Paul-Christian Bürkner

Testing for publication bias in diagnostic meta-analysis: a simulation study

The present study investigates the performance of several statistical tests to detect publication bias in diagnostic meta-analysis by means of simulation. While bivariate models should be used to pool data from primary studies in diagnostic meta-analysis, univariate measures of diagnostic accuracy are preferable for the purpose of detecting publication bias. In contrast to earlier research, which focused solely on the diagnostic odds ratio or its logarithm ( ln ω), the tests are combined with four different univariate measures of diagnostic accuracy. For each combination of test and univariate measure, both type I error rate and statistical power are examined under diverse conditions. The results indicate that tests based on linear regression or rank correlation cannot be recommended in diagnostic meta-analysis, because type I error rates are either inflated or power is too low, irrespective of the applied univariate measure. In contrast, the combination of trim and fill and ln ω has non-inflated or only slightly inflated type I error rates and medium to high power, even under extreme circumstances (at least when the number of studies per meta-analysis is large enough). Therefore, we recommend the application of trim and fill combined with ln ω to detect funnel plot asymmetry in diagnostic meta-analysis.

Effects of intranasal oxytocin on symptoms of schizophrenia: A multivariate Bayesian meta-analysis

Schizophrenia is a heterogeneous disorder in which psychiatric symptoms are classified into two general subgroups-positive and negative symptoms. Current antipsychotic drugs are effective for treating positive symptoms, whereas negative symptoms are less responsive. Since the neuropeptide oxytocin (OT) has been shown to mediate social behavior in animals and humans, it has been used as an experimental therapeutic for treating schizophrenia and in particular negative symptoms which includes social deficits. Through eight randomized controlled trials (RCTs) and three meta-analyses, evidence for an effect of intranasal OT (IN-OT) has been inconsistent. We therefore conducted an updated meta-analysis that offers several advantages when compared to those done previously: (1) We used a multivariate analysis which allows for comparisons between symptoms and accounts for correlations between symptoms; (2) We controlled for baseline scores; (3) We used a fully Bayesian framework that allows for assessment of evidence in favor of the null hypothesis using Bayes factors; and (4) We addressed inconsistencies in the primary studies and previous meta-analyses. Eight RCTs (n=238) were included in the present study and we found that oxytocin did not improve any aspect of symptomology in schizophrenic patients and there was moderate evidence in favor of the null (no effect of oxytocin) for negative symptoms. Multivariate comparisons between symptom types revealed that oxytocin was not especially beneficial for treating negative symptoms. The effect size estimates were not moderated, publication bias was absent, and our estimates were robust to sensitivity analyses. These results suggest that IN-OT is not an effective therapeutic for schizophrenia.

Distinct cognitive impairments in different disease courses of multiple sclerosis—A systematic review and meta-analysis

HighlightsCognitive impairment was contrasted between different disease courses of MS.47 studies reporting neuropsychological data from 4460 patients with MS were included.Patients with PPMS consistently showed more severe cognitive impairment than RRMS.Differences in e.g. age and disability could not fully explain cognitive heterogeneity.Patients with PPMS may need disease management specialized on cognitive deficits. Abstract Cognitive impairment (CI) is common and debilitating in patients with multiple sclerosis. However, little is known about how different disease courses affect CI, impeding prognosis and disease management. Here, we contrasted the magnitude and profile of CI measured with standardized neuropsychological tests in patients with primary progressive multiple sclerosis (PPMS) against relapsing‐remitting multiple sclerosis (RRMS) while considering potentially confounding demographic and clinical differences. Systematic literature review and meta‐analysis was performed finding 47 eligible studies (N = 4460 patients). Effect‐sizes for 12 cognitive domains were calculated as Hedges’ g. Results indicated more severe CI overall (g = −0.37, p < .001) and in each single cognitive domain (g = −0.28 to −0.65, p < .001) in patients with PPMS despite comparable degrees of fatigue and depression. Moderator analyses revealed that these differences were not fully attributable to clinical heterogeneity between disease courses (e.g., age, disability). Particularly verbal learning and memory differentiated PPMS and RRMS independent from demographic differences. Results imply that, previously under‐recognized, PPMS patients display severe degrees of CI and need more specialized disease management than RRMS patients.

