

The concept for this RFA was developed in consultation with AI/AN communities, recognizing the government-to-government relationship between Tribal and U.S. governments, and incorporated recommendations of a working group of AI/AN health experts (http://www.nhlbi.nih.gov/meetings/workshops/ai-an1.htm). Applicants were expected to develop and conduct their projects in collaboration with the communities.

The target population was identified as AI/AN adults and/or children at high risk for CVD but free of clinical (or previously diagnosed) CVD. Applicants could propose a primary prevention program aimed to improve CVD risk factor(s) in those who possess a risk factor (e.g., already obese or hypertensive). Alternatively, applicants could propose a primordial prevention program aimed to prevent the occurrence of the causative risk factor (e.g., preventing obesity or hypertension).

Applicants could propose a design to randomize either individuals or an entire family or, alternatively, a cluster-randomized design that would randomize either communities or schools. Applicants proposing the latter design were required to include at least two schools or communities in the intervention condition and at least two schools or communities in the comparison or control condition. The budget for the projects did not allow for a full-scale cluster-randomized design with 35–40 schools or communities, as was the case in the Pathways Study (Caballero et al., 2003). The intervention could target individuals, families, schools, or communities. All participants would continue to receive medical care from their usual source of care.

Applicants were also expected to negotiate with AI/AN communities regarding control and ownership of the data and biological samples. Because some researchers in the past have published analyses of data, including biological sample data, that were not covered in their research agreements with outside researchers, most communities now retain the right to review manuscripts and abstracts before submission. This ensures that researchers publish only data that were included in their research agreements with the communities, and neither constitutes censorship nor infringes upon academic freedom. Biological samples are considered to be sacred in Indian Country; therefore, communities wish to control access to these samples. Moreover, some communities also require that their community identity not be revealed but that they only be referred to in the publication or presentation using a geographic region of the country (National Institutes of Health 2004, 2005).

Trials

Funding of four trials was planned, but the quality of the applications resulted in the funding of five at a total cost of approximately $14.0 million. This special section of The Journal of Primary Prevention contains articles describing the design of each of the five funded trials. They are summarized below and in the Table 1.

Table 1 Design characteristics of the five randomized controlled trials targeting CVD risk factors among AI/AN communities
Full size table

The first contribution, A Community-Based Intervention to Prevent Obesity Beginning at Birth Among American Indian Children: Study Design and Rationale for the PTOTS Study, describes a two-arm cluster-randomized controlled trial with 600 AI infants and toddlers from six AI communities in the Northwest. Tribal representatives provide input to both study design and implementation through their participation as project staff and interventionists. This community-wide intervention, although carefully defined in terms of goals and strategies, is tailored to fit the needs of each tribal community. Community health workers from the tribal communities deliver the interventions to ensure that interventions are translated in a manner consistent with tribal values. The interventions are designed to reduce body mass index (BMI) by altering feeding practices. The components of the intervention are (1) promoting breastfeeding, (2) reducing sugar-sweetened beverage consumption, (3) delaying solid food introduction, and (4) guiding introduction of type of food. Another component is to influence parenting to reduce sedentary lifestyles: (1) limiting introduction of television viewing, (2) encouraging development of motor skills, and (3) creating safe play opportunities. The family-level interventions target feeding and activity behavior within families, and the community-wide intervention has an added family component to alter the target behaviors. Children in the three comparison communities receive dental exams. Assessment is conducted at baseline and at 6-month intervals after intervention initiation for a total of 24–30 months. The primary outcome variable is child BMI z-scores at ages 24–30 months, adjusted for potential differences in BMI at baseline between communities. Secondary outcome variables include child BMI at 0, 6, 12, and 18 months of age; breastfeeding initiation and duration rates; and parental knowledge, attitudes, beliefs, and practices about child feeding and physical activity.

The second contribution, Healthy Children, Strong Families, describes a two-arm family-randomized controlled trial of a home-based healthy lifestyle intervention for 150 AI families with children ages 2–5 years in four Wisconsin tribal communities. Members from government bodies of the three initial participating communities met with the researchers to discuss ideas for the intervention, which included: involving parents, using an intergenerational model, focusing on children, educating about nutrition, role modeling, forming partnerships between organizations, and incorporating family events. Ongoing meetings ensued, and key community personnel play important roles in project promotion, subject recruitment, data collection, and dissemination of results. The trial also includes a community advisory board. The comparison group receives a toolkit of lifestyle lessons via the mail and a monthly newsletter with relevant topics, such as healthy eating. The intensive intervention group receives mentored home visits by trained AI community mentors in the first year of the intervention, relating to healthy nutrition and physical activity; in the second year of the intervention, families also attend monthly group meetings, at which activities promoting healthy nutrition and physical activity occur. This family-based intervention utilizes the caregiver as the primary mediator of change for the family, but the index case for assessment is the child. Assessment is conducted at baseline and at 12 and 24 months post-randomization. The primary outcome variable is child BMI z-score at 24 months post-intervention; caregiver BMI is a contingent primary outcome variable if the intervention is effective. Secondary outcome variables include health-related behaviors: fruit/vegetable consumption, sweetened beverage and candy consumption, hours of TV viewing, sedentary time for children, moderate physical activity for children and caregivers, and caregiver improvements in mean levels of biochemical markers.

