Comparing two programs for weight loss among Marshallese adults living in the United States / Pearl A. McElfish.

Format:

Book

Author/Creator:

McElfish, Pearl A., author.

Language:

English

Subjects (All):

Obesity--Treatment.

Obesity.

Weight loss--Psychological aspects.

Weight loss.

Physical Description:

1 online resource (73 pages) : illustrations

Place of Publication:

Washington, D.C. : Patient-Centered Outcomes Research Institute (PCORI), 2021.

Summary:

BACKGROUND: Marshallese Pacific Islander adults experience high rates of overweight, obesity, prediabetes, and type 2 diabetes mellitus (T2DM). Community engagement efforts identified T2DM prevention and improving health behaviors as a top health concern. Using a community-based research approach, we engaged community stakeholders to help design and carry out a cluster randomized controlled trial (cRCT) designed to reduce body weight as an approach to preventing T2DM. OBJECTIVES: The primary objective was to compare the effectiveness of 2 Diabetes Prevention Program (DPP) interventions on reducing body weight. METHODS: We conducted a 2-arm cRCT comparing a faith-based DPP curriculum with a Pacific culture-focused DPP curriculum. The trial was conducted from January 2018 to May 2020. We randomized 31 churches and 380 overweight or obese Marshallese adults living in the United States to receive (1) the Wholeness, Oneness, Righteousness, Deliverance (WORD) DPP (n = 16 churches; n = 213 participants) or (2) the Partnership for Improving Lifestyle Intervention (PILI) DPP (n = 15 churches; n = 167 participants). The WORD DPP included 16 sessions and focused on using one's faith to make healthy lifestyle changes. The PILI DPP included 14 sessions and focused on engaging one's social and family networks to make healthy lifestyle changes. Both interventions were conducted over 24 weeks in group settings in churches in Arkansas and Oklahoma. Outcomes were measured over the course of 1 year: at preintervention (baseline), at 6 months (immediate postintervention), and at 12 months (6 months postintervention). The primary outcome of interest was change in body weight. Secondary outcomes included hemoglobin A1c (HbA1c), blood pressure, physical activity, dietary intake, self-efficacy, and family support. Outcomes were examined using descriptive statistics, t tests, and general linear mixed-effects regression models, adjusting for sociodemographic covariates and accounting for clustering of participants within churches. RESULTS: The mean age of participants was 42.3 years ((+/-) 11.6), and slightly more than half were female (56.6%). Unadjusted and adjusted reductions in weight were small and not clinically significant for both groups. In adjusted models, participants in the PILI DPP arm lost 0.52 kg, and participants in the WORD DPP arm lost 0.26 kg at 6 months postintervention. Adjusted linear mixed-effects regression models showed no significant differences in weight loss between the 2 arms at immediate postintervention (P = .3599) or at 6 months postintervention (P = .3207). No significant differences in changes in HbA1c between the 2 arms were found immediately postintervention or at 6 months postintervention. Significant differences in systolic and diastolic blood pressure favoring the PILI DPP arm were found between the 2 arms immediately postintervention (P = .0293 and P = .0068, respectively) but not at 6 months postintervention. No significant differences were found between the 2 arms for physical activity, dietary intake, self-efficacy, or family support. CONCLUSIONS: There were no significant differences in weight reductions between the PILI DPP arm and the WORD DPP arm. These findings do not support our hypothesis that the PILI DPP would result in greater weight loss compared with the WORD DPP. Future research is needed to identify chronic disease primary and secondary prevention interventions that can be successful for this at-risk population. LIMITATIONS: The trial was conducted with Marshallese adults in Arkansas and Oklahoma, thereby limiting generalizability of the findings to other populations or regions.

Contents:

ABSTRACT 5

BACKGROUND 7

Significant Health Disparities 7

Historical Trauma 8

Underrepresentation in Research 9

Research Questions 10

Significant/Potential Impact 10

Overview of Study Goals 11

Specific Aims 11

PARTICIPATION OF PATIENTS AND OTHER STAKEHOLDERS 13

Choosing Appropriate Interventions 13

Collaborative Study Design 14

Cultural Adaptations to Data-Collection Instruments 15

Recruitment and Retention Plans 16

Importance of Language 17

METHODS 18

Study Overview 18

Study Setting 18

Participants 19

Interventions 21

Table 1 WORD DPP and PILI DPP Session Topics 22

Study Outcomes 23

Table 2 Overview of Patient-Centered Outcomes and Data Collection Instruments 25

Sample Size and Power Calculations 26

Post Hoc Detectable Effect Calculations 27

Time Frame for the Study 27

Figure 1 Full Study Timeline 28

Data Collection and Sources 29

Analytical and Statistical Approaches 29

Missing Data 31

Heterogeneity of Treatment Effect 31

Changes to the Original Study Protocol 32

RESULTS 33

DISCUSSION 51

Summary of Results 51

Results in Context 52

Study Limitations 53

Future Research 54

CONCLUSIONS 55

REFERENCES 56

RELATED PUBLICATIONS 67

ACKNOWLEDGMENTS 68

APPENDICES 69

Appendix A Participant Characteristics by Missing Primary Outcome Status at

Immediate Postintervention and 6 Months Postintervention 69

Appendix B Unadjusted and Adjusted Changes in Weight From Baseline to

Postintervention Among Completers Only 70

Appendix C Unadjusted and Adjusted Changes in Weight From Baseline to 6 Months

Postintervention Among Completers Only 71

Appendix D Estimated Means for Primary Outcome by Study Arm and Time With

Between-Arm Intervention Effects From Multiple Imputation of Missing Data Under

Missing-at-Random Assumption 72.

Notes:

Description based on publisher supplied metadata and other sources.