This paper revisits the 1986 article "Meta-Analysis in Clinical Trials" where the authors introduced the random-effects model for meta-analysis. The method, now known as the "DerSimonian and Laird method," has become widely used in medical and clinical research due to its simplicity and ease of implementation. It is particularly useful for estimating the overall effect of a treatment and characterizing the heterogeneity of effects across studies. The authors review the background leading to the original article, describe the random-effects approach, and explore its use in various settings and trends over time. They recommend a refinement of the method using a robust variance estimator for testing overall effects and discuss its application in Big Data meta-analysis and Genome Wide Association Studies (GWAS) for studying genetic variants in complex diseases. The random-effects model assumes that the observed effect of each study is the sum of a "true" random effect and sampling error. The variance of the observed effect is the sum of variation in true means and within-study variance. The authors propose a method for estimating the mean of the "true" effects and the variation in true effects across studies. They use a method of moments approach to estimate the variation in true effects, but also consider maximum likelihood (ML) and restricted maximum likelihood (REML) methods. The method has been widely cited, with over 12,000 citations to date, and its popularity has continued to grow. The method is particularly useful in medical and clinical research, but it has also been applied in other areas such as GWAS and Big Data meta-analysis. The authors discuss the limitations of the method, including its potential for bias in hypothesis testing and the need for more robust variance estimators. They also highlight the importance of considering heterogeneity in meta-analysis and the potential for environmental factors to influence the effect of genetic variants. The authors conclude that the DerSimonian and Laird method remains a popular and widely used approach for meta-analysis, but it is important to consider refinements and alternative methods for improved accuracy and reliability, especially in settings with small sample sizes or high heterogeneity. The method is particularly well-suited for GWAS and Big Data meta-analysis, where the goal is to identify genetic variants associated with complex diseases. The authors recommend the use of a robust variance estimator for testing overall effects in these settings.This paper revisits the 1986 article "Meta-Analysis in Clinical Trials" where the authors introduced the random-effects model for meta-analysis. The method, now known as the "DerSimonian and Laird method," has become widely used in medical and clinical research due to its simplicity and ease of implementation. It is particularly useful for estimating the overall effect of a treatment and characterizing the heterogeneity of effects across studies. The authors review the background leading to the original article, describe the random-effects approach, and explore its use in various settings and trends over time. They recommend a refinement of the method using a robust variance estimator for testing overall effects and discuss its application in Big Data meta-analysis and Genome Wide Association Studies (GWAS) for studying genetic variants in complex diseases. The random-effects model assumes that the observed effect of each study is the sum of a "true" random effect and sampling error. The variance of the observed effect is the sum of variation in true means and within-study variance. The authors propose a method for estimating the mean of the "true" effects and the variation in true effects across studies. They use a method of moments approach to estimate the variation in true effects, but also consider maximum likelihood (ML) and restricted maximum likelihood (REML) methods. The method has been widely cited, with over 12,000 citations to date, and its popularity has continued to grow. The method is particularly useful in medical and clinical research, but it has also been applied in other areas such as GWAS and Big Data meta-analysis. The authors discuss the limitations of the method, including its potential for bias in hypothesis testing and the need for more robust variance estimators. They also highlight the importance of considering heterogeneity in meta-analysis and the potential for environmental factors to influence the effect of genetic variants. The authors conclude that the DerSimonian and Laird method remains a popular and widely used approach for meta-analysis, but it is important to consider refinements and alternative methods for improved accuracy and reliability, especially in settings with small sample sizes or high heterogeneity. The method is particularly well-suited for GWAS and Big Data meta-analysis, where the goal is to identify genetic variants associated with complex diseases. The authors recommend the use of a robust variance estimator for testing overall effects in these settings.