Testing a Motor Score Based on PANSS Ratings: A Proxy for Comprehensive Motor Assessment

Background and hypothesis: Abnormal psychomotor behavior is a core schizophrenia symptom. However, assessment of motor abnormalities with expert rating scales is challenging. The Positive and Negative Syndrome Scale (PANSS) includes 3 items broadly related to hypokinetic motor behavior. Here, we tested whether a sum score of the PANSS items mannerisms and posturing (G5), motor retardation (G7), and disturbance of volition (G13) corresponds to expert ratings, potentially qualifying as a proxy-marker of motor abnormalities.

Study design: Combining baseline datasets (n = 196) of 2 clinical trials (OCoPS-P, BrAGG-SoS), we correlated PANSS motor score (PANSSmot) and 5 motor rating scales. In addition, we tested whether the cutoff set at ≥3 on each PANSS motor item, ie, "mild" on G05, G07, and G13 (in total ≥9 on PANSSmot) would differentiate the patients into groups with high vs low scores in motor scales. We further sought for replication in an independent trial (RESIS, n = 102), tested the longitudinal stability using week 3 data of OCoPS-P (n = 75), and evaluated the validity of PANSSmot with instrumental measures of physical activity (n = 113).

Study results: PANSSmot correlated with all motor scales (Spearman-Rho-range 0.19-0.52, all P ≤ .007). Furthermore, the cutoff set at ≥3 on each PANSS motor item was able to distinguish patients with high vs low motor scores in all motor scales except using Abnormal Involuntary Movement Scale (Mann-Whitney-U-Tests: all U ≥ 580, P ≤ .017).

Conclusions: Our findings suggest that PANSSmot could be a proxy measure for hypokinetic motor abnormalities. This might help to combine large datasets from clinical trials to explore whether some interventions may hold promise to alleviate hypokinetic motor abnormalities in psychosis.