Meta-analysis is a systematic approach in selecting and integrating multiple finding across studies in order to give chances in control of potential bias. This paper aims to estimate the summary effect on the risk of mortality in sickle cell patient. The effect size index was risk ratio and date was sourced via Pubmed, Science Direct, Web of Science, Medline, Rechargegate and Google scholar. The random-effects model was employed for the analysis. The studies in the analysis were assumed to be random sample from a universe of sickle cell disease studies. The summary effect size was 0.877, with a 95% confidence interval of 0.672 to 1.146. The Z-value tested the null hypothesis that the summary effect size is 1. We found Z = -0.962 with p = 0.336 for α = 0.05; hence, we cannot rejected the null hypothesis and concluded that the summary effect size was precisely 1. The Q-statistic provided a test of the null hypothesis that 16 studies in the analysis share a common effect size; the Q-value is 77.927 with 15 degrees of freedom (k-1) and p < 0.001. For α = 0.100, we rejected the null hypothesis that the true effect size was the same in all the 16 studies since Q=k-1, k being the number of studies. The I-squared statistic was 81%, which tells us that some 81% of the variance in observed effects reflected variance in true effects rather than sampling error. Tau-squared, the variance of true effect sizes, was 0.196 in log units. Tau, the standard deviation of true effect sizes, was 0.443 in log units. Since we assumed that the true effects were normally distributed (in log units), we estimated the prediction interval to lie between 0.325 and 2.368.
Keywords: Mortality, forest plot, meta-analysis, risk ratio, sickle cell.
