Regression to the mean in multiple sclerosis

Area of Science:

• Neurology
• Clinical Trials
• Biostatistics

Background:

• Relapsing remitting multiple sclerosis (RRMS) patients in clinical trials often have high baseline disease activity.
• High baseline activity can lead to an unusual disease course in the pre-study period.
• This can result in a tendency towards more typical disease behavior during follow-up, a phenomenon known as regression to the mean.

Purpose of the Study:

• To evaluate the relevance of regression to the mean in a non-treated cohort of RRMS patients.
• To assess the impact of regression to the mean on relapse rates in RRMS.
• To inform the interpretation of long-term studies without placebo controls.

Main Methods:

• Prospective follow-up of 44 definite RRMS patients.
• Inclusion criteria: two or more relapses in the previous two years and baseline Expanded Disability Status Scale (EDSS) ≤ 5.5.
• Analysis of relapse rates before and during the follow-up period.

Main Results:

• The mean number of relapses decreased from 1.72 (SD: 1.4) in the year prior to enrollment to 1.0 (SD: 1.3) during the first year of follow-up (P < 0.05).
• Regression to the mean may explain up to 40% of the reduction in relapse rate observed from the baseline to the on-study period.
• This indicates a significant statistical effect influencing observed outcomes.

Conclusions:

• Regression to the mean is a relevant phenomenon in RRMS patient cohorts selected for high baseline activity.
• It can substantially influence the interpretation of relapse rate reductions in non-placebo-controlled studies.
• Careful consideration of this statistical bias is crucial for accurate clinical trial interpretation in multiple sclerosis research.