Interpreting Change in Scores on Patient-Reported Outcome Instruments

Abstract

Interpreting change in scores on patient-reported outcome instruments is a key aspect of instrument development. Without interpretation guidelines, the clinical meaning of significant improvements observed within a treatment group cannot be ascertained. While the field has contemplated this topic for several decades, there remains inconsistency in terminology, methods, and application. Careful selection of methods can result in determining when change is meaningful, but researchers must keep an open mind to the methods that best fit their study and instrument. In many cases, anchor-based methods are appropriate, but the statistical model that evaluates them should be defensible (eg, linear regression, repeated-measures modeling, logistic regression). Sometimes, researchers must entertain the use of novel methods that may be more appropriate for their planned studies and instrument (eg, standard setting, exit interviews, conjoint analysis). The selection of the method is best supported by clear, transparent communication with the regulatory agency to ensure that the method can support its goals.

Adaptive behavior in autism: Minimal clinically important differences on the Vineland‐II

“…B). These MCID estimation methods are consistent with recommendations from scientific literature and federal guidance alike [Coon & Cappelleri, ; Crosby, Kolotkin, & Williams, ; Food and Drug Administration, ; King, ; Revicki, Hays, Cella, & Sloan, ]. At a high‐level, distribution‐based methods compute the MCID as the change required to exceed some proportion of the intrinsic variability within the affected population; these estimates approximate those achieved through other methods (Norman, Sloan, & Wyrwich, ).…”

Peak Pruritus Numerical Rating Scale: psychometric validation and responder definition for assessing itch in moderate‐to‐severe atopic dermatitis

“…Analyses were conducted using the trial data to define a clinically meaningful within‐person change, or response, by using both distribution‐ and anchor‐based methods . The distribution‐based method used one‐half SD of the average Peak Pruritus NRS at baseline.…”

Clinically important difference on the Erectile Dysfunction Inventory of Treatment Satisfaction questionnaire in patients with erectile dysfunction

“…The current results should be interpreted on the level of the average patient (ie, as the difference in mean EDITS index scores between treatment groups). A clinically important responder threshold for individual patients (ie, change from baseline to follow‐up for an individual) may be higher.…”

“…With increased attention to PROs in ED and other settings, there has been a growing need to define thresholds that represent clinically relevant differences for a given PRO. Clinically important difference (CID) is defined in this manuscript as the difference in scores between treatment groups that can be considered clinically relevant, used specifically in the interpretation of group‐level differences and distinct from a clinically important responder threshold (for the interpretation of individual‐level changes) . Identifying the CID is essential for interpretation of the EDITS score and understanding the clinical relevance of EDITS outcomes with respect to treatment differences.…”

“…Clinically important difference (CID) is defined in this manuscript as the difference in scores between treatment groups that can be considered clinically relevant, used specifically in the interpretation of group-level differences and distinct from a clinically important responder threshold (for the interpretation of individual-level changes). 9 Identifying the CID is essential for interpretation of the EDITS score and understanding the clinical relevance of EDITS outcomes with respect to treatment differences. Unlike the IIEF EF domain, for which a CID has been derived from a supplemental analysis capturing 17 randomised, double-blind, placebo-controlled trials, 10 no published value of a CID for EDITS was ever established.…”