(PS3-B50) Examining the Specificity of Self-report Measures of Schizotypy

Friday, November 18, 2022

12:00 PM – 1:00 PM EST

Location: Westside Ballroom, 5th Floor

Author(s)

NW

Noelle A. Warfford, M.A.

Graduate Student
University of Toledo
Toledo, Ohio

Co-Author(s)

JC

Joshua Cahill, None

Undergraduate Research Assistant
University of Toledo
Toledo, Ohio
PM

Peter G. Mezo, Ph.D.

Associate Professor & Associate Chair
University of Toledo
Toledo, Ohio

Self-report measures of schizotypy are an important tool for identifying individuals who may be considered at higher risk for developing a schizophrenia-spectrum disorder. Previous research into the specificity of such measures has led to mixed results. Using latent profile analysis (LPA), the current study aimed to analyze what overlap may exist in the response patterns of individuals with schizophrenia (SCZ) symptoms, non-schizophrenia-spectrum (NSS) disorder symptoms, and non-symptomatic individuals. It was hypothesized that at least three latent profiles would emerge: (1) one profile characterized by high scores across all dimensions of schizotypy and comprised mostly of individuals with SCZ symptoms, (2) another characterized by low scores across all dimensions and comprised mostly of non-symptomatic individuals, and (3) one or more profiles with low to average scores across the dimensions and comprised mostly of individuals with symptoms of NSS disorders. A total of 676 respondents (M_age = 42.4 years, SD = 15.7; 82% female; 83% White) recruited online through ResearchMatch completed the current study. The sample was broken into five groups based on their endorsement of symptoms of the following disorders: SCZ (N = 73), autism spectrum disorder (ASD; N = 111), depression (DEP; N = 124), or obsessive-compulsive disorder (OCD; N = 149). The other 219 respondents endorsed no symptoms of clinical severity. Participants completed four symptom screening measures, four self-report measures of schizotypy, and demographic questions via a single Qualtrics survey. An LPA was conducted in Mplus using dimensional scores from the self-report measures of schizotypy. The optimal LPA model included four latent profiles and had an entropy value of 0.92, indicating high classification accuracy. The first profile was characterized by low scores across all schizotypy dimensions and comprised mostly of respondents from the non-symptomatic group (64%). The second profile was characterized by high scores in negative schizotypy and mid to low scores across all other dimensions. This profile was mostly comprised of individuals from the DEP (25%) and ASD (27%) symptom groups. The third profile was characterized by mid-range scores across all dimensions. This profile was mostly comprised of respondents from the DEP (27%), ASD (22%), and OCD (26.5%) symptom groups. The fourth profile was characterized by high scores across all dimensions. Respondents from the SCZ symptom group made up just over a third (38%) of this profile, and another third (34%) came from the OCD symptom group. The significant overlap among individuals from the SCZ and OCD symptom groups in the high schizotypy profile suggests that individuals who are not at risk for developing a schizophrenia-spectrum disorder respond to these measures in a similar way to individuals who are already exhibiting symptoms of such disorders. Another possible explanation is that schizotypy is more of a transdiagnostic construct than previously believed. Future research is needed to understand the limitations of the individual measures of schizotypy included in this study and determine which explanation is most appropriate.