The Physician Global Assessment as a Proxy for Photoaging: Validation in an Observational Real‐World Cohort Using SCINEXA as the Reference Standard

ABSTRACT Background Photoaging is a chronic skin condition caused by cumulative ultraviolet exposure superimposed on intrinsic aging. Although the SCore of INtrinsic and EXtrinsic skin Aging (SCINEXA) is a validated multidimensional instrument for assessing photoaging, the Physician Global Assessment (PGA) is a simpler clinician‐reported measure whose validity for photoaging has not been established. Objectives To evaluate the reliability, criterion validity and potential clinical utility of the PGA as an index assessment of photoaging severity, using SCINEXA as the reference standard. Methods In this retrospective exploratory cohort study, 175 adults undergoing routine skin cancer screening in the Netherlands between 2022 and 2024 underwent standardized digital skin imaging. The Physician Global Assessment (PGA) was evaluated as the index assessment and SCINEXA as the reference standard. Inter‐ and intra‐observer reliability were assessed in a separate substudy involving 64 facial image assessments from 16 individuals performed by three dermatologists. Logistic regression was used to identify independent SCINEXA predictors of a high PGA score (≥3), an exploratory threshold defined for this study. Results High PGA scores were associated with older age, male sex, smoking, lighter Fitzpatrick skin types and higher SCINEXA scores. The PGA demonstrated the highest inter‐observer agreement among the assessed photoaging parameters and good intra‐observer reliability. Four SCINEXA features independently predicted a high PGA score: sunburn freckles (odds ratio [OR]: 49.8; 95% confidence interval [CI]: 2.73–909.46), cutis rhomboidalis nuchae (OR: 14.84; 95% CI: 2.28–96.64), pseudoscars (OR: 14.21; 95% CI: 1.79–113.15) and poikiloderma (OR: 6.64; 95% CI: 1.08–40.91). A reduced model based on these variables achieved an accuracy of 96.0% and an area under the receiver operating characteristic curve of 0.985 (95% CI: 0.954–0.997). Limitations The retrospective exploratory design, relatively small reliability substudy and non‐blinded assessment of PGA and SCINEXA may have introduced bias. In addition, the PGA threshold and reduced prediction model were developed and evaluated in the same cohort without external validation. The predominantly Dutch occupational screening population may limit generalizability. Conclusions The PGA demonstrated good reproducibility and strong criterion validity against key SCINEXA features of chronic photodamage. Although it cannot replace a multidimensional reference instrument such as SCINEXA, the PGA may serve as a rapid proxy for overall photoaging severity in routine clinical practice and screening settings, pending prospective external validation.