A Latent Class Analysis of Planetary Health Diet Index-informed Dietary Patterns Association with Overweight and Obesity Among Middle-Aged and Older Women: A Cross-Sectional Study

Abstract Background Middle-aged and older women face a heightened risk of overweight and obesity driven by physiological and social transitions. The Planetary Health Diet Index (PHDI) quantifies adherence to sustainable, health-promoting eating guidelines across key plant and animal food groups, yet how specific PHDI dietary behavior combinations cluster and influence weight in this population is poorly understood. This study used latent class analysis (LCA) based on nine PHDI-informed dietary behavior indicators to identify distinct dietary patterns in this population and examine their associations with overweight and obesity. Methods This cross-sectional study included 21,856 middle-aged and older women who underwent health examinations at a tertiary hospital in southern China between 2015 and 2023. Sociodemographic characteristics, anthropometric measurements, medical history, and nine PHDI-informed dietary behavior indicators were extracted. LCA was applied to identify dietary patterns, and multivariable binary logistic regression was used to evaluate their associations with overweight/obesity. Age-stratified subgroup and sensitivity analyses excluding participants with baseline hypertension or diabetes were conducted to evaluate robustness. Results The prevalence of combined overweight and obesity was 31.94%. LCA identified four dietary patterns: High PHDI-Adherent (19.07%), Grain-Dominant (21.02%), Protein-Focused (26.93%), and Low PHDI-Adherent (32.98%). Compared with the High PHDI-Adherent pattern, the Protein-Focused pattern (AOR = 1.216, 95% CI: 1.114–1.328) and Low PHDI-Adherent pattern (AOR = 1.236, 95% CI: 1.133–1.349) were statistically significantly associated with higher odds of overweight/obesity, whereas the Grain-Dominant pattern showed no significant association (AOR = 1.046, 95% CI: 0.953–1.149). Older age, lower education, postmenopausal status, cardiovascular family history, smoking, alcohol consumption, and low physical activity were independent risk factors. Further sensitivity analyses confirmed the stability of these associations. Conclusions Among the identified dietary patterns, the Protein-Focused and Low PHDI-Adherent patterns were associated with higher odds of overweight and obesity among middle-aged and older women in southern China. These findings highlight the importance of overall dietary balance and provide an empirical basis for developing tailored dietary and weight management strategies for distinct population subgroups.