Latent class analysis of stages of change for multiple health behaviors: results from the Special Diabetes Program for Indians Diabetes Prevention Program.

This study sought to identify latent subgroups among American Indian and Alaska Native (AI/AN) patients with pre-diabetes based on their stages of change for multiple health behaviors. We analyzed baseline data from participants of the Special Diabetes Program for Indians Diabetes Prevention (SDPI-DP) Program, a lifestyle intervention program to prevent diabetes among AI/ANs. A total of 3,135 participants completed baseline questionnaires assessing stages of change for multiple health behaviors, specifically exercise, healthy eating, and weight loss. Latent class analysis was used to identify subgroups of people based on their answers to stages of change questions. Covariates were added to the latent class analyses to investigate how class membership was related to sociodemographic, behavioral, and psychosocial factors. Three classes were identified based on the distributions of the stages of change variables: Contemplation, Preparation, and Action/Maintenance classes. Male and retired participants were more likely to be in more advanced stages. Those who exercised more, ate healthier diets, and weighed less were significantly more likely to be in the Action/Maintenance class. Further, the participants who had higher self-efficacy, stronger family support, and better health-related quality of life had higher odds of being in the Action/Maintenance class. In conclusion, we found that stages of change for multiple behaviors can be summarized by a three-class model in this sample. Investigating the relationships between latent classes and intervention outcomes represents important next steps to extend the findings of the current study.