Skip to main content
ARS Home » Northeast Area » Boston, Massachusetts » Jean Mayer Human Nutrition Research Center On Aging » Research » Publications at this Location » Publication #230703
Title: Reporting of Systematic Reviews of Micronutrients and Health: A Critical Appraisal

Author
item CHUNG, MEI - TUFTS MEDICAL CENTER
item BALK, ETHAN - TUFTS MEDICAL CENTER
item IP, STANLEY - TUFTS MEDICAL CENTER
item RAMAN, GOWRI - TUFTS MEDICAL CENTER
item YU, WINIFRED - JM USDA HNRCA @ TUFTS
item TRIKALINOS, THOMAS - TUFTS MEDICAL CENTER
item Lichtenstein, Alice
item YETLEY, ELIZABETH - ODS/NIH
item LAU, JOSEPH - TUFTS MEDICAL CENTER

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/27/2009
Publication Date: 4/1/2009
Citation: Chung, M., Balk, E.M., Ip, S., Raman, G., Yu, W.W., Trikalinos, T., Lichtenstein, A.H., Yetley, E.A., Lau, J. 2009. Reporting of Systematic Reviews of Micronutrients and Health: A Critical Appraisal. American Journal of Clinical Nutrition. 89:1099-1113.

Interpretive Summary: Systematic reviews are valuable tools for evaluating nutrition-related research and formulating nutrition recommendations. However, the lack of standardized criteria for generating these reviews has restricted their potential utility. The objective of this paper is to evaluate the characteristics and reporting quality of published systematic reviews linking micronutrients and health outcomes using human data. We completed a Medline search up to July 2007 for English-language systematic reviews linking micronutrients and health outcomes in human. A standardized extraction form was developed to collect data elements pertaining to the clarity of reporting of methods, results, nutrient or diet variables, quantitative analyses, and presence of future research recommendations. Descriptive analyses on reporting items related to study quality were performed. We found 139 eligible systematic reviews of 21 micronutrients in 84 journals. Cardiovascular diseases and cancer were the most common clinical conditions. Among the reviews, 88 included intervention studies alone, 31 observational studies alone, and 20 both study types. Systematic reviews including intervention studies alone had more complete reporting than those including observational studies alone or both types of study design. Quality scales or checklists to assess the methodological quality of the primary studies were used by 57 systematic reviews. Systematic reviews containing meta-45 analyses were more likely to be published in higher impact factor journals and receive more citations. In summary, the reporting of many published systematic reviews on micronutrients and health outcomes are inadequate. This limits their potential use as a tool when formulating nutrition related policies or recommendations. We propose a checklist of reporting items to be considered in future nutrition-related systematic reviews.

Technical Abstract: Context: Systematic review is a valuable tool for evaluating nutrition-related research and formulating nutrition recommendation. Lack of standardized criteria for generating these reviews has restricted their potential utility. Objective: To evaluate the characteristics and reporting quality of published systematic reviews of micronutrients and health outcomes. Data Sources and Study Selection: Medline search up to July 2007 for English-language systematic reviews linking micronutrients and health outcomes in human. Data Extraction and Analyses: A standardized extraction form was used to collect data elements pertaining to the clarity of reporting of methods, results, nutrient or diet variables, quantitative analyses, and presence of future research recommendations. Descriptive analyses on reporting items related to study quality were performed. Results: We found 139 eligible systematic reviews of 21 micronutrients in 84 journals. Cardiovascular diseases and cancer were the most common clinical conditions. Among the reviews, 88 included intervention studies alone, 31 observational studies alone, and 20 both study types. Systematic reviews including intervention studies alone had more complete reporting (median number of reporting items: 17, interquartile range, IQR: 13, 19) than those including observational studies alone (median: 14, IQR: 13, 16) or both types of study design (median: 14, IQR: 12, 15) (P=0.002). Quality scales or checklists to assess the methodological quality of the primary studies were used by 57 systematic reviews (41%; 48 evaluated intervention studies, 1 evaluated observational studies, and 8 evaluated both designs). Systematic reviews containing meta-45 analyses were more likely to be published in higher impact factor journals (P=0.002) and to receive more citations (P=0.001).