Negative dimension stability across clinical stages and sociodemographic characteristics in schizophrenia

Negative symptoms are a challenge to schizophrenia treatment and an obstacle to functional recovery. Although a deeper knowledge of the disorder’s neurobiology is needed to change this, reliable assessment of these symptoms remains a shortcoming in the field.¹ The most accepted structures of the negative symptoms include two to five dimensions.² However, there is limited data on whether such structures are stable in different groups of patients with schizophrenia. Therefore, we aimed to verify the relationship between clinical variables and the fit of a two-factor model of the Positive and Negative Syndrome Scale’s (PANSS) negative dimension, considering different clinical stages, sex, age, and current antipsychotic use. As secondary objectives, we analyzed the effect of a multilevel structure on the psychometric quality of the two-factor dimensional structure in a Brazilian sample.

We recruited 692 individuals from four different Brazilian centers diagnosed with schizophrenia according to DSM-IV,³ of either sex, aged between 15 and 65 years, and without severe intellectual disability. The Universidade Federal de São Paulo ethics committee approved the study protocol (project 1.052.059).

We performed confirmatory factor analysis to test a correlated two-factor model in which items N1, N3, N6 and G7 expressed the “expressive deficits” factor and N2, N4 and G16 expressed the “social amotivation” factor. The items used to express the negative dimension were based on Higuchi et al.,⁴ while the model was based on Fervaha et al.,⁵ Khan et al.,⁶ and Kagan et al.⁷ We used multilevel modeling to determine the impact of pooled data in psychometric analyses. Finally, we tested the model’s invariance using the multiple causes and multiple indicators model⁸ according to sex, age, current antipsychotic type (first- vs. second-generation), and clinical stage (treatment-resistant vs. non-resistant).

The majority of the sample was men (64.3%), and the mean age was 34.9 years (SD, 10.31). Other demographic and clinical characteristics of the sample are provided in Table S1, available as online-only supplementary material. In traditional confirmatory factor analysis, the two-factor model of the negative dimension showed a poor fit (Table S2). However, it achieved good fit when a multilevel structure was included. Likewise, subsequent analysis with the multiple causes and multiple indicators method adjusted by multilevel structure revealed a good fit (Figure 1). No covariates directly affected item responses, and all showed model invariance: clinical staging, age, sex, and current antipsychotic type (first- vs. second-generation). Sex and age significantly influenced the means of both factors – male sex and younger age showed the highest means among the factors. Treatment-resistant patients had higher means only in the “expressive deficits” factor.

Figure 1
Multiple Causes and Multiple Indicators modeling to test the impact of demographics and clinical covariates on negative symptom factors in the Positive and Negative Syndrome Scale (n=606). Standardized estimates are showed. Values of categorical variables are interpreted as Cohen’s D effect sizes. ED = expressive deficits; SA = social amotivation; n1: Blunted affect; n2: Emotional withdrawal; n3: Poor rapport; n4: Passive/apathetic social withdrawal; n6: Lack of spontaneity and flow of conversation; g7: Motor retardation; g16: Active social avoidance. * p < 0.05, ** p < 0.001.

The results did not support a distinct underlying structure for negative symptoms in patients with treatment-resistant schizophrenia. The exclusivity of positive symptoms in treatment-resistant criteria may also explain the invariance of negative symptoms among treatment-resistant and non-resistant patients.

Our results suggest that a negative two-factor dimension of the Positive and Negative Syndrome Scale is stable across different groups of patients regardless of sex, age, or current antipsychotic type. The results also support the need for a multilevel approach when performing confirmatory factor analysis of the Positive and Negative Syndrome Scale using multicenter samples.⁹ Finally, we encourage the use of instruments specifically designed to assess negative symptoms in future investigations on negative dimension invariance, such as the Brief Negative Symptom Scale and the Clinical Assessment Interview for Negative Symptoms.

Acknowledgments

The authors acknowledge the financial support from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). HE has received research grants from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP). RB has received grants from the European Research Council, the Medical Research Council (UK) and Cyted (Programa Iberoamericano de Ciencia y Tecnología para el Desarrollo), as well as grants from CNPq, FAPESP, and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). CZ receives a Young Talent Research Scholarship by CAPES (grant 88887.575201/2020-00).

References

• ¹
  Krogmann A, Peters L, von Hardenberg L, Bödeker K, Nöhles VB, Correll CU. Keeping up with the therapeutic advances in schizophrenia: a review of novel and emerging pharmacological entities. CNS Spectr. 2019;24:38-69.
• ²
  Haguiara B, Koga G, Diniz E, Fonseca L, Higuchi CH, Kagan S, et al. What is the best latent structure of negative symptoms in schizophrenia? A systematic review. Schizophr Bull Open. 2021;2:sgab013.
• ³
  American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Arlington: American Psychiatric Publishing; 1994.
• ⁴
  Higuchi CH, Ortiz B, Berberian AA, Noto C, Cordeiro Q, Belangero SI, et al. Factor structure of the Positive and Negative Syndrome Scale (PANSS) in Brazil: convergent validation of the Brazilian version. Braz J Psychiatry. 2014;36:336-9.
• ⁵
  Fervaha G, Foussias G, Agid O, Remington G. Motivational and neurocognitive deficits are central to the prediction of longitudinal functional outcome in schizophrenia. Acta Psychiatr Scand. 2014;130:290-9.
• ⁶
  Khan A, Liharska L, Harvey PD, Atkins A, Ulshen D, Keefe RSE. Negative symptom dimensions of the positive and negative syndrome scale across geographical regions: implications for social, linguistic, and cultural consistency. Innov Clin Neurosci. 2017;14:30-40.
• ⁷
  Kagan S, Cogo‐Moreira H, Barbosa MG, Cavalcante D, Shinji A, Noto M, et al. Longitudinal invariance of the positive and negative syndrome scale negative dimension in antipsychotic naïve first‐episode schizophrenia. Early Interv Psychiatry. 2022;16:581-6.
• ⁸
  Brown TA. Confirmatory factor analysis for applied research. 2nd ed. New York, NY: The Guilford Press; 2015.
• ⁹
  Higuchi CH, Cogo-Moreira H, Fonseca L, Ortiz BB, Correll CU, Noto C, et al. Identifying strategies to improve PANSS based dimensional models in schizophrenia: accounting for multilevel structure, Bayesian model and clinical staging. Schizophr Res. 2021 Jul 22;S0920-9964(21)00249-8. doi: 10.1016/j.schres.2021.06.034 Online ahead of print.
  » 10.1016/j.schres.2021.06.034

Publication Dates

• Publication in this collection
  29 Aug 2022
• Date of issue
  Sep-Oct 2022

History

• Received
  06 Mar 2022
• Accepted
  30 Apr 2022