Health Maintenance Consortium
Supplemental Collaboration Grants

     The Office of Behavioral and Social Sciences Research is pleased to announce that seven supplements will be awarded in response to our HMC Collaboration Grant Announcement.  We received outstanding proposals, representing truly innovative thinking. We believe that the completion of these studies will do much to advance the state of the art.  Because all of the seven proposals were given high marks by the reviewers, OBSSR chose to come up with additional funding to make it possible to award the entire group of applicants.  Congratulations to all!

 

Adherence Trajectories Across Behaviors & Associated Impact on Quality of Life

Susan Hughes, DSW, Principal Investigator, Marcia Ory, Ph.D., MPH, Diane Elliot, MD, Robert Friedman, MD, Robert Klesges, PhD, Brian Martinson, PhD, Kimberly Kirby, PhD, John Roll, PhD, Geoffrey Williams, MD, PhD

     The effectiveness of specific behavioral change interventions on the reduction of morbidity, mortality, and health care costs across different illnesses and conditions is becoming recognized (CDC, 2003; Ory, 2005).  While some attention has been paid to multiple behavior research (Nigg, Allegrante, & Ory 2002; Persky et al, 2005) for the most part previous research has focused on the efficacy of interventions in changing a single behavior (e.g., smoking or physical activity or nutrition or substance abuse), despite the fact that individuals often present with multiple competing risk factors (Goldstein, 2003).  As a result, little is currently known about the trajectory of adherence across multiple behaviors.  Thus, this proposed supplement presents a unique opportunity to advance the current state of knowledge in this important area. Data from multiple sites participating in the Health Maintenance Consortium provide an important opportunity to examine the research gaps.  Addressing these issues can enable us to understand how to prioritize the selection of behavioral targets, how to design the most effective interventions, and how to bolster behavior change over initial and sustained periods of adherence to maximize the impact of sustained adherence on important common outcomes across settings and populations. This proposal seeks supplemental funding to pool adherence data across eight funded maintenance of behavior change studies in order to address three main research objectives: (1) to examine the degree of variability that occurs across behaviors in the way that adherence is assessed, (2) to assess the degree of variability in adherence across behaviors, and (3) to determine the level of adherence necessary across behaviors to impact a common outcome. This proposal will encompass 7,849 participants at eight sites which are examining four behaviors, specifically: physical activity, diet/nutrition, smoking cessation, and drug abstinence.

 

The Impact of the Physical and Socioeconomic Environment on
Changes in Diet, Physical Activity, and Obesity

Jeffrey Migneault, Ph.D., Les Boden, Ph.D., Russ Lopez, Amy Gorin, Ph.D., Deborah Toobert, Ph.D.,
Brian Martinson, Ph.D., Susan Hughes, DSW, Paul Williams, Ph.D., Kenneth McLeroy, Ph.D.,

     This supplement will form a collaboration of participating consortium members to investigate central scientific questions about the interaction of environment factors and health behavior interventions and to assist participating sites in the use of social, economic, and built environmental factors in their efforts promoting maintained behavior change to address critical health needs in the United States population. We will do this by involving a core set of Consortium projects targeting nutrition, physical activity, and obesity. These behaviors are ones that should be maintained for a life time and have strong relationships to disease prevention, disease management, and overall health status. A new but growing body of research has demonstrated that characteristics of neighborhoods can affect these health behaviors, but much of this phenomenon remains unexplored or specified. Additionally this project is innovative in extending this line of research to include how these environmental factors impact effectiveness of behavioral interventions.  Specific aims include:  (1) Geocode subject addresses provided by collaborating consortium members;  (2) Develop a database of neighborhood social, economic, and built environment factors that potentially affect health related risks and behaviors. Variables in the database include factors previously found to be associated with relevant health behaviors and additional factors suggested by consortium members. Data will be linked using geocoded information; (3) Develop a module of questions for inclusion in participating study questionnaires that assess the perception of neighborhood factors not available in national databases. Analyze the relationship between neighborhood level factors derived from databases and perceived neighborhood factors derived from data gathered through the questionnaire module; and (4) Test hypotheses on the impact of social and built environment factors on the baseline levels, the initiation, and the maintenance of healthy diet, physical activity, and weight management behaviors. We will test these hypotheses both within individual projects and across multiple projects.

 

Transbehavioral and Quality of Life Outcomes:  Cross-Site Analyses and Resources

Deborah Toobert, Ph.D., Russell Glasgow, Ph.D., Diane Elliot, MD, Geoffrey Williams, MD, Ph.D., Sue Hughes, DSW, John Roll, Ph.D., Marcia Ory, Ph.D.

