A meta-analysis methods: Prevention of errors and bias in research results and application

Meta-analysis is a method of combining all the studies have been done in the same subject. In the literature, it has been seen that the meta-analysis method is the best statistical evaluation that provides best evidence. Because Meta Analysis combines the results of research published in recent years, it yields stronger and more sensitive results than other methods which are providing evidence. This allows the investigator to obtain a more realistic and more generalized result for the subject. For this reason, before conducting a meta-analysis, it is necessary to perform a detailed and thorough literature search of the subject to be compiled. When applying the method of meta-analysis, there are common parameters in the researches that are studied on the same subject that is necessary for the researcher. Paying attention to this issue when searching for literature allows meta-analysis to reach more accurate estimates. In current study, it is emphasized how the meta-analysis should be used in the case of heterogeneity and biases by going from the researches within the scope of biostatistics. Although this important detail is well known in the field of statistics, it has not been adequately examined in health sciences. In current study, ten different studies on the effect of black tea on lung cancer have been designed and examined for the last 10 years by using Meta Analysis. Comprehensive Meta Analysis V3.0 package program was used. Relative Risk (RR) ratios in published research have been taken into account. By combining the common results of the investigations have been combined by the method of meta-analysis and as the result these outcomes have brought some general and unbiased results. It has been determined that the studies are heterogeneous by the result of the test. In this case, the model applied to the studies has to be "Random Effect Model. "Funnel Plot" graphic was found asymmetric in the study. For this reason, the asymmetrical "Funnel Plot" showed that it has publication bias. To evaluate the asymmetry of the "Funnel Plot", the "Begg Method" was applied. According to the result of the Begg Test, the results have not publication bias. In the current study, the result of this test has been taken into consideration. In the study, for the subjects such as heterogeneity and bias were presented. The common result of Meta Analysis in current study is; the effect of black tea consumption on lung cancer has been found significantly positive (RR = 0.836; p = 0.024). Key words: Meta Analysis, Comprehensive Meta Analysis (CMA), Tea Consumption and Lung Cancer