A likelihood-based sensitivity analysis for publication bias in meta-analysis

2012 ◽  
Vol 62 (1) ◽  
pp. 47-66 ◽  
Author(s):  
John B. Copas
Keyword(s):  
Sensitivity Analysis ◽  
Publication Bias ◽  
Meta Analysis

scholarly journals Prevalence of At-Risk Marriages among Couples attending Premarital Screening (PMS) Programs: A Systematic Review and Meta-Analysis

2020 ◽  
Author(s):  
BibiAsma Syed ◽  
Mashael Alshafai ◽  
Karam Turk-Adawi
Keyword(s):  
Systematic Review ◽  
At Risk ◽  
Sensitivity Analysis ◽  
Publication Bias ◽  
Meta Analysis ◽  
Genetic Diseases ◽  
World Health ◽  
Middle East Countries ◽  
Mesh Terms ◽  
Premarital Screening

Background: Hemoglobinopathies are among the most common inherited genetic diseases. The World Health Organization estimates that at least 5% of the world’s population are carriers for hemoglobinopathies (2.9% for thalassemia and 2.3% for sickle cell disease). Programs like premarital screening (PMS) have been developed in most Middle East countries on a mandatory basis to reduce atrisk marriages by providing counseling after a confirmed “genetic carrier” state for hemoglobinopathies. Aim/Objective: The aim of this systematic review and meta-analysis was to estimate the prevalence of atrisk marriages globally and see the variation by region, income level, ethnicity, study period, implementation year of PMS program, study design and consanguinity proportion. Methods: Different databases such as PubMed, Science Direct, and Scopus were searched systematically by using key terms and MeSH Terms. Studies from Google Scholar and reference lists of studies were also collected, and the author extracted all relevant data. Two reviewers independently conducted quality assessment by using Hoy et al (2012) risk of bias tool. Quality effects model (QEM) was used due to considerable heterogeneity observed between studies. Subgroup analysis and sensitivity analysis were also performed for assessing the causes of heterogeneity. Results: A total of 15 studies were included in this meta-analysis. The overall pooled prevalence of at-risk marriages among total couples at-risk was 64% (95% CI: 49%- 78%). Estimates of several subgroups were found to be different as compared to the overall pooled estimate. Funnel plot and Doi plot indicated the presence of publication bias. Sensitivity analysis including only studies with low risk led to a pooled estimate of 52% (CI: 46%, 57%) and indicated absence of publication bias. Conclusion and recommendations: The pooled estimates varied widely and there was a substantial heterogeneity among studies, therefore, there is a need for more well-designed studies across different countries. Moreover, the importance of the quality of counseling sessions should be stressed and combined with efforts in other community sectors, such as high schools where students can attain high knowledge regarding genetic diseases before the age of marriage.

scholarly journals Prognostic Value of Lactate Dehydrogenase in Patients with Hepatocellular Carcinoma: A Meta-Analysis

2018 ◽  
Vol 2018 ◽  
pp. 1-10 ◽  
Author(s):  
Weihao Kong ◽  
Xiaomin Zuo ◽  
Hao Liang ◽  
Jingxiong Hu ◽  
Huabing Zhang ◽  
...  
Keyword(s):  
Hepatocellular Carcinoma ◽  
Sensitivity Analysis ◽  
Lactate Dehydrogenase ◽  
Publication Bias ◽  
Subgroup Analysis ◽  
Prognostic Value ◽  
Meta Analysis ◽  
Progression Free Survival ◽  
Cochrane Library ◽  
Free Survival

