Examining publication bias-a simulation-based evaluation of statistical tests on publication bias

Second curettage versus conventional chemotherapy in avoiding unnecessary chemotherapy and reducing the number of chemotherapy courses for patients with gestational trophoblastic neoplasia: A systematic review and meta-analysis

BACKGROUND

Chemotherapy is the recommended treatment for gestational trophoblastic neoplasia (GTN). Second curettage had been advocated to avoid unnecessary chemotherapy and to reduce the courses of chemotherapy; however, consensus has not been reached as there are arguments claiming its inability of inducing complete regression.

OBJECTIVES

The present study was performed to clarify the effectiveness of second curettage for avoiding unnecessary chemotherapy and lowering the number of chemotherapy courses in patients with post-molar GTN.

SEARCH STRATEGY

Seven predominant electronic databases were searched, including four English databases and three Chinese databases, from the inception of each database until January 31, 2023.

SELECTION CRITERIA

Studies were included if they were: (1) human, (2) explicitly indicated exposure to second curettage, (3) explicitly indicated control to conventional chemotherapy, (4) explicitly indicated the participants were patients with gestational trophoblastic neoplasia (GTN), and (5) compared the outcome of interest as the number of the course of chemotherapy.

DATA COLLECTION AND ANALYSIS

Two authors extracted and analyzed the data independently. Disagreements were reconciled by reviewing the full text by a third author. The data of study location, data collection, study design, number of participants, intervention strategy, control strategy, the follow-up period, outcome, adverse events were analyzed.

MAIN RESULTS

With regard to avoiding unnecessary chemotherapy, the overall pooled effect size of the second curettage group had a significant advantage over the conventional chemotherapy group with an OR of 0.02 (95% CI: 0.00-0.06). Meanwhile, for reducing the number of chemotherapy courses, the overall pooled effect size of the second curettage group had significant advantage over the conventional chemotherapy group with a mean difference of -2.11 (95% CI: -3.72 to -0.51).

CONCLUSION

The second curettage group had a significant advantage over the conventional chemotherapy group in avoiding unnecessary chemotherapy and reducing the number of chemotherapy courses. Further larger multi-center randomized controlled trials should be conducted to confirm our results and to clarify the optimal patients' group for second curettage in patients with post-molar GTN.

Diffusion tensor imaging in the courtroom: Distinction between scientific specificity and legally admissible evidence

Interest and uptake of science and medicine peer-reviewed literature by readers outside of a paper’s topical subject, field or even discipline is ever-expanding. While the application of knowledge from one field or discipline to others can stimulate innovative solutions to problems facing modern society, it is also fraught with danger for misuse. In the practice of law in the United States, academic papers are submitted to the courts as evidence in personal injury litigation from both the plaintiff (complainant) and defendant. Such transcendence of an academic publication over disciplinary boundaries is immediately met with the challenge of application by a group that inherently lacks in-depth knowledge on the scientific method, the practice of evidence-based medicine, or the publication process as a structured and internationally synthesized process involving peer review and guided by ethical standards and norms. A modern-day example of this is the ongoing conflict between the sensitivity of diffusion tensor imaging (DTI) and the legal standards for admissibility of evidence in litigation cases of mild traumatic brain injury (mTBI). In this review, we amalgamate the peer-reviewed research on DTI in mTBI with the court’s rationale underlying decisions to admit or exclude evidence of DTI abnormalities to support claims of brain injury. We found that the papers which are critical of the use of DTI in the courtroom reflect a primary misunderstanding about how diagnostic biomarkers differ legally from relevant and admissible evidence. The clinical use of DTI to identify white matter abnormalities in the brain at the chronic stage is a valid methodology both clinically as well as forensically, contributes data that may or may not corroborate the existence of white matter damage, and should be admitted into evidence in personal injury trials if supported by a clinician. We also delve into an aspect of science publication and peer review that can be manipulated by scientists and clinicians to publish an opinion piece and misrepresent it as an unbiased, evidence-based, systematic research article in court cases, the decisions of which establish precedence for future cases and have implications on future legislation that will impact the lives of every citizen and erode the integrity of science and medicine practitioners.

