Psychosis risk screening in clinical high-risk adolescents: A longitudinal investigation using the Child Behavior Checklist

Abstract

This is the first study to investigate whether parent-reported social and behavioral problems on the Child Behavior Checklist (CBCL) can be used for psychosis risk screening and the identification of at-risk youth in the general population. This longitudinal investigation assessed 122 adolescent participants from three groups (at-risk, other personality disorders, non-psychiatric controls) at baseline and one year follow-up. The findings indicate that two individual CBCL rating scales, Withdrawn/Depressed and Thought Problems, have clinical and diagnostic utility as an adjunctive risk screening measure to aid in early detection of at-risk youth likely to develop psychosis. Furthermore, the findings shows that a cost-effective, general screening tool with a widespread use in community and pediatric healthcare settings has a promise to serve as a first step in a multi-stage risk screening process. This can potentially facilitate increased screening precision and reduction of high rate of false-positives in clinical high-risk individuals who present with elevated scores on psychosis-risk measures, but ultimately do not go on to develop psychosis. The findings of the present study also have significant clinical and research implications for the development of a broad-based psychosis risk screening strategy, and novel prevention and early intervention approaches in at-risk populations for the emergence of severe mental illness.

Introduction

In recent years, increased research attention has been focused on the identification and improvement of psychosis risk screening methods in at-risk populations. An extensive body of research provides evidence for social and behavioral precursors of vulnerability to psychosis long before the illness onset (Nuechterlein and Dawson, 1984, Olin and Mednick, 1996, Erlenmeyer-Kimling, 2000, Johnstone et al., 2000, Cornblatt et al., 2002). It is estimated that at least 70% of patients with schizophrenia manifest behavioral problems during adolescence (Cannon et al., 1999, Neumann et al., 1995). Early adulthood is the modal period for the onset of psychosis (Neumann and Walker, 2003). The premorbid indicators of vulnerability include schizotypal symptoms, such as social withdrawal and thought abnormalities (Walker et al., 1999), deficits in memory and executive function (Silverstein et al., 2003), neurological soft signs (Neumann and Walker, 2003), movement abnormalities (Mittal et al., 2007), and other. Also, the majority of individuals who succumb to psychotic disorders manifest prodromal signs of behavioral disturbance (Neumann et al., 1995, Larsen et al., 1996).

The general pattern of findings suggests that pre-psychotic youth are more socially isolated, withdrawn, emotionally labile, anxious, and aggressive than their healthy siblings and/or age-matched comparison subjects. They also have higher levels of impaired attention, which remain stable and elevated from childhood to adolescence, and are assumed to negatively affect social interactions leading to increased stress related to social situations (Cornblatt et al., 1997, Amminger et al., 1999, Hans et al., 2000, Miller et al., 2002a, Ballon et al., 2007). The divergence in developmental trajectories becomes more pronounced with age and is especially apparent in the adolescent period. Research findings also suggest that the behavioral expression of vulnerability to psychosis is characterized by sex differences, with males exhibiting more externalizing behavior problems, while females exhibiting more internalizing behavior problems (Neumann et al., 1995, Walker et al., 1995, Gutt et al., 2008).

Given evidence that early identification and treatment can prevent or delay the transition to psychotic illness (Stafford et al., 2013), efforts to enhance early intervention and prevention methods have become a central focus of attention. Clinician-administered assessments such as the Structured Interview for Prodromal Syndromes (SIPS; Miller et al., 2002b, Miller et al., 2003) and the Comprehensive Assessment of At-Risk Mental States (CAARMS; Yung et al., 2005) have been the standard measures used in specialty research clinics for early detection of patients at risk for psychotic illness. These measures, however, require substantial time for clinician training and patient participation, and are unlikely to be widely adopted in general clinical settings. One feasible strategy to enhance broad-based community screening of individuals at risk for psychosis is to assess the clinical and diagnostic utility of existing and widely used mental health screening tools. For instance, results from a recent study indicate that the Atypicality scale of the Behavior Assessment System for Children, Second Edition (BASC-2; Reynolds and Kamphaus, 2004) may be a useful measure for identifying youth in the early stages of psychosis (Thompson et al., 2013). Another study used the Child Behavior Checklist (CBCL; Achenbach, 1991) to assess its utility to distinguish within a clinical high-risk group of adolescents, individuals who subsequently converted to psychosis compared to those who did not convert (Simeonova et al., 2011). The findings indicate that within a clinical high-risk sample, the CBCL did not show promise as an alternative or adjunctive predictor of conversion to psychosis. No investigations, however, exist on whether the CBCL holds promise for the identification of at-risk youth in the general population. This is a feasible line of investigation, because the CBCL is a parent-report measure with extensive published normative data and its reliability and validity are well established. Although this instrument was not intended to differentiate between individuals at-risk for psychosis and control groups, it has the potential to serve as an inexpensive adjunctive screening measure in clinical practice. Also, the findings of the present study could have important implications for psychosis risk assessments in a variety of youth-oriented settings such as high-school, community centers, pediatric healthcare practice, and other.

The purpose of the present study is to shed light on two research questions: 1) Do CBCL rating scales significantly differentiate between at-risk youth and control groups? 2) At what level of accuracy do selected CBCL rating scales correctly classify individuals based on risk status? It was hypothesized that at-risk youth will be rated by parents as exhibiting more pronounced social and behavioral problems on the CBCL when compared to control groups. It was also predicted that the differences between the groups will become more pronounced over time. The adolescent period is the focus of this study because it is characterized by a rapid increase in risk for psychosis onset, and it is likely to be a critical period for early intervention and prevention (Walker, 2002).