China
Africa
Explore chapters and articles related to this topic
Why we need systematic reviews and initiatives like the Cochrane Library
Published in Debra Evans, Making Sense of Evidence-based Practice for Nursing, 2023
Have a look at the abstract of a quantitative systematic review – you will often see the effect measures within it. Can you interpret them? Within the SR you can view the forest plots and practice interpreting them (tip: remember for dichotomous data – the “no difference”/null result is 1; look to see if the RR, OR, or CI range includes this value, and for continuous data – the “no difference”/null result is 0; check to see if the MD or CI range includes this value).
Meta-Analysis
Published in Trevor F. Cox, Medical Statistics for Cancer Studies, 2022
The z-value for testing is with , and hence the evidence suggests that preoperative chemotherapy is a benefit for survival. The combined estimate is shown on the forest plot as a horizontally stretched diamond, the centre being placed at the value of HR and the left and right hand ends at the lower and upper points of the confidence interval. Here, the confidence interval is rather short and so the diamond here is not particularly stretched.
Meta-Analysis with Binary Data
Published in Ding-Geng (Din) Chen, Karl E. Peace, Applied Meta-Analysis with R and Stata, 2021
Ding-Geng (Din) Chen, Karl E. Peace
OR is a commonly used metric in the analysis of binomial data. We illustrate the meta-analysis using OR for the lamotrigine data. There are several methods implemented for OR in the R package meta as described in Section 4.2. To control the number of pages in this chapter, we only illustrate the MH weighting scheme (i.e., method="MH") here and leave the inverse weighting scheme (i.e., method="Inverse") and the Peto method (i.e., method="Peto") to interested readers as exercises. In addition, we encourage readers to produce the forest plots for each analysis.
Ever-use of the intra-uterine device and the risk of ovarian cancer
Published in Journal of Obstetrics and Gynaecology, 2021
Jacques Balayla, Yaron Gil, Ariane Lasry, Cristina Mitric
Risk of bias was assessed using the Cochrane Handbook for Systematic Reviews of Interventions (results not shown). We used relative risks (OR) and a fixed effects (FE) model with the Chi-square method to calculate weighted estimates and their 95% confidence intervals (CI) where appropriate. Forest plots are provided for visualisation of the results. Statistical heterogeneity between results of studies was examined by inspecting the scatter in the data points on the graphs and the overlap of CIs, and by checking the χ2 and I2 statistics. The Review Manager 5.3.5 software (Version 5.3 Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) was used to combine data for the meta-analysis. This meta-analysis was exempt from institutional review board (IRB) approval because of the nature of the research design (review article), as well as the lack of use of identified patient data.
Prevalence of acne vulgaris among women with polycystic ovary syndrome: a systemic review and meta-analysis
Published in Gynecological Endocrinology, 2021
Fahimeh Ramezani Tehrani, Samira Behboudi-Gandevani, Razieh Bidhendi Yarandi, Marzieh Saei Ghare Naz, Enrico Carmina
The software package STATA (version 12; STATA Inc., College Station, TX) was applied to conduct statistical analysis. Heterogeneity between studies was assessed using chi-squared statistics and p>.05 was interpreted as heterogeneity. Heterogeneous and non-heterogeneous results were analyzed using the fixed effects and random effects inverse variance models for calculating the pooled effect. ‘Meta-prop’ method was applied to estimate pooled prevalence of acne in both groups in different subgroups of age, PCOS diagnostic criteria, and region. Sensitivity analysis was done to assess the reliability of the estimate obtained in the Meta-prop analysis. Moreover, meta-regression was conducted to find the association between acne in women with and without PCOS. In this respect, publication bias was assessed by Begg’s test. In publication bias cases, the trim and fill method was conducted to correct. A forest plot was also drawn to summarize the result of each study’s effect sizes and its 95% confidence intervals (CIs). p>.05 was set as significance level.
Betaine Supplementation Moderately Increases Total Cholesterol Levels: A Systematic Review and Meta-Analysis
Published in Journal of Dietary Supplements, 2021
Emilia E. Zawieja, Bogna Zawieja, Agata Chmurzynska
We used the random effect model with the REML method. This model assumes that the unknown real normal distributed effects were similar, though not identical, across studies. Weights were then determined as the inverse of variance within the studies plus the variance between studies. Heterogeneity tests included the Cochran test Q, the Hartley test and I2. The effect sizes, weights, and confidence intervals of each study, and the total effects of all studies, were presented graphically as forest plots. We also conducted a sensitivity analysis. The bias analysis was conducted using funnel plots with pseudo-96% confidence intervals. The significance of asymmetry was checked using the regression test (Egger et al. 1997). This method checks the relationship between the standard error of the effect and its size. All calculations were performed using the “metaphor” package in R (Viechtbauer 2010).