No abstract

The effectiveness of group treatments for people with schizophrenia has not been examined on symptomspecific (positive and negative symptoms) outcomes, and the differential effects of the most popular group treatments remain unknown. We conducted a meta-analysis of randomized controlled trials that tested (a) the effectiveness of 7 frequently used group treatments on positive and negative symptoms and (b) if treatmentspecific outcome improvement was associated with improvement on schizophrenia symptoms. Major databases were searched from 1990 to 2018 for randomized controlled trials of group treatment for people with schizophrenia, including first-episode psychosis. A random effects meta-analysis and meta-regression was conducted on 52 studies representing 4,156 individuals that produced a significant, small effect on symptomspecific outcomes (g ϭ 0.30), with 4 group treatments (cognitive remediation, multifamily, psychoeducational, and social skills training) posting significant improvement. In addition, change on treatment-specific outcomes explained 16% of schizophrenia symptom and 44% of general functioning improvement. Results are discussed with respect to how they replicate past meta-analytic findings and possible revision of practice guidelines to incorporate evidence-based group treatments for schizophrenia. 1 Clinical Impact StatementQuestion: What is the effectiveness of common group treatments for patients diagnosed with schizophrenia? Findings: Cognitive remediation, multifamily, psychoeducational, and social skills training groups produce significant posttreatment improvement in schizophrenia symptom outcomes. Meaning: Select group treatment significantly improves schizophrenia symptom outcomes, with greater improvement when there is parallel change in treatment-specific outcomes. Next Steps: Group treatments with the strongest evidence (social skills and cognitive remediation) are candidates for practice guideline inclusion.

Psychological stress is increasingly associated with alterations in performance and affect. Yet, the relationship between experimentally induced psychological stress and neural indices of performance monitoring and error processing, as well as response inhibition, are unclear. Using scalp-recorded event-related potentials (ERPs), we tested the relationship between experimental stress, using the Trier Social Stress Test (TSST), and the error-related negativity (ERN), error positivity (Pe), and N2 ERP components. A final sample of 71 undergraduate students were randomly assigned to go through the TSST (n = 36; 18 female) or a brief mindfulness relaxation exercise (n = 35; 16 female) immediately followed by a go/no-go task while electroencephalogram (EEG) data were collected. Salivary cortisol, heart rate, and blood pressure confirmed increased physiological stress in the TSST group relative to control. Reaction times, accuracy, and post-error slowing did not differ by stress group. Two-group (TSST, control) by 2-trial type (correct, incorrect for ERN/Pe; go correct, no-go correct for N2) repeated measures ANOVAs for the ERN, Pe, and N2 showed the expected main effects of trial type; neither the ERN nor the N2 ERP components showed interactions with the stress manipulation. In contrast, the Pe component showed a significant Group by Trial interaction, with reduced Pe amplitude following the stress condition relative to control. Pe amplitude did not, however, correlate with cortisol reactivity. Findings suggest a reduction in Pe amplitude following experimental stress that may be associated with reduced error awareness or attention to errors following the TSST. Given the variability in the extant literature on the relationship between experimentally induced stress and neurophysiological reflections of performance monitoring, we provide another point of data and conclude that better understanding of moderating variables is needed followed by high-powered replication studies to get at the nuance that is not yet understood in the relationship between induced stress and performance monitoring/response inhibition processes.

No abstract