Iodine nutrition and papillary thyroid cancer

Thyroid cancer (TC) is the most frequent endocrine malignancy. The incidence of TC, especially papillary thyroid carcinoma (PTC), has continued to rise all over the world during the past few years, for reasons that are not entirely clear. Though the phenomenon of overdiagnosis is occurring, it is not the sole driver of the substantial increase in incidence. Lifestyle, environmental factors, or complications are considered to be potential risk factors. Among these factors, iodine is a micronutrient that is vital to thyroid function. The effect of iodine intake on PTC has been controversial for many years and the epidemiological or experimental studies provided diametrically opposite conclusions. Combining all these studies, we found that iodine nutrition may affect the overall prevalence, distribution of the histological types, and clinicopathological aggressiveness of TC, especially PTC. However, the available evidence is poor due to the impact of various internal and external related factors. Therefore, this article sums up available results from both epidemiological and experimental studies, future studies are also warranted to expound on the relationship between overall PTC prevalence and iodine intake.

Detecting publication selection bias through excess statistical significance

We introduce and evaluate three tests for publication selection bias based on excess statistical significance (ESS). The proposed tests incorporate heterogeneity explicitly in the formulas for expected and ESS. We calculate the expected proportion of statistically significant findings in the absence of selective reporting or publication bias based on each study's SE and meta-analysis estimates of the mean and variance of the true-effect distribution. A simple proportion of statistical significance test (PSST) compares the expected to the observed proportion of statistically significant findings. Alternatively, we propose a direct test of excess statistical significance (TESS). We also combine these two tests of excess statistical significance (TESSPSST). Simulations show that these ESS tests often outperform the conventional Egger test for publication selection bias and the three-parameter selection model (3PSM).

Reputation effects in peer-to-peer online markets: A meta-analysis∗

Most online market exchanges are governed by reputation systems, which allow traders to comment on one another's behavior and attributes with ratings and text messages. These ratings then constitute sellers' reputations that serve as signals of their trustworthiness and competence. The large body of research investigating the effect of reputation on selling performance has produced mixed results, and there is a lack of consensus on whether the reputation effect exists and what it means. After showing how the reputation effect can be derived from a game-theoretic model, we use meta-analysis to synthesize evidence from 107 studies investigating the reputation effect in peer-to-peer online markets. Our results corroborate the existence of the reputation effect across different operationalizations of seller reputation and selling performance. Our results also show the extent to which the reputation effect varies. We discuss potential explanations for the variation in reputation effects that cannot be attributed to sampling error and thereby point out promising avenues for future research.

Risk factors of postoperative delirium after liver transplantation: a systematic review and meta-analysis

INTRODUCTION

The aim of this study was to summarize the incidence and risk factors of postoperative delirium (POD) after liver transplantation (LT) and associations of POD after LT with outcomes.

EVIDENCE ACQUISITION

A literature search of Pubmed, EMBASE, and the Cochrane Databases was performed to identify studies reporting POD after LT. The Newcastle-Ottawa Scale was used to rate study quality. Effect estimates were extracted and combined using random-effect model. Pooled mean differences and odds ratios for individual risk factors were calculated using inverse-variance method and Mantel-Haenszel method, as appropriate.

EVIDENCE SYNTHESIS

Eight articles with 1434 patients were included in the meta-analysis. Overall, the pooled estimated incidence rates of POD after LT were 30% (95% confidence interval: 20-39%). Fourteen statistically significant risk factors were identified in the pooled analysis: alcohol excess, preoperative renal replacement therapy (RRT), preoperative hospital length of stay (LOS), depression, hepatic encephalopathy, alcohol etiology of liver failure, Child-Turcotte-Pugh Score, APACHE II Score, MELD Score, preoperative INR, preoperative bilirubin, intraoperative use of fentanyl, intraoperative RBC transfusion, postoperative ammonia. Patients with POD had a significantly increased mechanical ventilation, postoperative RRT, LOS and mortality rate compared with those without POD.

CONCLUSIONS

POD after LT was common and multifactorial in etiology. There are significant associations of POD after LT with some clinical outcomes. Effective interventions during perioperative period may be promising to reduce the risk of POD after LT.

