An Investigation of Longitudinal Measurement Invariance and Item Response Theory Analysis of the LPFS-BF 2.0 in a Large Clinical Norwegian Sample.

The LPFS-BF 2.0 (Level of Personality Functioning-Brief Form 2.0) is a brief self-report instrument that was developed in order to gain a quick first impression of personality dysfunction and was designed to align with the Level of Personality Scale as outlined in the DSM-5 Alternative Model for Personality Disorders (AMPD). In the current study, we used a factor analytic framework to evaluate the longitudinal measurement invariance of the Norwegian version of the LPFS-BF 2.0 in a sample of 1030 patients referred for personality disorder treatment with the R package lavaan. Additionally, we estimated item response theory (IRT) parameters for the 12 items, as well as local reliability, in order to facilitate comparisons with other studies. We found support for partial measurement invariance. Three items showed different thresholds across time points. However, these differences were minor and did not have a substantial impact on the estimated factor scores. The discrimination parameters were lower than could be expected for an instrument used in a clinical setting. Taking these findings together, we conclude that although the LPFS-BF 2.0 seems to measure the same latent trait across time, the items do not discriminate well between patients with high and low scores in a Norwegian population of adults seeking treatment for PD. Important next steps would be to estimate IRT parameters for comparable patient samples from other countries to assess the generalizability of our findings and to compare the LPFS-BF 2.0 scores to scores obtained through a clinical interview that aligns with the AMPD.