Trajectories of symptom severity and functioning over a three-year period in a psychosis high-risk sample: A secondary analysis of the Neurapro trial
Introduction
The notion that symptoms may develop along a continuum of illness course has gained traction in recent years (P. D. McGorry, Hartmann, Spooner, & Nelson, 2018; Patrick D. McGorry, Nelson, Goldstone, & Yung, 2010; van Os & Guloksuz, 2017). Despite this notion, most longitudinal studies in ultra-high risk (UHR) for psychosis samples2 are based on a categorical, dichotomized outcome measure. That is, outcome is defined as whether or not an UHR individual has progressed or ‘transitioned’ to full-threshold psychosis over the follow-up period, with the delineation between subthreshold and frank psychosis being an arbitrary decision (albeit informed by need for change in treatment approach) (Fusar-Poli & Van Os, 2012; Nelson, McGorry, Wichers, Wigman, & Hartmann, 2017; van Os & Guloksuz, 2017; Alison R. Yung, Nelson, Thompson, & Wood, 2010). Similarly, the traditional time to transition approach uses only one endpoint, ignoring the temporal pattern of change in clinical picture and functioning over time (Yuen et al., 2018). Continuous measures, such as severity of negative symptoms or severity of attenuated psychotic symptoms (APS), are routinely (and often repeatedly) obtained during the study period, however rarely adequately analysed and reported in outcome papers, which typically focus on the dichotomous final endpoint (transition vs non-transition).
Moreover, while heterogeneity of the UHR group has been recognized (Fusar-Poli, Cappucciati, Borgwardt, & et al., 2016; Nelson & Yung, 2009; van Os & Guloksuz, 2017), the existing literature treats the UHR population as a homogeneous construct with a common pathway, thereby masking individual symptom and functional trajectories for which course and thus treatment indication may differ considerably (Xie, McHugo, & Drake, 2009). Latent Class Growth Modelling (LCGM), however, can be used to investigate unobserved heterogeneity by identifying homogenous subpopulations with comparable growth trajectories over time (‘latent classes’), as well as their characteristics and correlates (Muthen, 2004; Nylund, Asparouhov, & Muthén, 2007). By identifying more homogeneous subclasses and their predictors, we may gain insight into UHR profiles, enabling the tailoring and targeting of interventions towards certain profiles. While this methodology has recently been employed in psychosis samples to map (natural or treatment-response) trajectories of symptomatology (e.g. Abdin et al., 2017; Austin et al., 2015; Levine, Rabinowitz, Case, & Ascher-Svanum, 2010; Levine, Rabinowitz, Faries, Lawson, & Ascher-Svanum, 2012; Marques et al., 2011; Schennach et al., 2012), suicidal ideation (Madsen, Karstoft, Secher, Austin, & Nordentoft, 2016), (social) functioning (Chang et al., 2018; Cole, Apud, Weinberger, & Dickinson, 2012; Hodgekins et al., 2015; Nordon et al., 2014; Velthorst et al., 2017), and cognition (J. H. Barnett et al., 2007; Thompson et al., 2013), we are not aware of any LCGM studies conducted in UHR samples to date.
While there has certainly been interest and attempts to identify course/trajectories in the UHR group (Polari et al., 2018), the sub-groups and corresponding thresholds in these studies have been defined using a consensus process among clinical and research experts (e.g., to define symptomatic remission). The advantage of LCGM is that it offers a data-driven opportunity to identify subgroups, their predictors and characteristics, without the need to define artificial cut-off points for dichotomous categories. Furthermore, factors influencing latent class membership can be explored and critical periods for interventions be identified.
The present study aims to: (1) identify classes of UHR individuals with distinct longitudinal symptom and functional trajectories (including APS, negative symptoms, depressive symptoms and level of functioning) over a 2-year follow-up period using LCGM and (2) identify sociodemographic, clinical, biological, and functional predictors of class membership. Based on the existing literature in early psychosis, we hypothesise that the analysis will identify heterogeneous trajectories with improving, stable and deteriorating temporal patterns.
Section snippets
Sample and procedures
The current study constitutes a secondary analysis of the Neurapro data set. The Neurapro trial is a multi-centre, double-blind randomised controlled trial investigating the effects of omega-3 polyunsaturated fatty acids (PUFA) versus placebo in UHR individuals (ACTRN 12608000475347) (Markulev et al., 2017). In this trial, participants meeting UHR criteria received either omega-3 PUFA or placebo over a six-month period. Both treatment arms received a cognitive-behavioural intervention
Results
The sample consisted of 304 participants (153 assigned to omega-3 PUFA treatment, 151 to placebo). Baseline demographic and symptomatic information is presented in Table 1. Other characteristics are presented in McGorry et al. (P. D. McGorry et al., 2017). According to the traditional ‘transition to psychosis’ categorical approach, 40 individuals (13%) were considered as ‘transitioned’ within the study period.
Discussion
In psychiatric prediction research, an important aim is to better understand symptomatic and functional change over time. Young people at UHR are recognised as being a very heterogeneous population with diverse outcomes and therefore, subgroups may differ with regard to the course of functional and symptomatic outcomes over time. The present study aimed to identify homogenous subclasses of UHR individuals based on trajectories of level of functioning and symptoms using LCGA.
