Article type
Year
Abstract
Background: Less than a quarter of developing countries are on-track for achieving the Millennium Development goal of halving under-nutrition. Rigorous and transparent systematic reviews are recognized internationally as a credible source for evidence of effects. However, in the nutrition field, there are a number of conflicting systematic reviews which has reduced their credibility.
Objective: We aimed to assess the reasons for conflicting findings of systematic reviews in four purposefully selected areas of nutrition.
Methods: We identified a purposive sample of systematic reviews in four areas of nutrition. We assessed the following possible reasons for differences in conclusions and Results: (1) methodological quality; (2) risk of bias assessment of included studies; (3) inclusion and exclusion criteria; (4) methods used (e.g. subgroup analyses, applicability assessment); and (5) external factors.
Results: We included 90 systematic reviews across nine content areas. We found that the definition of the question influenced the results of reviews. The generalizability was dependent upon which components of an intervention were included, whether the population was undernourished or healthy, whether outcomes were defined by clinical diagnosis or self-reported symptoms, and whether the control group received an active intervention or placebo. Methods used to conduct the systematic reviews also varied substantially across the 90 systematic reviews. There were variations in study designs, eligibility criteria and methods for assessing mediating effects.
Conclusion: There is a need for improved justification of methodological choices in systematic reviews on nutrition. We propose the need to follow standards for conducting and reporting systematic reviews, such as the Cochrane Handbook and PRISMA reporting guidelines. We propose there is a need to develop nutrition-specific guidance and centralized editorial review for nutrition-related reviews. The Cochrane and Campbell Collaborations are ideally positioned to carry this initiative forward due to their international reputation for rigour, transparency and freedom from conflicted funding.
Objective: We aimed to assess the reasons for conflicting findings of systematic reviews in four purposefully selected areas of nutrition.
Methods: We identified a purposive sample of systematic reviews in four areas of nutrition. We assessed the following possible reasons for differences in conclusions and Results: (1) methodological quality; (2) risk of bias assessment of included studies; (3) inclusion and exclusion criteria; (4) methods used (e.g. subgroup analyses, applicability assessment); and (5) external factors.
Results: We included 90 systematic reviews across nine content areas. We found that the definition of the question influenced the results of reviews. The generalizability was dependent upon which components of an intervention were included, whether the population was undernourished or healthy, whether outcomes were defined by clinical diagnosis or self-reported symptoms, and whether the control group received an active intervention or placebo. Methods used to conduct the systematic reviews also varied substantially across the 90 systematic reviews. There were variations in study designs, eligibility criteria and methods for assessing mediating effects.
Conclusion: There is a need for improved justification of methodological choices in systematic reviews on nutrition. We propose the need to follow standards for conducting and reporting systematic reviews, such as the Cochrane Handbook and PRISMA reporting guidelines. We propose there is a need to develop nutrition-specific guidance and centralized editorial review for nutrition-related reviews. The Cochrane and Campbell Collaborations are ideally positioned to carry this initiative forward due to their international reputation for rigour, transparency and freedom from conflicted funding.