This review aimed to arrange the concepts of a network meta-analysis (NMA) and to demonstrate the analytical process of NMA using Stata software under frequentist framework. The NMA tries to synthesize evidences for a decision making by evaluating the comparative effectiveness of more than two alternative interventions for the same condition. Before conducting a NMA, 3 major assumptions-similarity, transitivity, and consistency-should be checked. The statistical analysis consists of 5 steps. The first step is to draw a network geometry to provide an overview of the network relationship. The second step checks the assumption of consistency. The third step is to make the network forest plot or interval plot in order to illustrate the summary size of comparative effectiveness among various interventions. The fourth step calculates cumulative rankings for identifying superiority among interventions. The last step evaluates publication bias or effect modifiers for a valid inference from results. The synthesized evidences through five steps would be very useful to evidence-based decision-making in healthcare. Thus, NMA should be activated in order to guarantee the quality of healthcare system.

The objective of this study is to describe the general approaches to network meta-analysis that are available for quantitative data synthesis using R software. We conducted a network meta-analysis using two approaches: Bayesian and frequentist methods. The corresponding R packages were “gemtc” for the Bayesian approach and “netmeta” for the frequentist approach. In estimating a network meta-analysis model using a Bayesian framework, the “rjags” package is a common tool. “rjags” implements Markov chain Monte Carlo simulation with a graphical output. The estimated overall effect sizes, test for heterogeneity, moderator effects, and publication bias were reported using R software. The authors focus on two flexible models, Bayesian and frequentist, to determine overall effect sizes in network meta-analysis. This study focused on the practical methods of network meta-analysis rather than theoretical concepts, making the material easy to understand for Korean researchers who did not major in statistics. The authors hope that this study will help many Korean researchers to perform network meta-analyses and conduct related research more easily with R software.

The usefulness of the quick Sequential (Sepsis-related) Organ Failure Assessment (qSOFA) score in providing bedside criteria for early prediction of poor outcomes in patients with suspected infection remains controversial. We investigated the prognostic performance of a positive qSOFA score outside the intensive care unit (ICU) compared with positive systemic inflammatory response syndrome (SIRS) criteria. A systematic literature search was performed using MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials. Data were pooled on the basis of sensitivity, specificity, and diagnostic OR. Overall test performance was summarized using a hierarchical summary ROC and the AUC. Meta-regression analysis was used to identify potential sources of bias. We identified 23 studies with a total of 146,551 patients. When predicting in-hospital mortality in our meta-analysis, we identified pooled sensitivities of 0.51 for a positive qSOFA score and 0.86 for positive SIRS criteria, as well as pooled specificities of 0.83 for a positive qSOFA score and 0.29 for positive SIRS criteria. Discrimination for in-hospital mortality had similar AUCs between the two tools (0.74 vs. 0.71; P = 0.816). Using meta-regression analysis, an overall mortality rate ≥ 10% and timing of qSOFA score measurement could be significant sources of heterogeneity. For predicting acute organ dysfunction, although the AUC for a positive qSOFA score was higher than that for positive SIRS criteria (0.87 vs. 0.76; P < 0.001), the pooled sensitivity of positive qSOFA score was very low (0.47). In addition, a positive qSOFA score tended to be inferior to positive SIRS criteria in predicting ICU admission (0.63 vs. 0.78; P = 0.121). A positive qSOFA score had high specificity outside the ICU in early detection of in-hospital mortality, acute organ dysfunction, and ICU admission, but low sensitivity may have limitations as a predictive tool for adverse outcomes. Because between-study heterogeneity was highly represented among the studies, our results should be interpreted with caution.