Conclusion
Identified UHR trajectories demonstrated largely parallel, significant slopes, indicating significant symptomatic and functional improvement over the course of the study for all subclasses. Trajectories differed mainly in their intercept, i.e., severity of impairment at baseline. The pattern of improvement across trajectories may reflect the effects of CBCM, the natural course, or active engagement in a mental health service. While improving, higher symptom classes remained considerably
Funding sources
This work was supported by Grant 07TGF-1102 from the Stanley Medical Research Institute, a National Health and Medical Research Council (NHMRC) Australia Program Grant (ID: 566529; PDM, IBH, ARY,GPA) and a grant from the Colonial Foundation. JAH is supported by a University of Melbourne Postdoctoral McKenzie Fellowship. PDM was supported by a Senior Principal Research Fellowship from the NHMRC (ID: 1060996); GPA and ARY were supported by NHMRC Senior Research Fellowships (ID: 1080963 and 566593
Disclosure statement
PDM reported receiving grant funding from National Alliance for Research on Schizophrenia and Depression and unrestricted research funding from AstraZeneca, Eli Lilly, Janssen-Cilag, Pfizer, and Novartis, as well as honoraria for educational activities with AstraZeneca, Eli Lilly, Janssen-Cilag, Pfizer, Bristol-Myers Squibb, Roche, and the Lundbeck Institute. BN, IBH, ARY, and GPA have received National Health and Medical Research Council (NHMRC) funding. No other conflicts were reported.
Note.
References (54)
- et al.
Long-term trajectories of positive and negative symptoms in first episode psychosis: A 10 year follow-up study in the OPUS cohort
Schizophrenia Research
(2015) - et al.
Substance use in youth at risk for psychosis
Schizophrenia Research
(2017) - et al.
Patterns and predictors of trajectories for social and occupational functioning in patients presenting with first-episode non-affective psychosis: A three-year follow-up study
Schizophrenia Research
(2018) - et al.
The omega-3 index: A new risk factor for death from coronary heart disease?
Preventive Medicine
(2004) - et al.
The Brief assessment of cognition in schizophrenia: Reliability, sensitivity, and comparison with a standard neurocognitive battery
Schizophrenia Research
(2004) - et al.
Norms and standardization of the Brief assessment of cognition in schizophrenia (BACS)
Schizophrenia Research
(2008) - et al.
Treatment response trajectories and antipsychotic medications: Examination of up to 18 months of treatment in the CATIE chronic schizophrenia trial
Schizophrenia Research
(2012) - et al.
Trajectories of suicidal ideation in patients with first-episode psychosis: Secondary analysis of data from the OPUS trial
Lancet Psychiatry
(2016) - et al.
Psychotic-like experiences as overdetermined phenomena: When do they increase risk for psychotic disorder?
Schizophrenia Research
(2009) - et al.
Clinical trajectories in the ultra-high risk for psychosis population
Schizophrenia Research
(2018)
Response trajectories in "real-world" naturalistically treated schizophrenia patients
Schizophrenia Research
A randomised controlled trial of cognitive behaviour therapy versus non-directive reflective listening for young people at ultra high risk of developing psychosis: The detection and evaluation of psychological therapy (DEPTh) trial
Schizophrenia Research
Characterizing trajectories of cognitive functioning in older adults with schizophrenia: Does method matter?
Schizophrenia Research
The psychosis threshold in Ultra High Risk (prodromal) research: Is it valid?
Schizophrenia Research
Trajectories of positive, negative and general psychopathology symptoms in first episode psychosis and their relationship with functioning over a 2-year follow-up period
PLoS One
Substance use in clinical high risk for psychosis: A review of the literature
Early Intervention Psychiatry
A high-throughput method for the analysis of erythrocyte fatty acids and the omega-3 index
Lipids
Scale for the assessment of negative symptoms (SANS)
Gender differences in individuals at high-risk of psychosis: A comprehensive literature review
ScientificWorld Journal
Improvement and decline of cognitive function in schizophrenia over one year: A longitudinal investigation using latent growth modelling
BMC Psychiatry
Regression to the mean: What it is and how to deal with it
International Journal of Epidemiology
Using latent class growth analysis to form trajectories of premorbid adjustment in schizophrenia
Journal of Abnormal Psychology
The relationship of social function to depressive and negative symptoms in individuals at clinical high risk for psychosis
Psychological Medicine
Heterogeneity of psychosis risk within individuals at clinical high risk: A meta-analytical stratification
JAMA Psychiatry
Lost in transition: Setting the psychosis threshold in prodromal research
Acta Psychiatrica Scandinavica
Cognitive behavioral therapy for subjects at ultrahigh risk for developing psychosis: A randomized controlled clinical trial
Schizophrenia Bulletin
Revising axis V for DSM-IV: A review of measures of social functioning
American Journal of Psychiatry
Cited by (15)
Young migrants to Australia identified as being at ultra-high risk for psychosis: Pathways to care and clinical characteristics
2022, Schizophrenia ResearchCitation Excerpt :The intensity score was multiplied by the frequency score to determine the severity score for each item. A total severity score was determined by summing the severity scores of the four attenuated symptoms items, a method used in previous UHR studies (Hartmann et al., 2020; Morrison et al., 2012). The Patient Health Questionaire (PHQ9) was used to assess depressive symptoms and it is scored from 0 to 27, with higher scores indicating more severe depressive symptoms (Kroenke et al., 2001).
Cognitive-behavioural therapy and personalized treatment: An introduction to the special issue
2020, Behaviour Research and TherapyDevelopment of the PSYCHS: Positive SYmptoms and Diagnostic Criteria for the CAARMS Harmonized with the SIPS
2024, Early Intervention in Psychiatry
- 1
Contributed equally.