Article type
Year
Abstract
Background: The application of GRADE (Grading of Recommendations Assessment, Development and Evaluation) in systematic reviews (SRs) of animal studies can promote the translation from bench to bedside.
Objectives: We aim to explore the use of GRADE in SRs of animal studies.
Methods: We used a theoretical analysis method to explore the use of GRADE in SRs of animal studies and applied in one SRs of animal studies. Meanwhile, we discussed our results with relevant experts in two international conferences.
Results: Five downgrading factors were considered in SRs of animal studies.
1. Risk of bias: the SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation) tool can be used for assessing the risk of bias of animal studies.
2. Indirectness: we can assess indirectness in systematic reviews of animal studies from the PICO (population, intervention, comparator, outcomes).
3. Inconsistency: similarity of point estimates, extent of overlap of confidence intervals and statistical heterogeneity are also suitable to evaluate inconsistency of evidence from animal studies.
4. Imprecision: optimal information size (OIS) and 95% confidence intervals (CIs) are also suitable for systematic reviews of animal studies, like those of clinical trials.
5. Publication bias: we need to consider publication bias comprehensively through qualitative and quantitative methods.
Conclusions: The methods for the use of GRADE in systematic review of animal studies are explicit. However, the principle about GRADE in developing the policy based on the evidence from animal studies when there is an emergency of public health.
Objectives: We aim to explore the use of GRADE in SRs of animal studies.
Methods: We used a theoretical analysis method to explore the use of GRADE in SRs of animal studies and applied in one SRs of animal studies. Meanwhile, we discussed our results with relevant experts in two international conferences.
Results: Five downgrading factors were considered in SRs of animal studies.
1. Risk of bias: the SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation) tool can be used for assessing the risk of bias of animal studies.
2. Indirectness: we can assess indirectness in systematic reviews of animal studies from the PICO (population, intervention, comparator, outcomes).
3. Inconsistency: similarity of point estimates, extent of overlap of confidence intervals and statistical heterogeneity are also suitable to evaluate inconsistency of evidence from animal studies.
4. Imprecision: optimal information size (OIS) and 95% confidence intervals (CIs) are also suitable for systematic reviews of animal studies, like those of clinical trials.
5. Publication bias: we need to consider publication bias comprehensively through qualitative and quantitative methods.
Conclusions: The methods for the use of GRADE in systematic review of animal studies are explicit. However, the principle about GRADE in developing the policy based on the evidence from animal studies when there is an emergency of public health.