Negative symptoms of schizophrenia demonstrate measurement bias across sociodemographic variables

Negative symptoms are a core feature of schizophrenia and severely impact functioning and quality of life. The five symptom domains (anhedonia, asociality, avolition, blunted affect, and alogia) have emerged as the preferred structure of negative symptoms; however, it is unclear whether these dimensions demonstrate measurement bias across sociodemographic variables. We used regularized moderated nonlinear factor analysis (MNLFA) to evaluate potential sociodemographic bias in the measurement of negative symptoms among individuals with schizophrenia.

The study included two samples of people with schizophrenia (total n = 1175). Negative symptoms were assessed with the Brief Negative Symptom Scale. Regularized MNLFA was conducted to assess whether the five-factors of negative symptoms showed measurement bias at the item and latent levels as a function of age, education level, and sex.

At the item level, the asociality behavior item showed differential item functioning, such that older participants were more likely to endorse more severe deficits on this item after accounting for their latent level of asociality. In addition, those with higher education levels were more likely to endorse more severe deficits on the avolition internal experience item after accounting for their latent level of avolition. Measurement bias was also present across all five of the latent negative symptom variables.

As negative symptoms increasingly serve as treatment targets and surrogate outcomes, ensuring unbiased measurement is essential for advancing clinical precision. The presence of measurement bias underscores the need for careful consideration of age, education, and sex when interpreting symptom scores in both research and clinical settings.