Disentangling the symptoms of schizophrenia: Network analysis in acute phase patients and in patients with predominant negative symptoms.

BACKGROUND: The Positive and Negative Syndrome Scale (PANSS) is widely used in schizophrenia and has been divided into distinct factors (5-factor models) and subfactors. Network analyses are newer in psychiatry and can help to better understand the relationships and interactions between the symptoms of a psychiatric disorder. The aim of this study was threefold: (a) to evaluate connections between schizophrenia symptoms in two populations of patients (patients in the acutely exacerbated phase of schizophrenia and patients with predominant negative symptoms [PNS]), (b) to test whether network analyses support the Mohr 5 factor model of the PANSS and the Kahn 2 factor model of negative symptoms, and finally (c) to identify the most central symptoms in the two populations. METHODS: Using pooled baseline data from four cariprazine clinical trials in patients with acute exacerbation of schizophrenia (n = 2193) and the cariprazine-risperidone study in patients with PNS (n = 460), separate network analyses were performed. Network structures were estimated for all 30 items of the PANSS. RESULTS: While negative symptoms in patients with an acute exacerbation of schizophrenia are correlated with other PANSS symptoms, these negative symptoms are not correlated with other PANSS symptoms in patients with PNS. The Mohr factors were partially reflected in the network analyses. The two most central symptoms (largest node strength) were delusions and uncooperativeness in acute phase patients and hostility and delusions in patients with PNS. CONCLUSIONS: This network analysis suggests that symptoms of schizophrenia are differently structured in acute and PNS patients. While in the former, negative symptoms are mainly secondary, in patients with PNS, they are mainly primary. Further, primary negative symptoms are better conceptualized as distinct negative symptom dimensions of the PANSS.

Using informative prior based on expert opinion in Bayesian estimation of the transition probability matrix in Markov modelling-an example from the cost-effectiveness analysis of the treatment of patients with predominantly negative symptoms of schizophrenia with cariprazine.

BACKGROUND: When patient health state transition evidence is missing from clinical literature, analysts are inclined to make simple assumptions to complete the transition matrices within a health economic model. Our aim was to provide a solution for estimating transition matrices by the Bayesian statistical method within a health economic model when empirical evidence is lacking. METHODS: We used a previously published cost-effectiveness analysis of the use of cariprazine compared to that of risperidone in patients with predominantly negative symptoms of schizophrenia. We generated the treatment-specific state transition probability matrices in three different ways: (1) based only on the observed clinical trial data; (2) based on Bayesian estimation where prior transition probabilities came from experts' opinions; and (3) based on Bayesian estimation with vague prior transition probabilities (i.e., assigning equal prior probabilities to the missing transitions from one state to the others). For the second approach, we elicited Dirichlet prior distributions by three clinical experts. We compared the transition probability matrices and the incremental quality-adjusted life years (QALYs) across the three approaches. RESULTS: The estimates of the prior transition probabilities from the experts were feasible to obtain and showed considerable consistency with the clinical trial data. As expected, the estimated health benefit of the treatments was different when only the clinical trial data were considered (QALY difference 0.0260), its combination with the experts' beliefs were used in the economic model (QALY difference 0.0253), and when vague prior distributions were used (QALY difference 0.0243). CONCLUSIONS: Imputing zeros to missing transition probabilities in Markov models might be untenable from the clinical perspective and may result in inappropriate estimates. Bayesian statistics provides an appropriate framework for imputing missing values without making overly simple assumptions. Informative priors based on expert opinions might be more appropriate than vague priors.