     One of the reasons that behavior change research has not had more impact on societal decisions or generated wider adoption of behavioral interventions is that most research has been conducted in “silos” of a given topic area. Thus, most studies measure only the primary outcome related to their single risk factor intervention (i.e., smoking cessation, physical activity). This status of the literature makes it difficult for decision- and policymakers to evaluate or prioritize among the plethora of potential interventions for health promotion and disease prevention. Seldom are any common measures or ultimate impact indices, such as quality of life, collected to allow comparison across interventions (Tunis, 2003; Glasgow, 2004) and over time. Also, interventions on one behavioral risk factor seldom report data on impact on other outcomes — either positive generalization or negative unintended consequences (e.g., smokers who quit smoking gaining weight). This project will advance the science of behavior change and its evaluation by analyzing results using common metrics and relating interventions to outcomes and maintenance based on epidemiologic risk and patient functioning.  Fortunately, several key projects involved in the Health Maintenance Consortium (HMC) have collected such multiple, crosscutting measures, including behavior change across multiple behaviors and a common measure of quality of life (QoL). These data provide an opportunity for exploration and analyses related to important issues that can advance not only our understanding of outcomes and linkages, but also provide answers to bottom-line questions of central interest to clinicians, health and social service organizations with limited resources, and policymakers (Tunis, 2003; Glasgow in press). This project will advance the measurement and evaluation of behavioral intervention outcomes, investigate important linkages among these outcomes, and identify intervention and social-environmental factors influencing these outcomes.

Predictors of Long-Term Successful Weight Loss

Suzanne Phelan, Ph.D., Rena Wing, Ph.D., Amy Gorin, Ph.D., Michael Lowe, Ph.D.

     The major problem in obesity treatment is failure to maintain long-term weight loss.  To develop more effective maintenance interventions, it is important to learn more about the factors that determine long-term weight loss maintenance.  The purpose of the present study is to better understand the role of dispositional, environmental, and behavioral factors in determining long-term successful weight control across a variety of different treatment and patient populations. The study will: 1) identify the latent cluster of environmental, dispositional, and behavioral factors that best distinguish a group of Long-Term Successful Weight Losers from a group of overweight Unsuccessful Weight Losers; 2) examine whether the latent cluster of factors may be used to predict who, among those losing weight after 6 months of treatment, will remain successful at 12 and 18-month follow-ups; 3) compare the relative importance of each latent factor (i.e., environmental, dispositional, and behavioral) in predicting successful weight loss at 12 and 18-month follow-ups; and, 4) determine whether the same cluster of latent factors are associated with successful weight loss in Caucasian and African-American participants.  This study involves collaboration among three different studies that have agreed to include common measures of dispositional, behavioral, and environmental factors.  Results of this study will have important implications for the design of more effective weight loss interventions to improve long-term weight loss outcomes.

 

Identifying Components of HMC Interventions that Predict Outcomes

Jim McKay, Ph.D., Diane Elliot, MD, Geoffrey Williams, MD, Ph.D., Deborah Toobert, Ph.D., John Roll, Ph.D., Kim Kirby, Ph.D., Shirley Beresford, Ph.D., Brian Martinson, Ph.D., Barbara Rimer, DrPH, Steve Belle, Ph.D., Sara Czaja, Ph.D., Marcia Ory, Ph.D., Joseph Sharkey, Ph.D.  

     Multi-component interventions have become increasingly common approaches to initiating and maintaining positive change for many of the health problems addressed in the HMC (e.g., weight, nutrition, exercise, smoking, substance use, cancer screening) (Belle et al., 2003).  Although there is growing evidence that these approaches can be effective, less is known about the relative value of different components of the interventions.  If the most potent components can be identified, it may be possible to develop less costly and easier to implement versions of these interventions, without loss of effectiveness.  Recently, several investigators from a previously funded NIH multi-site initiative—the Resources for Enhancing Alzheimer’s Caregiver Health (REACH) program—developed a new technique for characterizing diverse multi-dimensional psychosocial and behavioral interventions across a set of common components and examining  the  relationships of the individual components  to study outcomes. These procedures enabled analyses that  identified individual intervention elements associated with outcomes, across interventions (Belle et al., 2003).  In the proposed supplement, a group of investigators from the HMC, with consultation by the REACH investigators involved in that cross-site project, are seeking funds to extend this new methodology to data that will be gathered as part of the HMC.  The work will carried out in collaboration with 8 other HMC projects, and consultants from the University of Pittsburgh and University of Miami, who developed the procedures with REACH investigators.  In year 1, the procedures will be implemented with data from three BCC studies.  In year 2, the procedures will be adjusted on the basis of year 1 results, and applied to four HMC sites that are testing interventions for substance abuse.  In year 3, the procedures will undergo further refinement, and will be applied to five other HMC sites that are testing interventions to increase physical activity, healthy diet, and cancer screening.  Products from the work will include 2-3 journal articles, several conference presentations, and training materials to facilitate application of the methods by other investigators.