Association of Serotonin Transporter Gene AluJb Methylation with Major Depression, Amygdala Responsiveness, 5-HTTLPR/rs25531 Polymorphism, and Stress

DNA methylation profiles of the serotonin transporter gene (SLC6A4) have been shown to alter SLC6A4 expression, drive antidepressant treatment response and modify brain functions. This study investigated whether methylation of an AluJb element in the SLC6A4 promotor was associated with major depressive disorder (MDD), amygdala reactivity to emotional faces, 5-HTTLPR/rs25531 polymorphism, and recent stress. MDD patients (n=122) and healthy controls (HC, n=176) underwent fMRI during an emotional face-matching task. Individual SLC6A4 AluJb methylation profiles were ascertained and associated with MDD, amygdala reactivity, 5-HTTLPR/rs25531, and stress. SLC6A4 AluJb methylation was significantly lower in MDD compared to HC and in stressed compared to less stressed participants. Lower AluJb methylation was particularly found in 5-HTTLPR/rs25531 risk allele carriers under stress and correlated with less depressive episodes. fMRI analysis revealed a significant interaction of AluJb methylation and diagnosis in the amygdala, with MDD patients showing lower AluJb methylation associated with decreased amygdala reactivity. While no joint effect of AluJb methylation and 5-HTTLPR/rs25531 existed, risk allele carriers showed significantly increased bilateral amygdala activation. These findings suggest a role of SLC6A4 AluJb methylation in MDD, amygdala reactivity, and stress reaction, partly interwoven with 5-HTTLPR/rs25531 effects. Patients with low methylation in conjunction with a shorter MDD history and decreased amygdala reactivity might feature a more stress-adaptive epigenetic process, maybe via theoretically possible endogenous antidepressant-like effects. In contrast, patients with higher methylation might possibly suffer from impaired epigenetic adaption to chronic stress. Further, the 5-HTTLPR/rs25531 association with amygdala activation was confirmed in our large sample.

Bayes Factors From Pooled Data Are No Substitute for Bayesian Meta-Analysis: Commentary on Scheibehenne, Jamil, and Wagenmakers (2016)

Scheibehenne, Jamil, and Wagenmakers (2016; SJW) recently introduced Bayesian evidence synthesis (BES). They used it to combine evidence from seven published studies that examined the influence of ...

Between-litter variation in developmental studies of hormones and behavior: inflated false positives and diminished power

Developmental studies of hormones and behavior often include littermates-rodent siblings that share early-life experiences and genes. Due to between-litter variation (i.e., litter effects), the statistical assumption of independent observations is untenable. In two literatures-natural variation in maternal care and prenatal stress-entire litters are categorized based on maternal behavior or experimental condition. Here, we (1) review both literatures; (2) simulate false positive rates for commonly used statistical methods in each literature; and (3) characterize small sample performance of multilevel models (MLM) and generalized estimating equations (GEE). We found that the assumption of independence was routinely violated (>85%), false positives (α=0.05) exceeded nominal levels (up to 0.70), and power (1-β) rarely surpassed 0.80 (even for optimistic sample and effect sizes). Additionally, we show that MLMs and GEEs have adequate performance for common research designs. We discuss implications for the extant literature, the field of behavioral neuroendocrinology, and provide recommendations.

D-cycloserine augmentation of behavior therapy for anxiety and obsessive-compulsive disorders: A meta-analysis.

