Nonparametric estimation of the random effects distribution for the risk or rate ratio in rare events meta-analysis with the arm-based and contrast-based approaches.

Rare events are events which occur with low frequencies. These often arise in clinical trials or cohort studies where the data are arranged in binary contingency tables. In this article, we investigate the estimation of effect heterogeneity for the risk-ratio parameter in meta-analysis of rare events studies through two likelihood-based nonparametric mixture approaches: an arm-based and a contrast-based model. Maximum likelihood estimation is achieved using the EM algorithm. Special attention is given to the choice of initial values. Inspired by the classification likelihood, a strategy is implemented which repeatably uses random allocation of the studies to the mixture components as choice of initial values. The likelihoods under the contrast-based and arm-based approaches are compared and differences are highlighted. We use simulations to assess the performance of these two methods. Under the design of sampling studies with nested treatment groups, the results show that the nonparametric mixture model based on the contrast-based approach is more appropriate in terms of model selection criteria such as AIC and BIC. Under the arm-based design the results from the arm-based model performs well although in some cases it is also outperformed by the contrast-based model. Comparisons of the estimators are provided in terms of bias and mean squared error. Also included in the comparison is the mixed Poisson regression model as well as the classical DerSimonian-Laird model (using the Mantel-Haenszel estimator for the common effect). Using simulation, estimating effect heterogeneity in the case of the contrast-based method appears to behave better than the compared methods although differences become negligible for large within-study sample sizes. We illustrate the methodologies using several meta-analytic data sets in medicine.

Rare events meta-analysis using the Bayesian beta-binomial model.

In meta-analyses of rare events, it can be challenging to obtain a reliable estimate of the pooled effect, in particular when the meta-analysis is based on a small number of studies. Recent simulation studies have shown that the beta-binomial model is a promising candidate in this situation, but have thus far only investigated its performance in a frequentist framework. In this study, we aim to make the beta-binomial model for meta-analysis of rare events amenable to Bayesian inference by proposing prior distributions for the effect parameter and investigating the models' robustness to different specifications of priors for the scale parameter. To evaluate the performance of Bayesian beta-binomial models with different priors, we conducted a simulation study with two different data generating models in which we varied the size of the pooled effect, the degree of heterogeneity, the baseline probability, and the sample size. Our results show that while some caution must be exercised when using the Bayesian beta-binomial in meta-analyses with extremely sparse data, the use of a weakly informative prior for the effect parameter is beneficial in terms of mean bias, mean squared error, and coverage. For the scale parameter, half-normal and exponential distributions are identified as candidate priors in meta-analysis of rare events using the Bayesian beta-binomial model.

Random-effects meta-analysis models for the odds ratio in the case of rare events under different data-generating models: A simulation study.

Meta-analysis of binary data is challenging when the event under investigation is rare, and standard models for random-effects meta-analysis perform poorly in such settings. In this simulation study, we investigate the performance of different random-effects meta-analysis models in terms of point and interval estimation of the pooled log odds ratio in rare events meta-analysis. First and foremost, we evaluate the performance of a hypergeometric-normal model from the family of generalized linear mixed models (GLMMs), which has been recommended, but has not yet been thoroughly investigated for rare events meta-analysis. Performance of this model is compared to performance of the beta-binomial model, which yielded favorable results in previous simulation studies, and to the performance of models that are frequently used in rare events meta-analysis, such as the inverse variance model and the Mantel-Haenszel method. In addition to considering a large number of simulation parameters inspired by real-world data settings, we study the comparative performance of the meta-analytic models under two different data-generating models (DGMs) that have been used in past simulation studies. The results of this study show that the hypergeometric-normal GLMM is useful for meta-analysis of rare events when moderate to large heterogeneity is present. In addition, our study reveals important insights with regard to the performance of the beta-binomial model under different DGMs from the binomial-normal family. In particular, we demonstrate that although misalignment of the beta-binomial model with the DGM affects its performance, it shows more robustness to the DGM than its competitors.

Persistent neuropsychiatric symptoms after COVID-19: a systematic review and meta-analysis.