Comparative efficacy and safety of different concentrations of atropine for myopia control. Atropine is known to be an effective intervention to delay myopia progression. Nonetheless, no well-supported evidence exists yet to rank the clinical outcomes of various concentrations of atropine. We searched PubMed, EMBASE, Cochrane Central Register of Controlled Trials, the World Health Organization International Clinical Trials Registry Platform, and ClinicalTrials.gov on April 14, 2021. We selected studies involving atropine treatment of at least 1 year's duration for myopia control in children. We performed a network meta-analysis (NMA) of randomized controlled trials (RCTs) and compared 8 atropine concentrations (1% to 0.01%). We ranked the atropine concentrations for the corresponding outcomes by P score (estimate of probability of being best treatment). Our primary outcomes were mean annual changes in refraction (diopters/year) and axial length (AXL; millimeters/year). We extracted data on the proportion of eyes showing myopia progression and safety outcomes (photopic and mesopic pupil diameter, accommodation amplitude, and distance and near best-corrected visual acuity [BCVA]). Thirty pairwise comparisons from 16 RCTs (3272 participants) were obtained. Our NMA ranked the 1%, 0.5%, and 0.05% atropine concentrations as the 3 most beneficial for myopia control, as assessed for both primary outcomes: 1% atropine (mean differences compared with control: refraction, 0.81 [95% confidence interval (CI), 0.58-1.04]; AXL, -0.35 [-0.46 to -0.25]); 0.5% atropine (mean differences compared with control: refraction, 0.70 [95% CI, 0.40-1.00]; AXL, -0.23 [-0.38 to -0.07]); 0.05% atropine (mean differences compared with control: refraction, 0.62 [95% CI, 0.17-1.07]; AXL, -0.25 [-0.44 to -0.06]). In terms of myopia control as assessed by relative risk (RR) for overall myopia progression, 0.05% was ranked as the most beneficial concentration (RR, 0.39 [95% CI, 0.27-0.57]). The risk for adverse effects tended to rise as the atropine concentration was increased, although this tendency was not evident for distance BCVA. No valid network was formed for near BCVA. The ranking probability for efficacy was not proportional to dose (i.e., 0.05% atropine was comparable with that of high-dose atropine [1% and 0.5%]), although those for pupil size and accommodation amplitude were dose related.

To verify the dose-response relation between the degree of myopia and open-angle glaucoma (OAG) risk DESIGN: Dose-response meta-analysis. We searched the PubMed, EMBASE, and Cochrane Library databases for population-based studies published until November 30, 2020, and reporting on both myopia and OAG. Random-effect models generated pooled odds ratios (OR) and 95% CIs. Results robustness was confirmed by influence and subgroup analyses. A 2-stage dose-response meta-analysis calculated the OAG risk per unit dose of myopia (spherical equivalent [SE] decrease of 1 diopter [D]) and examined the relationship pattern. The meta-analysis comprised 24 studies in 11 countries (514,265 individuals). The pooled OR of any myopia degree's association with OAG was 1.88 (95% CI, 1.66-2.13; I² = 53%). The OR differences based on ethnicity (Asians vs Westerners) or 5 geographic areas were not statistically significant (P = .80 and P = .06, respectively). The pooled ORs of the associations between low, moderate, moderate-to-high, high myopia, and OAG were 1.50 (95% CI, 1.29-1.76), 1.69 (95% CI, 1.33-2.15), 2.27 (95% CI, 1.74-2.96), and 4.14 (95% CI, 2.57-6.69), respectively. According to the dose-response meta-analysis, the pooled OR (per SE 1-D change) was 1.21 (95% CI, 1.15-1.28). The OAG risk accelerated at approximately -6 D, and further accelerated from -8 D, showing a nonlinear concave upward slope (P = .03). For each unit (1-D) increase in myopia, the risk of glaucoma increases by approximately 20%. The risk more steeply increases in high-degree myopia, representing a significant nonlinear relationship.

The objective of this paper is to describe general approaches of diagnostic test accuracy (DTA) that are available for the quantitative synthesis of data using R software. We conduct a DTA that summarizes statistics for univariate analysis and bivariate analysis. The package commands of R software were “metaprop” and “metabin” for sensitivity, specificity, and diagnostic odds ratio; forest for forest plot; reitsma of “mada” for a summarized receiver-operating characteristic (ROC) curve; and “metareg” for meta-regression analysis. The estimated total effect sizes, test for heterogeneity and moderator effect, and a summarized ROC curve are reported using R software. In particular, we focus on how to calculate the effect sizes of target studies in DTA. This study focuses on the practical methods of DTA rather than theoretical concepts for researchers whose fields of study were non-statistics related. By performing this study, we hope that many researchers will use R software to determine the DTA more easily, and that there will be greater interest in related research.