A comparison of randomized concurrent control groups with matched historical control groups: are historical controls valid?

Abstract

The use of a historical control group is predicated on the assumption that survival and relapse-free survival in the historical control group closely approximate the survival and relapse-free survival in a randomized concurrent control group. This assumption has never been tested. This study compares survival and relapse-free survival in randomized control groups with historical control groups matched for disease, stage, and follow-up. Of the 43 matched control groups, 42% varied by more than 10 percentage points, 21% varied by more than 20 percentage points, and 5% varied by more than 30 percentage points. Of the 18 that varied by greater than 10 percentage points, 17 had superior survival or relapse-free survival in the randomized concurrent control group. This study indicates that the assumption that historical control groups may replace randomized concurrent control groups is not valid.