The GRADE Working Group and CINeMA approaches provided inconsistent certainty of evidence ratings for a network meta-analysis of opioids for chronic noncancer pain

Abstract

ObjectiveAssessment of the certainty of evidence (CoE) from network meta-analysis is critical to convey the strength of inferences for clinical decision-making. Both the GRADE Working Group (GWG) and the CINeMA framework have been designed to assess the CoE of treatment effects informed by network meta-analysis; however, the concordance of results is uncertain. Study design and settingWe assessed the CoE for treatment effects of individual opioids on pain relief and physical functioning from a network meta-analysis for chronic noncancer pain using the GWG approach and the CINeMA framework. Both approaches evaluate the CoE as high, moderate, low or very low. We quantified the number of discrepant CoE ratings between approaches, and the magnitude of the difference (i.e., 1-level, 2-levels, or 3-levels). ResultsAcross 105 comparisons among individual opioids for pain relief, the GWG and CINeMA approaches provided different CoE ratings in 34% of cases (36 of 105). Across 66 comparisons for physical functioning, there was discordance in 17% of cases (11 of 66). All discrepancies were separated by 1-level. The CINeMA framework typically provided lower CoE ratings compared to the GWG approach, predominantly because of differences in the assessment of transitivity and heterogeneity. ConclusionOur findings suggest there are differences between the CoE ratings provided by the GWG and CINeMA approaches when applied to network meta-analyses. Further research is needed to replicate or refute our findings in other network meta-analyses and assess the implications for clinical decision-making.