A meta-analysis of the incidence rate of postoperative acute kidney injury in patients with congenital heart disease

Background

Acute kidney injury (AKI) is a common complication of cardiac surgery. However, the incidence rate of AKI in patients with congenital heart disease (CHD) greatly varies between reports owing to the different definitions used for AKI. Therefore, this study was designed as a meta-analysis aimed at summarizing the incidence rate of AKI in patients with congenital heart disease (CHD) on the basis of different AKI criteria.

Methods

Studies published till April 24, 2020, on the incidence rate of AKI in patients with CHD, were retrieved from electronic databases and printed literature. To pool data from the included studies, the effect size, a combined statistics, was chosen and presented with the incidence rate and 95% confidence interval (CI). Heterogeneity was evaluated using I² statistics and Cochran Q test. The incidence rates obtained from the subgroup analysis according to study location, type of surgery, type of cohort, age, and AKI criteria) were also evaluated to determine the correlation of AKI with these factors. Publication bias was estimated using the Egger test.

Results

Thirty studies, comprising 9925 patients with AKI who had CHD, were included. Overall, the pooled incidence rate of AKI in the patients with CHD was 38.4% (95% CI, 32.0–44.7%). However, the incidence rate was not significantly affected by gender, study location, type of surgery, type of cohort, and AKI criteria. Moreover, age was significantly associated with the incidence of AKI, and the incidence rate was higher in the patients aged < 1 month than in those aged 1 month to 18 years, < 18 years, and ≥ 18 years (P < 0.05).

Conclusions

In this study, the estimated incidence rate of AKI in patients with CHD was 38.4% and may be influenced by age. These findings highlight the importance of further investigation of the specific causes of and effective preventive measures for AKI.

Association between postoperative delirium and mortality in elderly patients undergoing hip fractures surgery: a meta-analysis

INTRODUCTION

Postoperative delirium (POD) is a common complication in elderly surgical patients. Patients undergoing hip fractures surgery who are often characterized by advanced age could be particularly prone to suffering POD. We performed a meta-analysis to assess the association between POD and mortality in elderly patients undergoing hip fractures surgery. This meta-analysis included twenty-one cohort studies, and the pooled outcomes demonstrated that approximated one-fourth of patients undergoing hipfracture surgery would develop POD, and delirium increased the mortality in these patients.

METHODS

We searched PubMed, Embase, Web of Science, and Scopus for studies that investigated the effect of POD on mortality in elderly patients undergoing hip fracture surgery. Two reviewers independently selected studies, assessed quality, and extracted data. Statistical analyses were performed by STATA 14.0 and RevMan 5.3. Risk ratios (RRs) with 95% confidence intervals (CIs) were derived using random or fixed-effects model.

RESULTS

Twenty-one cohort studies enrolling a total of 6288 patients were included, and the pooled prevalence (95% CI) of POD was 28% (23-34%). POD was associated with an increase in perioperative mortality (30-day or in-hospital mortality) [12 studies, 3123 patients, relative risk (RR) (95% CI) 2.79 (1.97-3.93)], 6-month mortality [6 studies, 1673 patients, 2.51 (1.99-3.16)], 1-year mortality [6 studies, 1896 patients, 1.98 (1.62-2.41)], and more than 1-year mortality [8 studies, 1926 patients, 2.06 (1.60-2.64)].

CONCLUSIONS

Our meta-analysis demonstrated that approximated one-fourth of patients undergoing hip fracture surgery would develop POD, and delirium increased the short-term and long-term mortality in these patients.

A kinked meta-regression model for publication bias correction

Publication bias distorts the available empirical evidence and misinforms policymaking. Evidence of publication bias is mounting in virtually all fields of empirical research. This paper proposes the endogenous kink (EK) meta-regression model as a novel method of publication bias correction. The EK method fits a piecewise linear meta-regression of the primary estimates on their standard errors, with a kink at the cutoff value of the standard error below which publication selection is unlikely. We provide a simple method of endogenously determining this cutoff value as a function of a first-stage estimate of the true effect and an assumed threshold of statistical significance. Our Monte Carlo simulations show that EK is less biased and more efficient than other related regression-based methods of publication bias correction in a variety of research conditions.