The nature and extent of persistent neuropsychiatric symptoms after COVID-19 are not established. To help inform mental health service planning in the pandemic recovery phase, we systematically determined the prevalence of neuropsychiatric symptoms in survivors of COVID-19. For this pre-registered systematic review and meta-analysis (PROSPERO ID CRD42021239750), we searched MEDLINE, EMBASE, CINAHL and PsycINFO to 20 February 2021, plus our own curated database. We included peer-reviewed studies reporting neuropsychiatric symptoms at post-acute or later time-points after COVID-19 infection and in control groups where available. For each study, a minimum of two authors extracted summary data. For each symptom, we calculated a pooled prevalence using generalized linear mixed models. Heterogeneity was measured with I 2. Subgroup analyses were conducted for COVID-19 hospitalization, severity and duration of follow-up. From 2844 unique titles, we included 51 studies (n = 18 917 patients). The mean duration of follow-up after COVID-19 was 77 days (range 14-182 days). Study quality was most commonly moderate. The most prevalent neuropsychiatric symptom was sleep disturbance [pooled prevalence = 27.4% (95% confidence interval 21.4-34.4%)], followed by fatigue [24.4% (17.5-32.9%)], objective cognitive impairment [20.2% (10.3-35.7%)], anxiety [19.1% (13.3-26.8%)] and post-traumatic stress [15.7% (9.9-24.1%)]. Only two studies reported symptoms in control groups, both reporting higher frequencies in COVID-19 survivors versus controls. Between-study heterogeneity was high (I 2 = 79.6-98.6%). There was little or no evidence of differential symptom prevalence based on hospitalization status, severity or follow-up duration. Neuropsychiatric symptoms are common and persistent after recovery from COVID-19. The literature on longer-term consequences is still maturing but indicates a particularly high prevalence of insomnia, fatigue, cognitive impairment and anxiety disorders in the first 6 months after infection.

Estimation of Effect Heterogeneity in Rare Events Meta-Analysis.

The paper outlines several approaches for dealing with meta-analyses of count outcome data. These counts are the accumulation of occurred events, and these events might be rare, so a special feature of the meta-analysis is dealing with low counts including zero-count studies. Emphasis is put on approaches which are state of the art for count data modelling including mixed log-linear (Poisson) and mixed logistic (binomial) regression as well as nonparametric mixture models for count data of Poisson and binomial type. A simulation study investigates the performance and capability of discrete mixture models in estimating effect heterogeneity. The approaches are exemplified on a meta-analytic case study investigating the acceptance of bibliotherapy.

Neurology and neuropsychiatry of COVID-19: a systematic review and meta-analysis of the early literature reveals frequent CNS manifestations and key emerging narratives.

There is accumulating evidence of the neurological and neuropsychiatric features of infection with SARS-CoV-2. In this systematic review and meta-analysis, we aimed to describe the characteristics of the early literature and estimate point prevalences for neurological and neuropsychiatric manifestations.We searched MEDLINE, Embase, PsycINFO and CINAHL up to 18 July 2020 for randomised controlled trials, cohort studies, case-control studies, cross-sectional studies and case series. Studies reporting prevalences of neurological or neuropsychiatric symptoms were synthesised into meta-analyses to estimate pooled prevalence.13 292 records were screened by at least two authors to identify 215 included studies, of which there were 37 cohort studies, 15 case-control studies, 80 cross-sectional studies and 83 case series from 30 countries. 147 studies were included in the meta-analysis. The symptoms with the highest prevalence were anosmia (43.1% (95% CI 35.2% to 51.3%), n=15 975, 63 studies), weakness (40.0% (95% CI 27.9% to 53.5%), n=221, 3 studies), fatigue (37.8% (95% CI 31.6% to 44.4%), n=21 101, 67 studies), dysgeusia (37.2% (95% CI 29.8% to 45.3%), n=13 686, 52 studies), myalgia (25.1% (95% CI 19.8% to 31.3%), n=66 268, 76 studies), depression (23.0% (95% CI 11.8% to 40.2%), n=43 128, 10 studies), headache (20.7% (95% CI 16.1% to 26.1%), n=64 613, 84 studies), anxiety (15.9% (5.6% to 37.7%), n=42 566, 9 studies) and altered mental status (8.2% (95% CI 4.4% to 14.8%), n=49 326, 19 studies). Heterogeneity for most clinical manifestations was high.Neurological and neuropsychiatric symptoms of COVID-19 in the pandemic's early phase are varied and common. The neurological and psychiatric academic communities should develop systems to facilitate high-quality methodologies, including more rapid examination of the longitudinal course of neuropsychiatric complications of newly emerging diseases and their relationship to neuroimaging and inflammatory biomarkers.

Blood-Based ATN Biomarkers of Alzheimer's Disease: A Meta-Analysis.