Background. Previous studies have shown the prognostic value of lactate dehydrogenase (LDH) in hepatocellular carcinoma (HCC), but the results are not persuasive. Therefore, the purpose of our study was to quantitatively explore the prognostic value of LDH in hepatocellular carcinoma.Methods. We searched the Web of Science, Embase, PubMed, and the Cochrane Library for literature published before October 2018 on the prognostic value of LDH in patients with hepatocellular carcinoma. The combined hazard ratios (HRs) and 95% confidence intervals (CIs) were utilized to assess the prognostic value of LDH in overall survival (OS), recurrence-free survival (RFS), and progression-free survival (PFS) of HCC. Subgroup analysis, sensitivity analysis, and metaregression were used to explore the source of heterogeneity. Funnel plots with Begg’s test and Egger’s test were used to detect potential publication biases. Furthermore, combined odds ratios (ORs) were utilized to assess the correlation between LDH and clinicopathological features.Results. A total of 10 nonrandomized controlled studies were included in this meta-analysis. The combined effects of LDH on HCC patients’ OS, RFS/DFS, and PFS were HR = 2.07, 95% CI: 1.63-2.62, P < 0.001; HR = 1.62, 95% CI: 1.37-1.90, P < 0.001; and HR = 1.96, 95% CI: 1.14-3.36, P = 0.014, respectively. Subgroup analysis and sensitivity analysis showed that the outcome was stable, and the results of the metaregression also identified statistical models as an important source of heterogeneity. Potential publication bias was detected in the OS studies, so the trim-and-fill method was used to explore publication bias, and the results showed stability. Furthermore, the combined OR suggests that LDH was significantly correlated with gender, Child-Pugh grade, alpha-fetoprotein, vascular invasion, and tumor size.Conclusions. Preoperative LDH elevation is significantly associated with poor prognosis in patients with HCC, which may be a promising factor in assessing the prognosis of patients with HCC.

scholarly journals Relationship of Obstructive Sleep Apnoea with Diabetic Retinopathy: A Meta-Analysis

2017 ◽  
Vol 2017 ◽  
pp. 1-5 ◽  
Author(s):  
Zhenliu Zhu ◽  
Fengying Zhang ◽  
Yunxia Liu ◽  
Shuqin Yang ◽  
Chunting Li ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Sensitivity Analysis ◽  
Diabetic Retinopathy ◽  
Publication Bias ◽  
Meta Analysis ◽  
Sleep Apnoea ◽  
Obstructive Sleep ◽  
Increased Risk ◽  
Relationship Of ◽  
The Relationship

Until now, the relationship of obstructive sleep apnoea (OSA) with diabetic retinopathy (DR) was controversial. This meta-analysis was performed to obtain definitive conclusion on this topic. Relevant articles were searched on databases of Pubmed, Google Scholar, and Chinese National Knowledge Infrastructure (CNKI). The articles were selected according to inclusion and exclusion criteria. Odds ratio (OR) with 95% confidence interval (CI) was used to evaluate the relationship of OSA with risk of DR.I2andPvalue were used to assess the presence of heterogeneity.I2≥ 50% orP<0.05indicated significant heterogeneity. Sensitivity analysis was performed to evaluate the robustness of pooled results. Begg’s funnel plot and Egger’s regression analysis were adopted to assess publication bias. 6 eligible studies were selected in the present meta-analysis. The pooled results indicated that OSA was significantly associated with increased risk of DR (OR = 2.01, 95% CI = 1.49–2.72). Subgroup analysis based on type of diabetes mellitus suggested that OSA was related to DR in both Type 1 and Type 2 diabetes mellitus. Sensitivity analysis demonstrated that pooled results were robust. No significant publication bias was observed (P=0.128). The results indicate that OSA is related to increased risk of DR.

The Clinical Effects of Abdominal Binder on Abdominal Surgery: A Meta-analysis

2020 ◽  
pp. 155335062097482
Author(s):  
Nanhui Jiang ◽  
Bihai Hao ◽  
Rong Huang ◽  
Fengying Rao ◽  
Ping Wu ◽  
...  
Keyword(s):  
Sensitivity Analysis ◽  
Publication Bias ◽  
Meta Analysis ◽  
Large Sample Size ◽  
Significant Heterogeneity ◽  
Clinical Effects ◽  
Clinical Value ◽  
Standard Mean Difference ◽  
Symptom Distress Scale ◽  
Distress Scale

Objective. We conducted a meta-analysis to quantitatively evaluate the effects of abdominal binder in abdominal surgeries. Methods. Through literature retrieval in globally recognized databases (MEDLINE, EMBASE, and Cochrane Central), trials investigating the application of abdominal binder in abdominal surgeries were systematically reviewed. The main outcomes, namely, 6-minute walk test (6MWT), visual analog scale (VAS) pain score, and symptom distress scale (SDS) score, were pooled to make an overall estimation. I2 index was calculated to identify heterogeneity, and sensitivity analysis was performed to validate the stability of main results and explore the source of heterogeneity. A funnel plot and Egger’s test were applied to assess publication bias. Results. Ten randomized controlled trials consisting of 968 subjects were ultimately included for the pooled estimation. Abdominal binder significantly increased the distance of 6MWT with standard mean difference (SMD) of .555 ( P < .001) and decreased the scores of VAS and SDS with SMD of −.979 ( P < .001) and −.716 ( P < .001), respectively. Despite of the significant heterogeneity indicated by I2 index statistic, the results of sensitivity analysis revealed the reliability of the main conclusions. While we identified no obvious publication bias regarding 6MWT (Egger’s test P = .321), it seemed that significant publication biases existed with respect to the estimation of VAS ( P < .001) and SDS ( P = .006). Conclusion. The current meta-analysis verified that abdominal binder efficiently promoted recovery after abdominal surgeries in terms of facilitating mobilization, alleviating pain, and reducing postoperative distress. More rigorously designed clinical trials with large sample size are expected to further elaborate its clinical value.

