| BackgroundWith the popularity of evidence-based medicine in China, researches on systematic reviews in public health have gradually increased in recent years. However, there are no comprehensive summary of the development of systematic reviews in public health in China and no strict assessment for quality of other methods and reports. Good evidence not only needs good methods, but also needs standardized reporting methods, which could truly become the important information source in medical practice. Therefore, we classified the Chinese systematic reviews in public health to understand the research status and used bibliometric methods to analyze their characteristics in order to summarize the development trends and characteristics. We selected the systematic reviews in prevention and control of six kinds of diseases which are the most studied in Chinese systematic reviews in public health, including tumor, cerebrovascular disease, cardiovascular disease, hepatitis B, tuberculosis and AIDS, and then assessed the methodological quality and report quality of these reviews using OQAQ and PRISMA scales. MethodWe searched the reviews enrolled in the database of CQVIP, WANFANG, CNKI and the Chinese Biomedical Literature until June, 2010. Chinese systematic reviews/meta analysis in public health and Chinese systematic reviews/meta analysis on prevention and control of six major diseases, including cancer, cerebrovascular disease, cardiovascular disease, hepatitis B, tuberculosis and AIDS were included. Two researchers independently screened and evaluated the studies, disagreements were resolved by discussion. We classified the literatures and conducted bibliometric analysis on the issuing time, journals, authors and geographical distributions. Methodology quality and report quality of included reviews were evaluated by OQAQ and PRISMA scales, which provided evidences for the standardization of the implementation of systematic reviews in China.Results(1) Literature analysis of systematic reviews in public health: A total of 22140 literatures were acquired after initial search and 312 were included. The 312 literatures were divided into 15 categories, of which the numbers of literatures in chronic non-communicable diseases, maternal and child health and infectious diseases accounted the most. The total numbers of literatures in public health showed a rising trend after the year 2000. The coverage of journals was wide with the top 5 journals being all core journals and the numbers of the author publishing single literature were accounted for 79.81%. Beijing, Sichuan, Shandong, Guangdong and Jiangsu provinces led the development.(2) Quality assessment of the literatures on six major diseases in public health: A total of 9 230 literatures were acquired after initial search and 139 articles(142 reviews) were included. Methodology quality (OQAQ): The score range was [1.5, 6.5] and the average score was 4.66±0.92. 6 studies got more than 6 score. 106(74.7%)reviews′scores were lower than the average . The search strategies of 128 reviews were not complete. 90 reviews mentioned that the selection bias can be avoided, but only 7 of them introduced the methods in detail. 101 reviews did not take the quality evaluation comprehensively. Report quality (PRISMA): The score range was [8.5, 24] and the average score was 15.28±2.91. Among them, only one review′score was over 24, 79 reviews were below 15 score. 130 reviews did not describe the detailed research process,122 reviews did not describe the detailed screening process and 118 reviews did not describe the detailed data extraction process, 94 reviews lacked quality assessment on included original studies,130 reviews did not avoid bias and all reviews lacked the description for the role of funders played in the research.ConclusionThe Chinese systematic reviews in public health has gained big progress in recent years. The exisitng reviews have covered most areas and hot research areas of public health and key research orientations have emerged. But there are still blanks in some areas. The scores of quality of the reviews on six major diseases were at the middle level. There are still some defects in methodology and report. Reviewers should pay more attention to the clear description of the methods in search, screening, data extraction, quality assessment, risk of bias and data analysis in future reviews and reporting. |
PDF Full Text Request