The third contribution, A Cardiovascular Risk Reduction Program for American Indians with Metabolic Syndrome: The Balance Study, describes a two-arm individual-randomized controlled trial enrolling 200 AI adults with metabolic syndrome from seven major tribes in Southwest Oklahoma. The Balance Study was initiated by the researchers and developed through collaboration with community groups including the seven tribes, the IHS clinics and Lawton Hospital, the IHS regional Institutional Review Board, and Native community members who helped develop the intervention components. The intervention is a holistic lifestyle modification program with four components: cultural/spiritual, emotional/motivational, nutritional, and physical fitness and exercise. The comparison group is self-managed; participants receive only the CVD prevention materials in written form. The intensive intervention group is guided; participants receive the written materials and also attend intervention sessions, which include discussions on AI culture and health, mental and physical health, nutritional information and healthy cooking demonstrations, and physical activities, such as walking, chair exercises, and AI dancing. Assessment is conducted at baseline and at 6-month intervals after intervention initiation for a total of 24 months. The primary outcome variable is weight; secondary outcome variables include level of physical activity, consumption of saturated fat and sodium, status of metabolic syndrome, and incident CVD.

The fourth contribution, Project həli?dx w /Healthy Hearts Across Generations, describes a two-arm individual-randomized controlled trial enrolling 120 parents at risk for CVD, who live on a Northwest reservation. The goal is to evaluate the effectiveness of a 12-month motivational CVD risk reduction intervention. Early in the planning stages, the university and tribal research teams met with tribal health clinic personnel, met with tribal elders, and attended community functions to identify relevant content, design, and process, and to identify cultural concerns and challenges. Themes that emerged included the importance of developing a non-stigmatizing intervention with a focus on parents to promote intergenerational wellness and future generations’ health; moreover, the community wanted a second intervention to be tested as a comparison condition, rather than using a concurrent no-treatment control or wait-list control condition. The proposed intervention for both conditions contains the AI-related domains of mind, body, spirit, and emotion. Both groups begin with a seminar and then have individual coaches and receive motivational interviewing (Rollnick, Miller, & Butler, 2008). Participants in the CVD prevention intervention group receive accelerometers, Rez Chef cooking classes, exercise/movement classes, and meditation classes. Participants in the time-and-attention comparison condition receive a family life skills intervention that involves approximately the same amount of contact as the intervention arm with subject matter focusing on increasing family communication skills, parenting skills, and family connectedness. Assessment is conducted at baseline and at 4, 12, and 18 months. The primary outcome variable is BMI. Secondary outcome variables include blood pressure, cholesterol, inflammation, CVD-related behavioral risk factors (food habits, physical activity, sedentary and passive activities), and other indices of psychosocial functioning including trauma, mental health, and alcohol, tobacco, and other drug use.

The fifth contribution, The Lakota Oyate Wicozani Pi Kte (literally “moving toward a healthier Lakota population”), or LOWPK, describes a two-arm individual-randomized controlled trial, enrolling 180 adults with type 2 diabetes and dyslipidemia, hypertension, or both. Participants must reside on or near the Cheyenne River Sioux Reservation in South Dakota and be eligible to receive medical care on the reservation. The local IHS hospital and the tribally owned telephone authority are partners in the project. The project was approved by the tribe and by the regional IHS Institutional Review Board. Focus groups were held with tribal health care providers to solicit their input into the design of the trial, followed by additional focus groups to solicit their input into the design suggestions that they made during their initial focus group. The goal of the trial is to determine whether a web-based diabetes and nutritional intervention, which has previously been found to improve diabetes control in disadvantaged inner-city populations (Goldberg, Lessler, Mertens, Eytan, & Cheadle, 2004), can significantly reduce the risk of CVD in this population. All participants received a computer, computer training, and internet access to health websites. Participants in the intervention group also received individual case management and access to a special diabetes management website. Assessment is conducted at baseline and at 6-month intervals for a total of 18 months. The primary endpoint is change in glycosylated hemoglobin level at 18 months. Secondary endpoints include changes in cholesterol, blood pressure, BMI, and smoking status. A cost-effectiveness component of the trial will describe the cost-effectiveness of the intervention. LOWPK may serve as a guide for future chronic disease intervention trials in remote, technologically challenged settings.

Discussion

The five studies described here share a number of features, but they also have some important differences (see Table 1). Although they are all randomized controlled trials, three have individual randomization, one has family randomization, and one has community randomization. One trial is focused on infants and toddlers, one on preschool-aged children and three on adults. All five trials feature strong community partnerships, culturally tailored interventions, and a comparison group that receives benefits from participating in the study, consistent with community standards. All five trials have biological primary outcome variables to measure changes, as required by the RFA; two trials use BMI z-score, one uses BMI, one uses weight, and one uses glycosylated hemoglobin level. Four of the five trials also have biological secondary outcome measures. The interventions in all five trials are conducted by AI staff members; two of the five principal investigators, several co-investigators, and the funding institution’s project scientist are themselves AI.

The high burden of disease among AI/AN populations will worsen unless behaviors and lifestyles affecting CVD risk are changed. Widespread concern exists among AI/AN communities and health organizations that, despite extensive documentation of these problems in observational studies, few rigorous randomized controlled trials have been initiated to test culturally relevant interventions in their unique settings. Tribal leaders and AI/AN health experts have urged that health research in their communities focus on the most serious health issues they face, and they have identified CVD, obesity, and diabetes as being among these issues. These five trials, if successful, represent a start in addressing these significant health disparities related to CVD, along with the IHS CVD prevention intensive case-management demonstration project (Manson et al., 2011). However, much more remains to be done in the AI/AN population.