Risk Factors and Nomogram Model for Poor Prognosis in Atrophic Scar Patients Treated With Fractional Laser Therapy: The Role of Nursing Care in a Retrospective Cohort Study

ABSTRACT Background Accurate prognosis prediction following fractional laser therapy for atrophic scars remains challenging in clinical practice. Current models predominantly focus on clinical parameters while neglecting the synergistic effects of psychosocial factors and nursing interventions, limiting their predictive accuracy and clinical applicability. Aims This study aimed to develop and validate an integrative nomogram incorporating biological, clinical, and nursing‐related variables to predict poor scar prognosis after fractional CO 2 laser therapy. Methods In this retrospective cohort study, 180 patients receiving laser therapy (2020–2024) were randomly allocated into training ( n = 125) and validation ( n = 55) cohorts. Multivariate logistic regression with LASSO penalty was employed to analyze clinical parameters (age, skin type, ECCA severity), comorbidities, and psychosocial factors (emotional care status, negative emotions). A nomogram was constructed using R software and validated through ROC analysis, bootstrap resampling, and decision curve analysis. Results Seven independent predictors were identified: advancing age (OR = 1.23/year, 95% CI: 1.11–1.38), combination skin type (OR = 12.46 vs. neutral), concurrent skin diseases (OR = 3.92), severe ECCA classification (OR = 12.72 vs. mild), laser energy (OR = 5.01 vs. high), the presence of negative emotions (OR = 10.98), and inadequate emotional care (protective OR = 0.12). The model demonstrated excellent discrimination (training AUC = 0.96, validation AUC = 0.93) and calibration accuracy (slope = 0.98, intercept = −0.03). Decision curve analysis confirmed clinical utility across 15%–35% risk thresholds. Conclusion This novel nomogram represents the first prognostic tool integrating biological susceptibility and nursing‐mediated resilience, providing clinicians with a validated instrument for personalized risk stratification and optimized care pathway design. The model's high accuracy and generalizability support its implementation in both clinical practice and research settings.