Meta-analysis of randomised controlled trials

2018

J Rheum Dis

Diagnosis is a critical step for clinical treatment. Many individual studies have been conducted to determine the accuracy of various diagnostic tests, but they had small sample sizes and correspondingly inadequate statistical strength. Combining the results from several such studies can help increase the statistical strength and precision of their results. Meta-analysis is a useful tool for evaluating the accuracy of diagnostic tests and can be used to obtain precise estimates when multiple small studies for a given test and subject pool are available. The need for meta-analysis on studies examining diagnostic test accuracy has increased noticeably, and more meta-analyses on diagnostic test accuracy studies are being published. A meta-analysis of diagnostic test accuracy studies differs from a typical meta-analysis because diagnostic test accuracy studies report a pair of statistics, such as sensitivity and specificity, rather than a single statistic. Therefore, meta-analyses of the diagnostic test accuracy need to deal with two summary statistics simultaneously. More complex statistical methods are required for conducting meta-analyses using diagnostic test accuracy studies compared to that required for conventional meta-analysis. This is because the sensitivity and specificity are generally inversely correlated due to a threshold effect, and there is considerable heterogeneity in the results of test accuracy studies. This review provides an overview of the process of meta-analysis of the diagnostic test accuracy. (J Rheum Dis 2018;25:3-10)

Section: )mentioning

confidence: 99%

Overview of the Process of Conducting Meta-analyses of the Diagnostic Test Accuracy

2018

J Rheum Dis

“…Odds ratios (ORs) and 95% CIs were calculated for dichotomous data. Cochran's Q-statistic was used to assess within-and between-study variations and heterogeneities [18]. This heterogeneity test was used to assess the probability of the null hypothesis that all studies evaluated showed the same effect.…”

Section: Evaluation Of Statistical Associationsmentioning

confidence: 99%

“…When a significant Q-statistic (p＜0.10) indicated heterogeneity across studies, the random effects model was used for meta-analysis, but when heterogeneity across studies was not indicated, the fixed effects model was used. This model assumes that genetic factors have similar effects on disease susceptibility across all studies and that observed variations between studies are caused by chance alone [18]. In contrast, the random effects model assumes that different studies show substantial diversity, and assesses both within-study sampling errors and between-study variances [19].…”

Section: Evaluation Of Statistical Associationsmentioning

confidence: 99%

Comparison of Disease Activity Score 28 Using C-reactive Protein and Disease Activity Score 28 Using Erythrocyte Sedimentation Rate in Assessing Activity and Treatment Response in Rheumatoid Arthritis: A Meta-analysis

Song

2016

J Rheum Dis

Objective. We compared the Disease Activity Score 28 (DAS28) using C-reactive protein (DAS28-CRP) with DAS28 using erythrocyte sedimentation rate (DAS28-ESR) in assessing rheumatoid arthritis (RA) activity and determining European League Against Rheumatism (EULAR) response criteria. Methods. We searched the PubMed, EMBASE, and Cochrane databases and performed a meta-analysis to examine comparisons between DAS28-CRP and DAS28-ESR by RA activity and EULAR response criteria. Results. A total of ten studies were included in this meta-analysis. Significantly more patients were classified as having remission or low disease activity when using DAS28-CRP than when using DAS28-ESR (odds ratio [OR]=1.869, 95% confidence interval [CI]=1.180 to 2.959, p=0.008; OR=1.411, 95% CI=1.256 to 1.586, p=7.0×10 −8 ), whereas fewer patients were classified as having high disease activity when using DAS28-CRP than when using DAS28-ESR (OR=0.534, 95% CI=0.388 to 0.734, p=1.1×10 −4 ). More patients were classified as having good response with criteria were based on DAS28-CRP than with DAS28-ESR (OR=1.390, 95% CI=1.183 to 1.632, p=6.10×10 −5 ). Conclusion. Our meta-analysis demonstrates that DAS28-CRP underestimates disease activity and overestimates response by the EULAR response criteria compared to DAS28-ESR. (J Rheum Dis 2016;23:241-249)

“…The chi-square test was used to establish whether observed genotype frequencies in the control groups conformed to Hardy-Weinberg equilibrium (HWE). Point estimates of risks, odds ratios (ORs), and 95 % confidence intervals (CIs) were determined for each study, and Cochran's Q statistic was used to assess withinand between-study variation and heterogeneity [32]. The heterogeneity test was used to assess the probability of the null hypothesis that all studies were evaluating the same effect.…”

Section: Evaluation Of Statistical Associationsmentioning

confidence: 99%

Association between IL-6 −174 G/C, IL-6 −634 G/C, and IFN-γ +874 A/T polymorphisms and susceptibility to recurrent pregnancy loss: a meta-analysis

Choi

2015

J Assist Reprod Genet

Objective The aim of this study was to determine whether interleukin-6 (IL-6) −174 G/C, IL-6 −634 G/C, and interferon-γ (IFN-γ) +874 A/T polymorphisms are associated with susceptibility to recurrent pregnancy loss (RPL). Methods We conducted a literature search using PubMed and EMBASE databases and performed a meta-analysis using fixed-or random-effects models. Results A total of 15 articles met the study inclusion criteria. When all study subjects were considered together, metaanalysis showed no association between RPL and the IL-6 −174 GG+GC genotype (odds ratio [OR]=0.794, 95 % confidence interval [CI]=0.542-1.163, p=0.236). However, stratification of the data by ethnicity indicated an association between this genotype and RPL in non-Caucasians (OR=0.528, 95 % CI=0.302-0.925, p=0.028), but not in Caucasian populations. Moreover, meta-analysis revealed an association between RPL and the IL-6 −634 GG+GC genotype in all study subjects (OR=0.556, 95 % CI=0.383-0.806, p=0.002), while stratification by ethnicity revealed a negative association between this genotype and RPL in Asian (OR=0.545, 95 % CI= 0.371-0.800, p=0.002) but not Middle Eastern populations. Furthermore, a relationship between the IFN-γ +874 A allele and RPL was identified in non-Caucasians (OR=1.403, 95 % CI=1.133-1.734, p=0.002), but not in Caucasians.Conclusions This meta-analysis demonstrates that IL-6 −174 G/C, IL-6 −634 G/C, and IFN-γ +874 A/T polymorphisms are associated with susceptibility to RPL, particularly in nonCaucasians.