Background: Type 2 diabetes mellitus (T2DM) is a significant comorbidity of psoriasis that exerts a long-term detrimental effect on patients. Early T2DM detection in psoriasis patients is essential for timely medical intervention and the prevention of further complications.
Objective: To develop a diagnostic model and introduce a web-based calculator for assessing the risk of T2DM in patients with psoriasis.
Methods: A diagnostic model was constructed using the least absolute shrinkage and selection operator and logistic regression on data from 17,124 patients with psoriasis from 49 hospitals. The model's performance was evaluated using receiver operating characteristic curve, decision curve analysis, and clinical impact curve.
Results: Eleven significant independent predictors were identified in our model. The area under the receiver operating characteristic curve of this model was 0.875 (95% confidence interval: 0.862-0.889) and the result of Hosmer-Lemeshow goodness of fit test showed a p-value of 0.06. Within the threshold probability of 0-0.2, the diagnostic model showed a significant net benefit.
Limitations: The limitations of retrospective studies, including a potential data bias to arise during pandemics, must be considered.
Conclusion: A multifactorial diagnostic model for diagnosing T2DM in patients with psoriasis was constructed to help clinicians make better decisions.
Keywords: Decision curve analysis; Diagnostic Model; Psoriasis; Type 2 Diabetes Mellitus; Web-based calculator.
Copyright © 2024. Published by Elsevier Inc.