In many clinical trials, there are several endpoints of comparable importance, rather than one primary endpoint. In stroke treatment, a number of scales have been used to measure improvement in outcome and no one scale is believed to assess all dimensions of recovery. The restriction to one primary endpoint when designing or analyzing such a clinical trial may be inappropriate. We discuss several hypothesis tests for multiple endpoints in two-armed clinical trials as well as their implementation, including group sequential monitoring (1). Procedures to follow the primary hypothesis test in order to determine which individual endpoints differ will also be described (2). We apply the results to the NIH t-PA Stroke clinical trial, originally published in 1995 in the New England Journal of Medicine (3).

References:

1. Tang, D.-I., Geller, N.L. and Pocock, S.J. (1993). On the design and analysis of randomized clinical trials with multiple endpoints. Biometrics 49:23-30.
2. Tang, D.-I. and Geller, N.L. (1999). Closed testing procedures for group sequential clinical trials with multiple endpoints. Biometrics 55:1188-1192.
3. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group (1995). Tissue plasminogen activator for acute ischemic stroke. New England Journal of Medicine 333:1581-87.