Article type
Abstract
Background: Minimal important differences (MIDs) are used in guidelines to determine the clinical importance of effects and to aid decision making alongside quality of the evidence, trade-offs between benefits and harms, current context and equality considerations. The EQ-5D is a preference based health-related quality-of-life measure for which there is no agreed MID.
Objectives: Preliminary work undertaken during development of a NICE guideline explored the variation in opinions and methods used to apply MIDs for the EQ-5D. This research furthers that work by assessing the impact the choice of MID had on the recommendations made and the effect alternative methods may have had, with the overall objective of informing future guideline-development methodology.
Methods: Evidence from the EQ-5D in one clinical guideline (NICE NG59) was reviewed to determine whether the clinical importance rating and/or recommendations may have differed if alternative methodologies had been employed. The methods that were compared were; 1) committee determined MID of 0.03 on a 0-1 scale; 2) considering any positive change in the EQ-5D as clinically important; 3) only assessing the EQ-5D as an economic outcome; and, 4) GRADE default values for imprecision and clinical importance.
Results: Data were available for 43 comparisons within 16 review questions. Use of the GRADE default value resulted in 26 changes to clinical importance when compared to a committee determined MID. Considering any change to be clinically important resulted in only 5 changes in this decision. The key finding was that irrespective of the methodological approach applied, the recommendations that resulted did not change.
Conclusions: The choice of criteria used to determine clinical importance of a single outcome within a review does not appear to impact on the recommendation made. This is likely due in part to the fact that decision making within a guideline considers the whole body of evidence, including other critical and important outcomes, quality of the evidence, cost-effectiveness, safety and tolerability, patient acceptability, current context and equality issues, amongst other considerations.
Objectives: Preliminary work undertaken during development of a NICE guideline explored the variation in opinions and methods used to apply MIDs for the EQ-5D. This research furthers that work by assessing the impact the choice of MID had on the recommendations made and the effect alternative methods may have had, with the overall objective of informing future guideline-development methodology.
Methods: Evidence from the EQ-5D in one clinical guideline (NICE NG59) was reviewed to determine whether the clinical importance rating and/or recommendations may have differed if alternative methodologies had been employed. The methods that were compared were; 1) committee determined MID of 0.03 on a 0-1 scale; 2) considering any positive change in the EQ-5D as clinically important; 3) only assessing the EQ-5D as an economic outcome; and, 4) GRADE default values for imprecision and clinical importance.
Results: Data were available for 43 comparisons within 16 review questions. Use of the GRADE default value resulted in 26 changes to clinical importance when compared to a committee determined MID. Considering any change to be clinically important resulted in only 5 changes in this decision. The key finding was that irrespective of the methodological approach applied, the recommendations that resulted did not change.
Conclusions: The choice of criteria used to determine clinical importance of a single outcome within a review does not appear to impact on the recommendation made. This is likely due in part to the fact that decision making within a guideline considers the whole body of evidence, including other critical and important outcomes, quality of the evidence, cost-effectiveness, safety and tolerability, patient acceptability, current context and equality issues, amongst other considerations.