Sensitivity, specificity and interrater reliability of a clinical staging model for personality disorders in older adults: A case series study

A recent Delphi study found expert-consensus that clinical staging (CS) could provide a valid framework in the assessment of personality disorders (PDs) in older adults (Conjaerts et al., 2025). The categorical models used currently to assess PDs are poorly suited for treatment selection, in older adults even more so as age-specific factors were not considered in the development. The aim of this study was to investigate the interrater reliability, sensitivity and specificity of the recently proposed CS model, using the level of personality functioning and the lifetime course of personality pathology as profilers. Clinical vignettes were composed to reflect various stages of PD. 35 international expert clinicians in the field of PDs in older adults were presented three vignettes and asked to assign the PD to a clinical stage. Interrater reliability was good (κ = .86; 95% CI: 0.79-0.94). Overall sensitivity (0.80) and specificity (0.93) rates of the model were good. Sensitivity for stages 2, 3 and 4 was very good (0.92, 0.85, and 0.85, respectively), whereas sensitivity for stage 1 was insufficient (0.60). The overall satisfactory interrater reliability, sensitivity and specificity rates indicate that CS might be a valid and reliable new approach of assessing PDs in older adults.