 

The Role of Depression and Context in Health Behavior An Administrative Supplement to:  Preventing Suicide Risk Behaviors:  Long Term Change

Brooke P. Randell, DNSc, CS., Ralph J. DiClemente, PhD,
Deborah J. Toobert, PhD, Geoffrey C. Williams, PhD

     The purpose of this proposal is to apply an ecological framework of neighborhood effects on individual health.  In general models of individual health and behavioral change, depression often plays a key role.  It is well documented that neighborhood context influences individual health behavior but it is unclear what role depression may play in this relationship.  No study to date has explored the interplay between depression and long-term health change within this ecological framework. We posit that there is a link between health and depression. Does neighborhood context (neighborhood economic disadvantage and residential instability) strengthen or weaken the expected negative association between depression and health outcomes?  We posit that these facets of neighborhood will directly influence depression and health as well as interact with our posited negative relation between depression and healthy outcomes. In the proposed study we will explore the relationship between depression and health change, for a broad set of health outcomes – suicide risk, sexual practices, diet adherence, and smoking cessation – and specifically explore how the context in which an individual is embedded contributes to and moderates these effects. Specifically, we test for cross-sectional association and interaction between neighborhood ecological context, depression, and health outcomes and in a longitudinal framework, for the moderating effects of neighborhood on the effects of depression on change and maintenance across this set of specific health outcomes.  To address these questions we will combine data from four Health Maintenance Consortium sites representing an approximate sample size of 2400 that varies in age, sex, race and ethnicity, health outcome of interest, and broad representation of context.  In collaboration with each site, data from the US Census (indicators of disadvantage and instability) will be linked to the location of each individual in the pooled data base (from all 4 sites) at their baseline (entry into each study) and for changes in their location at exit of intervention, and at follow-up.  To address these questions we will employ both a hierarchical linear modeling approach and a structural equation approach. This proposal affords the unique and innovative opportunity to address the relationship among depression, neighborhood context and diverse outcomes. 

 

I-5-A-Day Collaboration Project

Diane Elliot, MD., Deborah Toobert, Ph.D., Shirley Beresford, Ph.D.

     Increasing intake of fruits and vegetables (F&V) is a U.S. health goal. However, little is known about the factors relating to and sustaining appropriate dietary behaviors. The I-5-a-Day Collaborative Project will apply recent statistical techniques to define the common patterns, mediators and theoretical mechanisms of achieving and maintaining health dietary behaviors. We will examine these issues in a collaborative project of three studies that involve longitudinal (50 to 84 months) follow-up of participants from controlled intervention trials that successfully increased F&V intake. Although the projects’ subjects are diverse, the three studies' theoretical underpinnings, data collection design and instruments share common features in the domains of outcomes; demographics; individual characteristics; family, coworker and group social influences; and features of the physical environment. We will harmonize and combine the centers' data and perform theory driven analyses to define and rank mediators of achieving and maintaining healthy levels of F&V intake; use longitudinal repeated measures data to assess durability/decay trajectories and evaluate the effects of hypothesized moderators (e.g., participant characteristics) and mediators (group cohesion, self-efficacy, worksite and other environmental characteristics); and define characteristics of individuals or groups of individuals, who sustain healthy F&V intake, to profile for whom and under what conditions the interventions appear most effective. The three sites are all located in the Pacific Northwest along Interstate 5, which will facilitate the collaboration, and our statistical consultants are recognized experts in these types of analyses, who are established collaborators and also have existing working relationships with the study centers. Importantly, in the context of the Health Maintenance Consortium, we also can share these statistical techniques and disseminate both our findings and methods to others studying health promotion behaviors. Anticipated products from this collaboration include presentation of the initial cross-sectional mediation analysis; an HMC workshop on mediation; linkage of the mediation and HMC websites; and a manuscript describing our cross-center mediation analyses, its findings and our conclusions concerning dietary behavior change.