Question

4. Meta-analysis-Mean risk ratio Aa Aa To help reduce the risk of alcohol-related health problems in heavy drinkers, many health care institutions offer therapy sessions known as brief interventions. These are a small number (1-4) of short, motivational counseling sessions (usually under 30 minutes each), which include education and advice about the dangers of heavy drinking A meta-analytical review conducted in 2002 of 56 clinical trials comparing outcomes of brief interventions with control conditions found positive evidence for brief intervention, increasing the likelihood of successful moderation or cessation of alcohol abuse among heavy drinkers. [Source: Moyer, A., Finney, J. W., Swearingen, C. E., & Vergun, P. (2002). Brief interventions for alcohol problems: A meta-analytic review of controlled investigations in treatment-seeking and non-treatment-seeking populations. Addiction, 97(3), 279-292.] Suppose you conduct a similar review of 5 clinical studies comparing the outcomes of brief interventions with those of control conditions. Results are shown in the following table, where the number of successes corresponds to the number of patients in each study that decreased or moderated their drinking during the trial period. Compute the risk ratios and log risk ratios for each study to complete the following table Total Risk Success Success Total 0 Study Treatment Treatment Control Contro Ratio Study 1 Study 2 Study 3 Study 4 Study 5 Risk Ratio 118 633 69 59 317 56 105 34 9 158 40 1130 241 1026 104 314 286 28 You use meta-analysis software to obtain the following mean log risk ratio and its 95% confidence interval: Mean Log Risk Ratio CIlower CIupper -0.02645 -0.16873 0.11583

Answer 1

Risk Success Total Success Total Control Log Study Treatment Trea Study 156 Study 2 105 Study 3 34 Study 49 Study 5158 118 633 69 59 317 40 1130 241 1026 104 314 286 28 study1- probability of success treatment- _ 116 probability of success controL sk ratio- probability of sucsent probability of success control log(2.413):ьзе D.382 Log risk ratio : same procedure for all cases- ikto 05/6ss - 0.597 Log risk ratio - loq (o 51)- - 0 224 study2- 314/30 34/4g 66(241 qq) :0.2.55 -1.79q Log risk ratio =bgU. study3 risk ratio34/q 0.547 Log risk ratio-Lo0 547)0262 study4 risk ratio -/ sq 206/1026 study 5 Lo g(As) risk ratio-I5 6.2674 i, 1 5 I Log risk ratio = 28/10H