BACKGROUND: The Amyloid Tau Neurodegeneration (ATN) framework was proposed to define the biological state underpinning Alzheimer's disease (AD). Blood-based biomarkers offer a scalable alternative to the costly and invasive currently available biomarkers. OBJECTIVE: In this meta-analysis we sought to assess the diagnostic performance of plasma amyloid (Aß40, Aß42, Aß42/40 ratio), tangle (p-tau181), and neurodegeneration (total tau [t-tau], neurofilament light [NfL]) biomarkers. METHODS: Electronic databases were screened for studies reporting biomarker concentrations for AD and control cohorts. Biomarker performance was examined by random-effect meta-analyses based on the ratio between biomarker concentrations in patients and controls. RESULTS: 83 studies published between 1996 and 2020 were included in the analyses. Aß42/40 ratio as well as Aß42 discriminated AD patients from controls when using novel platforms such as immunomagnetic reduction (IMR). We found significant differences in ptau-181 concentration for studies based on single molecule array (Simoa), but not for studies based on IMR or ELISA. T-tau was significantly different between AD patients and control in IMR and Simoa but not in ELISA-based studies. In contrast, NfL differentiated between groups across platforms. Exosome studies showed strong separation between patients and controls for Aß42, t-tau, and p-tau181. CONCLUSION: Currently available assays for sampling plasma ATN biomarkers appear to differentiate between AD patients and controls. Novel assay methodologies have given the field a significant boost for testing these biomarkers, such as IMR for Aß, Simoa for p-tau181. Enriching samples through extracellular vesicles shows promise but requires further validation.

Retest effects in working memory capacity tests: A meta-analysis.

The repeated administration of working memory capacity tests is common in clinical and research settings. For cognitive ability tests and different neuropsychological tests, meta-analyses have shown that they are prone to retest effects, which have to be accounted for when interpreting retest scores. Using a multilevel approach, this meta-analysis aims at showing the reproducibility of retest effects in working memory capacity tests for up to seven test administrations, and examines the impact of the length of the test-retest interval, test modality, equivalence of test forms and participant age on the size of retest effects. Furthermore, it is assessed whether the size of retest effects depends on the test paradigm. An extensive literature search revealed 234 effect sizes from 95 samples and 68 studies, in which healthy participants between 12 and 70 years repeatedly performed a working memory capacity test. Results yield a weighted average of g = 0.28 for retest effects from the first to the second test administration, and a significant increase in effect sizes was observed up to the fourth test administration. The length of the test-retest interval and publication year were found to moderate the size of retest effects. Retest effects differed between the paradigms of working memory capacity tests. These findings call for the development and use of appropriate experimental or statistical methods to address retest effects in working memory capacity tests.

Análisis de 2.058 pacientes con neoplasia intraepitelial cervical / Analysis of 2.058 patients with cervical intraepithelial neoplasia

Este estudio investiga el comportamiento de diferentes grados de neoplasia intraepitelial cervical (cervical intraepithelial neoplasia, CIN) incluyendo la relación estadística entre progresión persistencia y regresión en un gran número de pacientes. El estudio se realizó a partir de 5.782 resultados citológicos e histopatológicos de la investigación del cáncer cervical de 2.058 pacientes con CIN. Se comparó el comportamiento de diferentes lesiones de CIN y se analizaron estadísticamente las tasas de progresión, regresión y persistencia. Se detectó CIN en 5.778 muestras; 4 casos demostraron un carcinoma microinvasor del cérvix uterino. El 23 por ciento de CIN se asoció al papilomavirus humano (human papillomavirus HPV). Para todas las lesiones CIN, se observó una tasa de progresión del 12,9 por ciento. La tasa de progresión hacia la malignidad fue de 0,07 por ciento. La tasa de progresión más elevada fue para CIN I (41,9 por ciento). La progresión a CIN III/CIS fue significativamente superior para CIN II (14,5 por ciento) que para CIN I (7,5 por ciento) (test chi-cuadrado: p < 0,001). Según el estado de HPV, detectamos una progresión significativamente menor de CIN II HPV positivo (5,7 por ciento) que CIN II HPV negativo (11,6 por ciento) (test chi-cuadrado: p<0,0001). La neoplasia intraepitelial cervical en progresión es un proceso continuo. La progresión del CIN I se produce con mayor frecuencia a través de CIN II; asimismo CIN III/CIS se desarrolla más a menudo a partir del CIN II. Sin embargo, el 7,5 por ciento del CIN I mostró una progresión muy rápida hacia CIN III/CIS. Para estudiar este fenómeno se requiere investigaciones adicionales. Así mismo, se detectó una tasa notablemente baja de regresión para el CIN I y una tasa de progresión baja inesperada de CIN II HPV positivo