Differential item functioning in the Observer Scale of the POSAS for different scar types.

PURPOSE

To investigate whether the Observer Scale of the Patient and Observer Scar Assessment Scale (POSAS) can serve as a generic measure for scar quality across different scar types.

METHODS

A collection of POSAS scores derived from several clinical trials on burn (n = 404), linear (n = 384), and keloidal scars (n = 282) was analyzed using the partial credit model of the Rasch analysis package RUMM2030.

RESULTS

Differential item functioning (DIF) was observed for the Observer Scale of the POSAS between the three scar types for the items pliability, thickness, and surface area, which could be solved by item splitting. The items pigmentation and thickness showed disordered thresholds, considerable misfit, and unpredictability.

CONCLUSION

Users of the Observer Scale of the POSAS must be aware that the raw scores obtained from burn, linear, and keloidal scars cannot be compared without the scar-specific DIF adjustment of the items pliability, thickness, and surface area.