scholarly journals High Prevalence of Hypovitaminosis D in Patients with Low Back Pain: Evidence from Meta-Analysis

2018 ◽  
Vol 1 (21;1) ◽  
pp. E389-E399 ◽  
Author(s):  
Dipika Bansal
Keyword(s):  
Vitamin D ◽  
Sensitivity Analysis ◽  
Low Back Pain ◽  
Back Pain ◽  
Publication Bias ◽  
Meta Analysis ◽  
Hypovitaminosis D ◽  
Low Back ◽  
Pooled Prevalence ◽  
High Prevalence

Background: Emerging evidence suggests an association between vitamin D deficiency and low back pain (LBP). Objective: To pool evidence on the prevalence of hypovitaminosis D in patients with LBP. Study Design: Meta-analysis. Methods: A comprehensive literature search was done in PubMed, Cochrane Database, and Google scholar for observational studies including cohort, cross sectional (CS), and case control (CC) evaluating the prevalence of hypovitaminosis D in LBP patients. The primary outcome assessed was a prevalence of hypovitaminosis D in patients with LBP, presented as weighted pooled prevalence ratio (WPPR) with 95% confidence interval (CI) using the random effects model. Heterogeneity and inconsistency of the measurements were identified through Cochran’s Q statistic and I2 statistic. We also performed sensitivity analysis, publication bias (using funnel plot and Begg’s test), and subgroup analysis. Results: Fourteen studies (6 were CC, 6 CS, and 2 cohort) involving 2602 patients were included in the final analysis. The WPPR (95% CI) of hypovitaminosis D in patients with LBP was found to be 0.72 (0.60–0.83). Marked heterogeneity was observed, median quality score of all studies was 7.5 interquartile range (IQR) (6.2 - 8.7) on a scale of 0 to 11. Sensitivity analysis showed robustness of the results. The WPPR of hypovitaminosis D was lower in CS at 0.60 (0.35–0.85) as compared to CC studies at 0.81 (0.72–0.90) (P < 0.01). The WPPR was lower in men at 0.74 (0.63–0.86) as compared to women at 0.84 (0.78–0.89) (P < 0.01). No publication bias was observed. Limitations: Heterogeneity in the cut off level of vitamin D to classify the included patients as vitamin D deficient. Conclusions: The high prevalence of hypovitaminosis D was observed in patients with LBP. This provides a chance to screen the deficiency and correct it by supplementation, which can be therapeutic adjunct in the management of LBP patients. Key words: Low back pain, hypovitaminosis D, meta-analysis, pooled prevalence, systematic review

scholarly journals Predictive Value of D-dimer on Mortality in Patients with Acute Aortic Dissection

2020 ◽  
Author(s):  
Shuangshuang Li ◽  
Pengcheng Du ◽  
Jian Dong ◽  
Jian Zhou ◽  
Zaiping Jing
Keyword(s):  
Sensitivity Analysis ◽  
Aortic Dissection ◽  
Confidence Interval ◽  
Hospital Mortality ◽  
Publication Bias ◽  
Odds Ratio ◽  
Acute Aortic Dissection ◽  
Meta Analysis ◽  
Cutoff Value ◽  
D Dimer

