Table 4Comparison of six frameworks for assessing the evidence about global health interventions

FrameworksDomains
Assessment of Strength of Evidence
How Explicit?
How Strict?
Assessment Of Magnitude Of Benefit vs. HarmAssessment of ContextAssessment of Implementation
How Much is Known?
Assessment of Implementation
How Easy is Implementation?
CostsSustainabilityComments/Ease of Application
Tang et al.13Not as explicit as some others, not as strict as GRADEExplicit assessment of magnitude of benefit, no explicit assessment of harmPotentially assessed as part of “widely demonstrated”Not directly assessedCould potentially be included in repeatabilityNot assessedNot assessedRelatively easy to apply, but operationalizing “repeatability” is a challenge
GRADE14Very explicit, strictest criteriaExplicit assessmentNot assessedNot assessedNot assessedNot assessed*Not assessedEasy to apply, widely used and recognized framework
HASTE15Not explicitly assessed, more lenient criteria than GRADENot explicitly assessedNot assessedDetailed assessment of implementation science dataNo explicit assessment of ease of implementation but could readily be incorporatedNot assessedNot assessedRelatively easy to apply, some challenge with categories (for example, the situation where efficacy is consistent, biologically plausible and implementation data scarce has no category). The published example contains far more data collection on implementation than what would normally be feasible
Community Preventive Services Task Force16Explicitly assessed, more lenient criteria than GRADEExplicit assessment of benefits with expert judgment for threshold, no explicit assessment of harmsNot assessedNot assessedNot assessedNot assessedNot assessedEasy to apply
Australian NHMRC17Explicitly assessed, more lenient criteria than GRADEExplicit assessmentYesNot directly assessed, could potentially be included in applicabilityNot directly assessed, could potentially be included in applicabilityNot assessed, could potentially be included in applicability.Not directly assessed, could potentially be included in applicabilityVery difficult to apply. Potential is large for poor inter-rater reliability.
NHS Health Development Agency18Explicitly assessed, criteria approximately as strict as GRADENot explicitly assessedNot directly assessed Indirectly included in corroborationIncluded as a component of corroborationNot directly assessed, but potentially could be part of corroborationNot assessedNot assessedMost difficult to apply. Potential is large for poor inter-rate reliability
*

Costs are considered in GRADE's factors to consider when making a recommendation19

From: Results

Cover of Global Health Evidence Evaluation Framework
Global Health Evidence Evaluation Framework [Internet].
Shekelle PG, Maglione MA, Luoto J, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.