European Validation of the Self-Evaluation of Negative Symptoms (SNS): A Large Multinational and Multicenter Study.

Sonia Dollfus,Camille Delouche,W Wolfgang Fleischhacker,Oleg Papsuev,Larisa Movina,Stefan Kaiser,Francesco Brando,Gabriele Sachs,Birte Glenthøj,Alp Üçok,Ingrid Melle,Stephen F Austin,Silvana Galderisi,Karoline Gütter,Andreas Erfurth,Jan Libiger,István Bitter,Armida Mucci,Alex Hofer,Janusz Rybakowski ,Jan Hubeňák ,Mette Ødegaard Nielsen ,Giulia Giordano ,Paweł Wójciak

doi:10.3389/fpsyt.2022.826465

Abstract

BackgroundNegative symptoms are usually evaluated with scales based on observer ratings and up to now self-assessments have been overlooked. The aim of this paper was to validate the Self-evaluation of Negative Symptoms (SNS) in a large European sample coming from 12 countries. We wanted to demonstrate: (1) good convergent and divergent validities; (2) relationships between SNS scores and patients' functional outcome; (3) the capacity of the SNS compared to the Brief Negative Symptom Scale (BNSS) to detect negative symptoms; and (4) a five-domain construct in relation to the 5 consensus domains (social withdrawal, anhedonia, alogia, avolition, blunted affect) as the best latent structure of SNS.MethodsTwo hundred forty-five subjects with a DSM-IV diagnosis of schizophrenia completed the SNS, the Positive and Negative Syndrome Scale (PANSS), the BNSS, the Calgary Depression Scale for Schizophrenia (CDSS), and the Personal and Social Performance (PSP) scale. Spearman's Rho correlations, confirmatory factor analysis investigating 4 models of the latent structure of SNS and stepwise multiple regression were performed.ResultsSignificant positive correlations were observed between the total score of the SNS and the total scores of the PANSS negative subscale (r = 0.37; P < 0.0001) and the BNSS (r = 0.43; p < 0.0001). SNS scores did not correlate with the level of insight, parkinsonism, or the total score of the PANSS positive subscale. A positive correlation was found between SNS and CDSS (r = 0.35; p < 0.0001). Among the 5 SNS subscores, only avolition subscores entered the regression equation explaining a lower functional outcome. The 1-factor and 2-factor models provided poor fit, while the 5-factor model and the hierarchical model provided the best fit, with a small advantage of the 5-factor model. The frequency of each negative dimension was systematically higher using the BNSS and the SNS vs. the PANSS and was higher for alogia and avolition using SNS vs. BNSS.ConclusionIn a large European multicentric sample, this study demonstrated that the SNS has: (1) good psychometric properties with good convergent and divergent validities; (2) a five-factor latent structure; (3) an association with patients' functional outcome; and (4) the capacity to identify subjects with negative symptoms that is close to the BNSS and superior to the PANSS negative subscale.

Highlights

Negative symptoms are usually evaluated with scales based on observer ratings and up to now self-assessments (SA) have been overlooked, probably because of the idea that patients with schizophrenia with negative symptoms are unable to accurately report their own symptoms [1, 2]
Assessment of Frequency of at Least Moderate Severity Negative Symptoms As the European Medicines Agency (EMA) guidelines on clinical trial design for negative symptoms require the inclusion of subjects with at least moderate severity of negative symptoms, defined on an accepted and validated rating scale, we investigated the frequency of these symptoms using the Self-evaluation of Negative Symptoms (SNS), the Brief Negative Symptom Scale (BNSS), and the Positive and Negative Syndrome Scale (PANSS) negative subscale
Analyses in the Whole Sample Significant positive correlations were observed between the total scores of the SNS and the total scores of negative PANSS subscale (r = 0.37; P < 0.0001) and BNSS (r = 0.43; p < 0.0001) the coefficients were considered as moderate (Table 3)

Summary

Introduction

Negative symptoms are usually evaluated with scales based on observer ratings (as named hetero-assessment, HA) and up to now self-assessments (SA) have been overlooked, probably because of the idea that patients with schizophrenia with negative symptoms are unable to accurately report their own symptoms [1, 2]. The aim of this paper was to validate the Self-evaluation of Negative Symptoms (SNS) in a large European sample coming from 12 countries. We wanted to demonstrate: [1] good convergent and divergent validities; [2] relationships between SNS scores and patients’ functional outcome; [3] the capacity of the SNS compared to the Brief Negative Symptom Scale (BNSS) to detect negative symptoms; and [4] a five-domain construct in relation to the 5 consensus domains (social withdrawal, anhedonia, alogia, avolition, blunted affect) as the best latent structure of SNS

Objectives

Results

Conclusion