Abstract ObjectiveTo clarify the relation between D-dimer and in-hospital mortality of acute aortic dissection, a meta-analysis was performed by summarizing all relevant studies.MethodsAll related studies were retrieved and identified in PubMed and Embase databases. Precise data was extracted from standard articles, such as sample size, odds ratio, and 95% confidence interval. Then pooled odds ratio (OR) accompanying 95% confidence interval (CI) were calculated using random model. Study heterogeneity examined by Q text and I2 statistic. Sensitivity analysis was performed to assess the stability of the results. Publication bias was assessed by Egger’s test. ResultsTen studies (1954 patients) that met the inclusion criteria were included in this review. The results suggested a link between D-dimer and in-hospital mortality of acute aortic dissection (OR=1.17 95%CI=1.08-1.27). With higher of cutoff value of D-dimer, the closer for the mortality of AAD, with ORs (95% CIs) ranging from 1.13(1.09–1.16) to 4.12 (1.56–10.93). The relationship was also found in six Type A AAD studies without heterogeneity (OR=1.13 95%CI=1.08-1.18). According to sensitivity analysis, the link was stable after exclusion of one study at a time. Publication bias was find among studies (P=0.02). ConclusionsThe result of this meta-analysis indicated that D-dimer maybe a predictor in-hospital mortality of acute aortic dissection patients. What’s more, the higher of cutoff value of D-dimer, the stronger for the predictive ability.

Is shift-work associated with increased risk of esophageal cancer for males? A meta-analysis.

2021 ◽  
Vol 39 (3_suppl) ◽  
pp. 190-190
Author(s):  
Chenyu Sun ◽  
Ce Cheng
Keyword(s):  
Sensitivity Analysis ◽  
Esophageal Cancer ◽  
Cancer Risk ◽  
Publication Bias ◽  
Shift Work ◽  
Meta Analysis ◽  
Night Shift ◽  
Night Shift Work ◽  
Increased Risk ◽  
Subgroup Analyses

190 Background: Globally, more than 570,000 people are diagnosed of esophageal cancer each year. Shift-work involving circadian disruption was designated as a probable cause of cancer by The International Agency for Research on Cancer. Previous studies investigating the relationship between shift-work and esophageal cancer among showed controversial results. Thus, this meta-analysis was conducted. Methods: A comprehensive literature search on PubMed was conducted to identify all relevant studies published prior to September 2020 according to the established inclusion criteria. The quality assessment was performed by the Newcastle-Ottawa Scale (NOS). The pooled odds risk (OR) and 95% confidence intervals (CI) were calculated to estimate the association between the shift-work and esophageal cancer risk. Random-effect or fixed-effect model was used to calculate the pooled OR, based on heterogeneity significance. Subgroup analyses were conducted based on night-shift versus rotating-shift. Sensitivity analysis and publication bias detection were also performed. All statistical analyses were performed using RevMan software (version 5.3; Cochrane library) and STATA 12.0 statistical software (Stata Corp., College Station, TX), and all P values were two-tailed, the test level was 0.05. Results: 21 articles were obtained from database searching, and 9 articles were obtained from other sources. 3 articles involving 52,098 participants were included. All studies were considered moderate to high quality. All included studies investigated only males on the association between shift-work and esophageal cancer risk. A statistically significant association between shift-work and increased esophageal cancer risk among males was found (OR 2.09, 95%CI: 1.48, 2.94, P< 0.0001, I 2= 29%). In subgroup analyses, night-shift work was associated with a non-statistically significant increased risk of esophageal cancer (OR 1.56, 95%CI: 0.96, 2.53, P= 0.07, I 2= 0%). In contrast, Rotating-shift was associated with increased esophageal cancer risk (OR 2.80, 95%CI: 1.72, 4.57, P < 0.0001, I 2= 0%). Sensitivity analysis confirmed the stability of the result. Funnel plot, Egger's test, and Begg's test found no publication bias of analysis (P = 0.572). Conclusions: The current meta-analysis demonstrates that shift-work is associated with increased esophageal cancer risk for males. However, no association between night-shift work and esophageal cancer risk was found. In contrast, association between rotating-shift work and increased esophageal cancer risk was found. Original studies on females regarding shift-work and esophageal cancer risk are lacking. More original studies on this topic for both male and female are needed to further explore shift-work impacts on esophageal cancer risk.

scholarly journals Association of Matrix Metalloproteinase 9 (MMP-9) Polymorphisms with Asthma Risk: A Meta-Analysis

2019 ◽  
Vol 2019 ◽  
pp. 1-10 ◽  
Author(s):  
Fenfang Zou ◽  
Jianpeng Zhang ◽  
Guoan Xiang ◽  
Hongbin Jiao ◽  
Hongmei Gao
Keyword(s):  
Sensitivity Analysis ◽  
Publication Bias ◽  
Model Comparison ◽  
Genetic Model ◽  
Meta Analysis ◽  
Recessive Model ◽  
Published Data ◽  
China National Knowledge Infrastructure ◽  
Asthma Susceptibility ◽  
Asthma Patients

