Peter M McEvoy1, Matthew P Hyett2, Thomas Ehring3, Sheri L Johnson4, Suraj Samtani5, Rebecca Anderson2, Michelle L Moulds5. 1. Centre for Clinical Interventions, Perth, Western Australia, Australia; School of Psychology, Curtin University, Perth, Western Australia, Australia. Electronic address: peter.mcevoy@curtin.edu.au. 2. School of Psychology, Curtin University, Perth, Western Australia, Australia. 3. Department of Psychology, LMU Munich, Germany. 4. Department of Psychology, University of California Berkeley, Berkeley, CA, USA. 5. School of Psychology, The University of New South Wales, UNSW Sydney, New South Wales, Australia.
Abstract
BACKGROUND: Repetitive negative thinking (RNT) is a cognitive process that is repetitive, passive, relatively uncontrollable, and focused on negative content, and is elevated in emotional disorders including depression and anxiety disorders. Repetitive positive thinking is associated with bipolar disorder symptoms. The unique contributions of positive versus negative repetitive thinking to emotional symptoms are unknown. The first aim of this study was to use confirmatory factor analyses to evaluate the psychometrics of two transdiagnostic measures of RNT, the Repetitive Thinking Questionnaire (RTQ-10) and Perseverative Thinking Questionnaire (PTQ), and a measure of repetitive positive thinking, the Responses to Positive Affect (RPA) Questionnaire. The second aim was to determine incremental predictive utility of these measures. METHOD: All measures were administered to a sample of 2088 undergraduate students from the Netherlands (n = 992), Australia (n = 698), and America (n = 398). RESULTS: Unidimensional, bifactor, and three-factor models were supported for the RTQ-10, PTQ, and RPA, respectively. A common factor measured by all PTQ items explained most variance in PTQ scores suggesting that this measure is essentially unidimensional. The RNT factor of the RTQ-10 demonstrated the strongest predictive utility, although the PTQ was also uniquely although weakly associated with anxiety, depression, and mania symptoms. The RPA dampening factor uniquely predicted anxiety and depression symptoms, suggesting that this scale is a separable process to RNT as measured by the RTQ-10 and PTQ. LIMITATIONS: Findings were cross-sectional and need to be replicated in clinical samples. CONCLUSIONS: Transdiagnostic measures of RNT are essentially unidimensional, whereas RPA is multidimensional. RNT and RPA have unique predictive utility.
BACKGROUND: Repetitive negative thinking (RNT) is a cognitive process that is repetitive, passive, relatively uncontrollable, and focused on negative content, and is elevated in emotional disorders including depression and anxiety disorders. Repetitive positive thinking is associated with bipolar disorder symptoms. The unique contributions of positive versus negative repetitive thinking to emotional symptoms are unknown. The first aim of this study was to use confirmatory factor analyses to evaluate the psychometrics of two transdiagnostic measures of RNT, the Repetitive Thinking Questionnaire (RTQ-10) and Perseverative Thinking Questionnaire (PTQ), and a measure of repetitive positive thinking, the Responses to Positive Affect (RPA) Questionnaire. The second aim was to determine incremental predictive utility of these measures. METHOD: All measures were administered to a sample of 2088 undergraduate students from the Netherlands (n = 992), Australia (n = 698), and America (n = 398). RESULTS: Unidimensional, bifactor, and three-factor models were supported for the RTQ-10, PTQ, and RPA, respectively. A common factor measured by all PTQ items explained most variance in PTQ scores suggesting that this measure is essentially unidimensional. The RNT factor of the RTQ-10 demonstrated the strongest predictive utility, although the PTQ was also uniquely although weakly associated with anxiety, depression, and mania symptoms. The RPA dampening factor uniquely predicted anxiety and depression symptoms, suggesting that this scale is a separable process to RNT as measured by the RTQ-10 and PTQ. LIMITATIONS: Findings were cross-sectional and need to be replicated in clinical samples. CONCLUSIONS: Transdiagnostic measures of RNT are essentially unidimensional, whereas RPA is multidimensional. RNT and RPA have unique predictive utility.
Authors: Paolo Fusar-Poli; Marco Solmi; Natascia Brondino; Cathy Davies; Chungil Chae; Pierluigi Politi; Stefan Borgwardt; Stephen M Lawrie; Josef Parnas; Philip McGuire Journal: World Psychiatry Date: 2019-06 Impact factor: 49.548