Minimal important differences and self-identifying treatment response in chronic inflammatory demyelinating...

Area of Science:

• Neurology
• Clinical Trials
• Patient-Reported Outcomes

Background:

• Outcome measures are recommended for chronic inflammatory demyelinating polyneuropathy (CIDP).
• The implications of minimal important differences (MID) for determining responder status in CIDP are not well understood.
• The reliability of patient-reported treatment response relative to clinically meaningful changes needs further investigation.

Purpose of the Study:

• To determine the minimal important differences (MID) and minimum detectable change with 95% confidence intervals (MDC95) for four commonly used scales in CIDP.
• To assess the sensitivity of these scales, using MID-defined cutoffs (aMIDc), in identifying patients who self-report treatment response.
• To evaluate the reliability of patient-reported treatment response in relation to clinically relevant changes defined by MID.

Main Methods:

• Retrospective study of 72 patients with definite or probable CIDP.
• Calculation of MID and MDC95 for four outcome measures (Medical Research Council Sum Score, Overall Neuropathy Limitation Score (ONLS), centile Inflammatory Rasch-built Overall Disability Scale (cI-RODS), grip strength).
• Determination of scale sensitivities using MID-defined cutoffs (aMIDc) to detect self-reported treatment responders.

Main Results:

• MID was not valid for the Medical Research Council Sum Score as MDC95 exceeded MID.
• Optimal aMIDc values and sensitivities were found for ONLS (1; 84.7%), cI-RODS (8; 62.3%), and grip strength (4 kg; 79.1%).
• Using MID-defined improvement in any of ONLS, cI-RODS, or grip strength significantly increased sensitivity for detecting treatment responders compared to ONLS alone (P=.008). Patient-reported improvement demonstrated high reliability with MID-defined changes.

Conclusions:

• Amelioration of any one of ONLS, cI-RODS, or grip strength, defined by MID, provides optimal relevance and sensitivity for identifying self-reported treatment responders in CIDP.
• Patient reliability in single-question response ascertainment is high when correlated with MID-defined clinical relevance.
• These findings advocate for the use of multiple outcome measures in CIDP monitoring and support greater patient involvement in treatment assessment.