Depression-Related Nutritional Risk and Physical Performance in Middle- to Older-Aged Adults: Cross-Sectional Secondary Analysis of Tree-Based and Regression Approaches.

Background: Depressive symptoms are linked to nutritional vulnerability and functional decline in aging populations, but their relationships with nutritional risk and lower-extremity physical performance are often examined separately. Tree-based exploratory approaches may provide transparent subgroup characterization. Objective: This study aimed to examine the cross-sectional associations of depressive symptoms with nutritional risk and lower-extremity physical performance among community-dwelling middle- to older-aged adults and to compare a tree-based exploratory approach with regression-based methods for characterizing depression-related patterns within the study sample. Methods: This cross-sectional secondary analysis included 1010 community-dwelling adults aged ≥50 years recruited from 1 hospital and 3 community centers in northern Taiwan. Depressive symptoms were assessed using the Geriatric Depression Scale-Short Form (GDS-15). Nutritional status was measured using the Mini Nutritional Assessment-Short Form (MNA-SF), and physical performance was evaluated using the Short Physical Performance Battery (SPPB), gait speed, and the timed up and go (TUG) test. Sociodemographic, lifestyle, and clinical variables were included as covariates. Outcomes were dichotomized using established clinical cutoffs. χ² automatic interaction detection (CHAID) decision trees were used as a tree-based exploratory approach for subgroup characterization. Multivariable logistic regression and least absolute shrinkage and selection operator (LASSO) logistic regression models were used for comparison. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and classification accuracy. Results: Of the 1010 participants, 143 (14.2%) screened positive for depressive symptoms (GDS-15 ≥5). Compared with participants without depressive symptoms, those with depressive symptoms had poorer nutritional status and lower-extremity physical performance, including a higher prevalence of nutritional risk (53/143, 37.1% vs 105/867, 12.1%; P<.001), SPPB impairment (53/143, 37.1% vs 118/867, 13.6%; P<.001), gait speed impairment (73/143, 51% vs 270/865, 31.2%; P<.001), and TUG impairment (25/143, 17.5% vs 55/864, 6.4%; P<.001). Across CHAID models, GDS-15 score was consistently selected as the primary splitting variable, while BMI, calf circumference, age, education level, and comorbidity severity provided additional hierarchical refinement of subgroup patterns. In comparative model analyses, LASSO logistic regression analysis had the highest classification performance for the MNA-SF, SPPB, and gait speed outcomes, whereas CHAID provided transparent, rule-based subgroup characterization with acceptable within-sample classification performance. Model performance for TUG was less consistent across approaches. Conclusions: In this community-based sample of adults aged 50 years and older, depressive symptom severity was associated with nutritional vulnerability and poorer lower-extremity physical performance. CHAID identified hierarchical subgroup patterns linking depressive symptoms, including subthreshold levels, with nutritional and functional vulnerability. Although LASSO logistic regression analysis had higher classification performance, CHAID offered transparent, rule-based subgroup characterization. These findings support the relevance of integrating depression screening with nutritional and functional assessment in prevention-oriented geriatric care, although further validation is needed.