Objective The present meta-analysis investigates whether the antibiotic D-cycloserine (DCS), a partial agonist at the glutamatergic N-methyl-D-aspartate receptor, can augment the effect of behavior therapy in humans with anxiety and obsessive-compulsive disorders. Method A keyword-based computer search was conducted using common electronic databases. Only studies investigating the effect of DCS in humans with anxiety and obsessive-compulsive disorders were included, resulting in 23 studies with a combined sample size of 1314 patients. Effect sizes were coded as Hedges’ g and SMCC, the latter also incorporating differences in pre-treatment values. Bayesian multilevel meta-analysis was applied to take dependencies of effect sizes obtained from the same study into account. Results While previous meta-analyses found small to moderate improvements, the current results including the most recent research indicate that the overall effect of DCS is very small and almost indistinguishable from zero (g = -0.12, CI = [-0.27, 0.02]; SMCC = -0.10, CI = [-0.29, 0.07]). Slightly larger effects were found for social anxious patients. Further, study quality and year of publication were relevant moderators, with higher quality / more recent studies reported smaller effects of DCS. Conclusions These findings raise the question of the usefulness of DCS as an augmentation of exposure therapy for anxiety and obsessive-compulsive disorders. At least, it seems to be less promising than initially thought. The fact that study quality was inversely related to the reported effect sizes underlines the importance of high quality primary research in order to avoid over-estimation of treatment effects in clinical psychology.

Optimal design of the Wilcoxon-Mann-Whitney-test.

In scientific research, many hypotheses relate to the comparison of two independent groups. Usually, it is of interest to use a design (i.e., the allocation of sample sizes m and n for fixed N=m+n) that maximizes the power of the applied statistical test. It is known that the two-sample t-tests for homogeneous and heterogeneous variances may lose substantial power when variances are unequal but equally large samples are used. We demonstrate that this is not the case for the nonparametric Wilcoxon-Mann-Whitney-test, whose application in biometrical research fields is motivated by two examples from cancer research. We prove the optimality of the design m=n in case of symmetric and identically shaped distributions using normal approximations and show that this design generally offers power only negligibly lower than the optimal design for a wide range of distributions.

Effects of expressive writing on depressive symptoms-A meta-analysis

Correspondence Paul-Christian B€urkner, Institute of Psychology, Westf€alische WilhelmsUniversit€at M€unster, M€unster, Germany. Email: paul.buerkner@wwu.de This meta-analysis addresses the question of whether expressive writing shows an effect on reducing depressive symptoms. It focuses on samples of physically healthy adults with varying degrees of stress but without posttraumatic stress disorder. A total of 39 randomized controlled trials with 64 intervention–control group comparisons were obtained through keyword search in databases and backward search. Expressive writing did not yield significant long-term effects on depressive symptoms. However, effects were larger when the number of sessions was higher and when the writing topic was more specific. The results of this meta-analysis did not support the effectiveness of brief, self-directed expressive writing as an intervention that decreases depressive symptoms in physically healthy adults with varying degrees of psychological stress. Future research should examine whether longer, more directed writing interventions with additional therapeutic support would lead to different results.

Shedding light on the association between repetitive negative thinking and deficits in cognitive control – A meta-analysis

Individuals who experience recurrent negative thoughts are at elevated risk for mood and anxiety disorders. It is thus essential to understand why some individuals get stuck in recurrent negative thinking (RNT), whereas others are able to disengage eventually. Theoretical models propose that individuals high in recurrent negative thinking suffer from deficits in controlling the contents of working memory. Empirical findings, however, are inconclusive. In this meta-analysis, we synthesize findings from 94 studies to examine the proposed association between RNT and deficits in cognitive control. We included numerous effect sizes not reported in the primary publications. Moderator analyses tested the influence of variables, such as stimuli valence, cognitive control function (e.g., shifting, discarding), or type of RNT (i.e., rumination or worry). Results demonstrated an association between repetitive negative thinking and deficits in only one specific cognitive control function, namely difficulty discarding no longer relevant material from working memory (r = -0.20). This association remained significant after controlling for level of psychopathology. There was no substantial association between RNT and deficits in any other cognitive control function. All other moderators were not significant. We discuss limitations (e.g., primary sample sizes, reliability of paradigms) and highlight implications for future research and clinical interventions.