Urinary iodine is increased in papillary thyroid carcinoma but is not altered by regional population iodine intake status: a meta-analysis and implications

Excessive iodine intake has been associated with increased risk of thyroid cancer (TC) in many studies, but the results have not been consistent. Since it was common knowledge that urinary iodine (UI) is considered a sensitive marker of current iodine intake, we conducted a meta-analysis to clarify the association between high UI and TC. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement, and the Cochrane Collaboration. Between-group meta-analyses were performed to compare UI between TC patients and the healthy/euthyroid subjects in local residents and benign thyroid nodules (BTN) patients. Then, between-group meta-analyses to compare the incidence rate of iodine excess were also conducted. The 22 case-control studies included in the meta-analyses represented 15,476 participants. It is the first time to clarify that UI was increased in PTC patients, but was not altered by regional population iodine intake status. Compared with BTN patients, PTC patients exhibited both higher UIC and higher odds ratio of iodine excess only in adequate iodine intake status subgroup; UIC, not the odds ratio of iodine excess, was higher in patients with PTC than those with BTN in above requirements iodine intake subgroup. A novel insight is offered that high UI in PTC patients was less influenced by regional population iodine intake status. It is indicted that high iodine intake is not a risk factor for PTC and high urinary iodine is just a specific characteristic of the disease.

Comparison of artificial neural network and logistic regression models for prediction of outcomes in trauma patients: A systematic review and meta-analysis

BACKGROUND

Currently, two models of artificial neural network (ANN) and logistic regression (LR) are known as models that extensively used in medical sciences. The aim of this study was to compare the ANN and LR models in prediction of Health-related outcomes in traumatic patients using a systematic review.

METHODS

The study was planned and conducted based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist. A literature search of published studies was conducted using PubMed, Embase, Web of knowledge, Scopus, and Google Scholar in May 2018. Joanna Briggs Institute (JBI) checklists was used for assessing the quality of the included articles.

RESULTS

The literature searches yielded 326 potentially relevant studies from the primary searches. Overall, the review included 10 unique studies. The results of this study showed that the area under curve (AUC) for the ANN was 0.91, (95% CI 0.89-0.83) and 0.89, (95% CI 0.87-90) for the LR in random effect model. The accuracy rate for ANN and LR in random effect models were 90.5, (95% CI, 87.6-94.2) and 83.2, (95% CI 75.1-91.2), respectively.

CONCLUSION

The results of our study showed that ANN has better performance than LR in predicting the terminal outcomes of traumatic patients in both the AUC and accuracy rate. Using an ANN to predict the final implications of trauma patients can provide more accurate clinical decisions.

Total hysterectomy versus uterine evacuation for preventing post-molar gestational trophoblastic neoplasia in patients who are at least 40 years old: a systematic review and meta-analysis

Background

The clinical value of total hysterectomy for patients with hydatidiform mole (HM) being at least 40 years old remains highly controversial. Since the practice of hysterectomy has been applied globally for decades, there is an urgent need to perform a systematic review to assess its risks and benefits.

Methods

Six electronic databases, including four English databases and one Chinese database, were searched from the inception of each database till October 6th 2017. Studies were included if they: 1) were human studies, 2) explicitly indicated exposure to hysterectomy, 3) explicitly indicated control to uterine evacuation, 4) explicitly indicated the participants were older patients with HM being at least 40 years in age, 5) compared the outcome of interest as the incidence of post-molar GTN. Two authors independently conducted the literature search, study selection, data extraction. Pooled odds ratios were analyzed using Review Manager 5.3.

Results

The overall pooled effect size of total hysterectomy had a significant advantage in preventing post-molar gestational trophoblastic neoplasia over uterine evacuation with an OR of 0.19 (95% CI, 0.08–0.48; P = 0.0004) and a low heterogeneity (I² = 21%, P = 0.28). Subgroup analysis and sensitivity analysis also showed similar results.

Conclusions

Total hysterectomy, as compared to uterine evacuation, is a better therapeutic method for patients with HM being at least 40 years old unless fertility is still desired.

Electronic supplementary material

The online version of this article (10.1186/s12885-018-5168-x) contains supplementary material, which is available to authorized users.