Composite Lifestyle Patterns and Perceived Resilience in Young Adults: A Cross-Sectional Analysis

Introduction: Psychological resilience is shaped by behavioral and social factors, thus far most studies have evaluated lifestyle habits in isolation rather than as coordinated patterns. However, it remains unclear whether resilience relates to individual behaviors or to wider behavioral organization. This study examines associations between demographic characteristics, lifestyle behaviors, and perceived resilience, and to test whether clustered behavioral patterns differentiate resilience levels. Methods: A cross-sectional anonymous online survey was completed by 440 young adults, predominantly university students (mean age, 21.96 ± 6.95 years; 90.0% students). Resilience was measured using the Brief Resilience Scale. Lifestyle variables included sleep duration, diet quality, substance use, and physical activity. Ordinal logistic regression was used to evaluate independent predictors of resilience level (Low, Normal, High). Principal component analysis (PCA) identified higher-order behavioral patterns, and group differences were tested using nonparametric comparisons. Results: Higher resilience was associated with male gender (OR = 2.48, p < 0.001) and doctoral degree level (OR = 4.90, p = 0.018); transgender-male participants showed an elevated but non-significant association (OR = 3.12, p = 0.102), and physical activity was not independently associated with resilience after adjustment for other lifestyle behaviors (OR = 1.39, p = 0.127). PCA identified two components representing distinct patterns of covariance among the measured lifestyle indicators (diet category, sleep duration, exercise status, and substance use). Component scores were subsequently compared across resilience categories. PC2 represented a general lifestyle pattern dominated by diet quality and exercise, whereas PC1 reflected a distinct pattern characterized primarily by exercise status, sleep duration, and substance use. PC1 did not differ significantly across resilience levels (p = 0.369). In contrast, PC2 differed significantly and increased monotonically across resilience categories (p = 0.001; η2 = 0.026). Post hoc testing identified significant differences between the Low and Normal resilience groups and between the Low and High resilience groups, but not between the Normal and High resilience groups. Conclusions: Resilience in young adults was associated with a difference in a composite lifestyle pattern, primarily reflecting diet quality and physical activity, rather than with individual lifestyle behaviors considered separately. These findings suggest that the measured lifestyle indicators may exhibit different patterns of covariation across resilience categories, and that the specific behavioral combination associated with resilience may be sensitive to the composition of the analytic sample. However, the cross-sectional design does not establish the direction or mechanisms of this association.