Published data on the association between MMP-9 polymorphisms (−1562 C > T, rs3918242; Gln279Arg, rs17576 Arg668Gln, rs17577) and asthma susceptibility are inconclusive. To derive a more precise estimation of this association, a meta-analysis was performed. A literature search was conducted in PubMed, Web of Science, EMBASE, Wanfang, and China National Knowledge Infrastructure (CNKI) databases to identify eligible studies. The pooled odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were used to calculate the strength of association. Sensitivity analysis was performed to evaluate the influence of individual studies on the overall effect estimates, and funnel plots and Egger’s test were inspected for indication of publication bias. Seven studies with 1592 asthma patients and 1987 controls were finally identified. Overall, we found no significant association between −1562 C > T, rs3918242 polymorphism, and asthma susceptibility in any of the genetic model comparisons. After categorizing studies into different subgroups on the basis of ethnicity and age, there is still no significant association. For the Gln279Arg, rs17576 polymorphism, there seems to be a significant association in the allelic genetic model in regard to the P value (OR = 1.11, 95% CI = 1.00–1.22, I2 = 0%, PZ=0.044); however, the value of lower 95% CI is 1.0. For the Arg668Gln, rs17577 polymorphism, a high significant association was observed in the dominant model comparison (OR = 1.65, 95% CI = 1.28–2.11, I2 = 22.50%, PZ=0), recessive model comparison (OR = 2.40, 95% CI = 1.23–4.72, I2 = 0%, PZ=0.011), homozygote genotype comparison (OR = 2.69, 95% CI = 1.36–5.33, I2 = 0%, PZ=0.004), and allelic genetic model (OR = 1.59, 95% CI = 1.29–1.97, I2 = 36.9%, PZ=0). Sensitivity analysis demonstrated the stability of our results, and publication bias was not evident. The present meta-analysis suggests that MMP-9 Arg668Gln, rs17577 polymorphism may be the risk factor for asthma susceptibility.

scholarly journals CHA2DS2-VASc Score as a Predictor for Left Atrial Thrombus or Spontaneous Echo Contrast in Patients with Nonvalvular Atrial Fibrillation: A Meta-Analysis

2020 ◽  
Vol 2020 ◽  
pp. 1-8
Author(s):  
Ping Sun ◽  
Zhi Hao Guo ◽  
Hong Bin Zhang
Keyword(s):  
Atrial Fibrillation ◽  
Sensitivity Analysis ◽  
Publication Bias ◽  
Left Atrial ◽  
Meta Analysis ◽  
Cochrane Library ◽  
Left Atrial Thrombus ◽  
Nonvalvular Atrial Fibrillation ◽  
Atrial Thrombus ◽  
Spontaneous Echo Contrast

Objective. This meta-analysis aimed at exploring the predictive value of CHA2DS2-VASc score for the left atrial thrombus (LAT) or left atrial spontaneous echo contrast (LASEC) in patients with nonvalvular atrial fibrillation (NVAF). Methods. PubMed, Embase, Web of Science, ScienceDirect, Cochrane Library, and Chinese core journals of the CNKI and Wanfang databases were searched to identify all the relevant papers that were published up to January 2020. The data were extracted for pooled odds ratios (ORs) with 95% confidence intervals (CIs), heterogeneity, subgroup, publication bias, and sensitivity analysis. Results. Overall, 15 studies containing 6223 patients with NVAF were enrolled. All studies were evaluated for LAT, and 12 studies were evaluated for LASEC. The pooled analysis using a random-effects model showed that a high CHA2DS2-VASc score was related with LAT/LASEC (pooled OR=1.59, 95% CI: 1.35–1.88, P<0.001) with high heterogeneity (I2=76.9%, P<0.001) and LAT (pooled OR=1.83, 95% CI: 1.44–2.33, P<0.001) with high heterogeneity (I2=79.4%, P<0.001). The subgroup analysis demonstrated that the sample size may be the main source of heterogeneity. Although the Begg’s funnel plot based on 15 studies for LAT/LASEC (P=0.029) and 12 studies for LAT (P=0.046) indicated the presence of publication bias among the included studies, the trim-and-fill method verified the stability of the pooled outcomes. In addition, sensitivity analysis indicated that all effects were stable. Conclusion. The results of this meta-analysis showed that the CHA2DS2-VASc score is related with LAT and LASEC in patients with NVAF. However, more studies are warranted to address this issue.

Sign in / Sign up

Export Citation Format

Share Document