Participant Religiosity in a Novel Diabetes Prevention Program: Preliminary Findings
Peter Moreau, MS, 4th year medical student (student investigator), Virginia Tech Carilion School of Medicine
BACKGROUND: Recently, the field of religion and health has gained the attention of scientific researchers. However, there is a paucity of research investigating the relationship between religious behavior and diabetes, especially diabetes prevention. Based on research showing associations between religious behaviors and obesity, it is plausible that religious behaviors play a role in participant performance in diabetes prevention programs.
OBJECTIVES: This study will explore how participant religiosity is related to performance in a novel diabetes prevention program clinical trial.
METHODS: This study will have a prospective cohort design and will use the 5-question Duke University Religiosity Index (DUREL) to measure the self-reported religiosity of participants in the DiaBEAT-it clinical trial. DiaBEAT-it will explore the reach and effectiveness of two technology-centered lifestyle interventions in regards weight loss, physical activity, and eating habits in those at risk for developing diabetes. DiaBEAT-it has 5 treatment arms and will enroll 600 individuals with prediabetes over the age of 18 in the Roanoke Valley area. In this religiosity sub-study, religiosity measured at baseline with the DUREL will be compared to weight loss over time—primarily 6 months—in DiaBEAT-it participants. DUREL measures three distinct aspects of religiosity: organizational religious activity (ORA), non-organizational religious activity (NORA), and intrinsic religiosity (IR).
PRELIMINARY RESULTS: To date, 90% (n=339) of participants have completed the DUREL questionnaire at baseline, and 50% of eligible participants have returned for 6 month follow up (n=62). Participants in this study tend to have high self-reported religiosity. Using regression analysis and controlling for several demographic variables, none of the measured aspects of religiosity—ORA, NORA, and IR— significantly predict the baseline weight of participants (p=0.21, 0.32, and 0.74 respectively). Additionally, there is no strong collinearity among the measures of religiosity and other demographic information (all tolerance >0.5). However, correlation analysis, ANOVA, and post-hoc analysis revealed several statistically significant relationships between religiosity categories and demographic information; participants with high ORA, NORA, and IR values tend to be female (p=0.01, 0.00, and 0.00 respectively), and older (p=0.01, 0.03, and 0.02 respectively). Those with highest ORA and NORA tend to be married (p=0.00 and 0.04 respectively). However, the Pearson correlation values for the overall relationship between religiosity and these variables are relatively low (all <0.22).
The 6-month average % weight change for participants was -2.37%, indicating a loss of weight. regression analysis revealed that the religiosity measures do not significantly predict 6-month % weight change (p=0.74, 0.67, and 0.61 respectively).
CONCLUSIONS: Overall, participants in this study have a high self-reported religiosity. However, higher levels of religiosity are not related to better or poorer outcomes in this diabetes prevention trial. it should be noted that the sample size for the 6 month analysis was relatively small, and the analysis could change as more participants complete six months of intervention. Regardless, the large number of religious participants may necessitate the development of a religious-based intervention for this region.
BACKGROUND: Recently, the field of religion and health has gained the attention of scientific researchers. However, there is a paucity of research investigating the relationship between religious behavior and diabetes, especially diabetes prevention. Based on research showing associations between religious behaviors and obesity, it is plausible that religious behaviors play a role in participant performance in diabetes prevention programs.
OBJECTIVES: This study will explore how participant religiosity is related to performance in a novel diabetes prevention program clinical trial.
METHODS: This study will have a prospective cohort design and will use the 5-question Duke University Religiosity Index (DUREL) to measure the self-reported religiosity of participants in the DiaBEAT-it clinical trial. DiaBEAT-it will explore the reach and effectiveness of two technology-centered lifestyle interventions in regards weight loss, physical activity, and eating habits in those at risk for developing diabetes. DiaBEAT-it has 5 treatment arms and will enroll 600 individuals with prediabetes over the age of 18 in the Roanoke Valley area. In this religiosity sub-study, religiosity measured at baseline with the DUREL will be compared to weight loss over time—primarily 6 months—in DiaBEAT-it participants. DUREL measures three distinct aspects of religiosity: organizational religious activity (ORA), non-organizational religious activity (NORA), and intrinsic religiosity (IR).
PRELIMINARY RESULTS: To date, 90% (n=339) of participants have completed the DUREL questionnaire at baseline, and 50% of eligible participants have returned for 6 month follow up (n=62). Participants in this study tend to have high self-reported religiosity. Using regression analysis and controlling for several demographic variables, none of the measured aspects of religiosity—ORA, NORA, and IR— significantly predict the baseline weight of participants (p=0.21, 0.32, and 0.74 respectively). Additionally, there is no strong collinearity among the measures of religiosity and other demographic information (all tolerance >0.5). However, correlation analysis, ANOVA, and post-hoc analysis revealed several statistically significant relationships between religiosity categories and demographic information; participants with high ORA, NORA, and IR values tend to be female (p=0.01, 0.00, and 0.00 respectively), and older (p=0.01, 0.03, and 0.02 respectively). Those with highest ORA and NORA tend to be married (p=0.00 and 0.04 respectively). However, the Pearson correlation values for the overall relationship between religiosity and these variables are relatively low (all <0.22).
The 6-month average % weight change for participants was -2.37%, indicating a loss of weight. regression analysis revealed that the religiosity measures do not significantly predict 6-month % weight change (p=0.74, 0.67, and 0.61 respectively).
CONCLUSIONS: Overall, participants in this study have a high self-reported religiosity. However, higher levels of religiosity are not related to better or poorer outcomes in this diabetes prevention trial. it should be noted that the sample size for the 6 month analysis was relatively small, and the analysis could change as more participants complete six months of intervention. Regardless, the large number of religious participants may necessitate the development of a religious-based intervention for this region.