Hepatitis C infection in the general population of Iran: a systematic review.

TLDR: It seems that the meta-analysis method would not have been a suitable method to achieve the objectives of this systematic review because the weighting system of this method only considers sample size and not the size of the provincial population.

Abstract: books by two independent reviewers. We also searched the research projects of 29 out of 40 Iranian universities of medical sciences from their websites. We contacted the Center for Disease Control (CDC) of the Iranian Ministry of Health and the Iranian Blood Transfusion Organization (IBTO) for searching national reports from the study time period. Medical students’ theses were also evaluated by two independent reviewers from the Iranian center for scientific documents and records (IranDoc). Finally, we consulted eight expert HCV researchers in Iran and searched their personal archives for additional citations. Forward and backward citations of searched items were also performed. Critical appraisal and selection of studies Two independent reviewers reviewed all citations thoroughly and checked for eligibility criteria to include the studies in the analysis. The inclusion criteria were all cross-sectional studies that specified temporal and geographic characteristics of the study; sufficiently declared objectives; and that used a valid 213 Seyed Moayed Alavian et al. Hepatitis Monthly, Summer 2009; 9(3): 211-223 sampling method that allowed for a generalization of the findings to the target population, valid measurement instruments for all study subjects, and appropriate analytic methods for the given sampling design and demographic characteristics. We revised the criteria developed by Sharifi et al. for this purpose (12). Data extraction After evaluating studies on these criteria, we extracted the findings of the included studies to Excel spreadsheets. The extracted data were year of the study, first author, province and district of the study, sample population, sampling method, sample size, HCV Antibody detection method, HCV Antibody kit name, mean age and standard error (SE) of subjects, percentage of male subjects, and HCV point prevalence in study subjects and/or in males/females and its SE. If there were other parameters reported other than SE, such as standard deviation, confidence interval, and/or P.value, the proper modifications were performed to calculate SE. Statistical analysis We analyzed the extracted data to estimate the point prevalence of HCV infection and its 95% confidence interval (CI) and used a Cochrane Q-test with a significance level of < 0.1 for checking the statistical heterogeneity of the results. We used a meta-analysis method with the “meta” command using fix/random model based on the results of the heterogeneity test. It seems that the meta-analysis method would not have been a suitable method to achieve the objectives of this systematic review because the weighting system of this method only considers sample size and not the size of the provincial population. We used a survey data analysis method to calculate the estimate of the nationwide prevalence rate considering the weight of each province as the ratio of the provincial population to the sample size(s), where the population of each province was retrieved from the Iranian national census at 2006 (7). In provinces with more than one prevalnce study (Tehran and Sistan-va-Baluchestan), the provincial prevalence was calculated by a meta-analysis of the studies, and the total sample size was determined by adding all study sample sizes. The analysis was performed with STATA 9.1 software (STATA Corp. LP). The results were shown in geographic maps using Arc View 3.2a software (ESRI Inc. NY).