A latent profile analysis of self-management behavior among patients after metabolic bariatric surgery

TL;DR

This study identifies different levels of self-management behavior in patients after bariatric surgery and finds that most patients fall into a moderate group.

Contribution

The study introduces a latent profile analysis to categorize self-management behaviors post-metabolic bariatric surgery.

Findings

Three distinct self-management behavior groups were identified: high, moderate, and low.

Age and educational level significantly differ among the self-management behavior groups.

Employment status, smoking, and drinking history predict self-management behavior.

Abstract

Obesity has emerged as a global public health epidemic with far-reaching health consequences. While metabolic bariatric surgery (MBS) is an established therapeutic modality for moderate-to-severe obesity and associated metabolic disorders, enabling rapid weight reduction and metabolic improvement, postoperative weight regain remains a critical barrier to sustaining long-term treatment efficacy. Indeed, the durability of surgical outcomes is heavily contingent upon patients’ ability to engage in sustained self-management behaviors.

To characterize the patterns of self-management behavior among patients after metabolic bariatric surgery using latent profile analysis, and to examine the relationships among these latent profiles.

A cross-sectional study was carried out at one general hospital. A total of 242 patients after metabolic bariatric surgery completed the socio-demographic questionnaire, Bariatric Surgery Self-Management Questionnaire, General Self-efficacy Scale, and International Physical Activity Questionnaire.

Three latent profiles were identified: high self-management behavior group (n = 28, 11.57%), moderate self-management behavior group (n = 156, 64.46%) and low self-management behavior group (n = 58, 23.97%). The ANOVA and chi-square tests demonstrated significant differences among three groups concerning age and educational level. Multinomial logistic regression analysis indicated that employment condition, smoking and drinking history significantly predicted self-management behavior.

The statistical analysis indicated that the majority of patients fall into the moderate self-management group. Further regression analysis demonstrated significant associations between self-management proficiency and both age and educational level. These findings highlight the need for tailored interventions targeting specific patient profiles.

Figures