Article type
Abstract
Background
Decision-making in healthcare interventions typically relies on high-certainty evidence from randomized controlled trials (RCTs). However, finding and using solely high-certainty evidence from RCTs is uncommon for many health interventions and questions. Nonrandomized studies of interventions (NRSIs) offer valuable information as supplementary evidence. Authors of evidence synthesis using GRADE including both types of study designs need proper guidance.
Methods
This article presents guidance developed through an iterative process with discussions within the GRADE project group, in-person and online meetings, and email communication. Results were summarized and feedback was sought from GRADE Working Group experts.
Results
Guidance is offered for synthesizing evidence from both RCTs and NRSIs in health interventions. Specific advice is provided for assessing their interaction on the certainty of evidence (CoE) and decision frameworks for including one or both types of evidence. The congruency of effect estimates is highlighted as a crucial consideration, with approaches provided for making judgments about congruency and the overall certainty of evidence when pooling RCTs and NRSIs is feasible.
Conclusion
The GRADE guidance addresses the challenge faced by evidence synthesizers and healthcare decision-makers when high-certainty evidence from RCTs is lacking, which is often the case. This guidance enhances the robustness of decision-making in healthcare interventions by offering a structured approach to integrating evidence from RCTs and NRSI.
Decision-making in healthcare interventions typically relies on high-certainty evidence from randomized controlled trials (RCTs). However, finding and using solely high-certainty evidence from RCTs is uncommon for many health interventions and questions. Nonrandomized studies of interventions (NRSIs) offer valuable information as supplementary evidence. Authors of evidence synthesis using GRADE including both types of study designs need proper guidance.
Methods
This article presents guidance developed through an iterative process with discussions within the GRADE project group, in-person and online meetings, and email communication. Results were summarized and feedback was sought from GRADE Working Group experts.
Results
Guidance is offered for synthesizing evidence from both RCTs and NRSIs in health interventions. Specific advice is provided for assessing their interaction on the certainty of evidence (CoE) and decision frameworks for including one or both types of evidence. The congruency of effect estimates is highlighted as a crucial consideration, with approaches provided for making judgments about congruency and the overall certainty of evidence when pooling RCTs and NRSIs is feasible.
Conclusion
The GRADE guidance addresses the challenge faced by evidence synthesizers and healthcare decision-makers when high-certainty evidence from RCTs is lacking, which is often the case. This guidance enhances the robustness of decision-making in healthcare interventions by offering a structured approach to integrating evidence from RCTs and NRSI.