What is the methodological quality and predictive performance of prognostic prediction models for long-term outcomes...

The clinical course of psoriasis is characterised by marked heterogeneity. Despite the availability of standardised assessment instruments, long-term outcomes at the individual level-encompassing the durability of treatment response, the risk of relapse and the development of comorbidities-remain difficult to predict. To address this uncertainty, numerous prognostic prediction models have been developed; however, their methodological quality, predictive performance and clinical applicability have not yet been systematically appraised.

A systematic search will be conducted across six databases-China National Knowledge Infrastructure, Wanfang Data, VIP Chinese Journal Database, PubMed, Cochrane Library and Embase-from inception to 31 December 2025, using a search strategy combining controlled vocabulary and free-text terms related to psoriasis, prognostic prediction models and relevant outcome measures. Two reviewers will independently perform study selection, data extraction using the Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modelling Studies checklist and risk-of-bias assessment using the Prediction model Risk Of Bias ASsessment Tool. Model performance metrics, including discrimination (eg, the C-statistic) and calibration, will be systematically extracted. Where feasible, random-effects meta-analysis will be performed to pool discrimination estimates across included models. Prespecified subgroup analyses and meta-regression will be employed to investigate potential sources of heterogeneity.

Ethical approval is not required because this study will analyse publicly available de-identified data from published studies. The results will be submitted to a peer-reviewed journal and presented at relevant conferences.