Table 1 Summary of methods used to estimate responder definition (RD) thresholds

Anchor-based Approaches

Target Measures

➣ Mean Change

Mean change in NIS/NIS-LL scores for patients who did not exhibit meaningful worsening on the anchor

➣ Linear Regression

Linear regression models with change in NIS/NIS-LL scores as the outcome and change in the anchor (LLF/PND) as the predictor

➣ Receiver Operating Characteristic Curve

Receiver operating characteristic curves to identify the optimal cut-off point on NIS/NIS-LL scores for classifying patients showing meaningful worsening or not based on the anchor measure

Anchor: Polyneuropathy Disability (PND) score

• NIS Total

• NIS Sensation

• NIS-LL Sensation

Anchor: Lower Limb Function (LLF) test

• NIS-LL Total

• NIS-LL Muscle Weakness

Distribution-based Approaches

Target Measures

➣ Effect Size

Group difference or change over time relative to the standard deviation at baseline

➣ Standard Error of Measurement

Measurement error of a scale based on the standard deviation of baseline scores and the scale’s intra-rater reliability

➣ Standardized Response Mean

Group difference or change over time relative to the standard deviation of change scores

• All NIS and NIS-LL total and subscales