Examination of Cluster Groups of Risk Behaviors and Beliefs Associated with Non-Communicable Diseases with Latent Class Analysis: A Cross-Sectional Study in Rural Bangladesh

Abstract

This cross-sectional observational study examined the cluster groups of risk behaviors and beliefs associated with non-communicable diseases (NCDs) and the demographic factors that influence these cluster groups. The questionnaire survey was conducted in Lohagara Upazila in Narail District, Bangladesh and included basic demographics and items associated with NCDs. The inclusion criteria for the participants in this study included those who were aged between 20 and 80 years and both sexes. The survey items were based on risk behavior, belief, and improvement behavior. To identify the several cluster groups based on NCD-related behavior and belief patterns, a log-likelihood latent class analysis was conducted. Then, a multinomial regression analysis was performed to identify the factor associated with each cluster group. Of the 600 participants, 231 (38.5%) had hypertension, 87 (14.5%) had diabetes, and 209 (34.8%) had a body mass index of 25 or more. Finally, risk behaviors and beliefs associated with NCDs were classified into three cluster groups: (1) very high-risk group (n = 58); (2) high-risk group (n = 270); and (3) moderate-risk group (n = 272). The very high-risk group was significantly associated with female gender, older age, fewer years spent in education, and the absence of daily medication compared to the moderate-risk group. Educational interventions in rural Bangladesh should be immediately implemented to improve the risk behaviors and beliefs associated with NCDs.