We consider a clinical trial in which the outcome can be assessed by a
continuous measure and where dropouts tend to have poorer efficacy th
an completers. When each subject can act as his/her own control, effic
acy is measured by the difference between the outcome measurements at
two times. When all subjects complete the protocol, a paired t-test ca
n be used to test for a treatment effect, i.e., whether or not the mea
n difference is zero. When a patient does not return for the final eva
luation, a measure of efficacy cannot be computed for that subject. Of
ten, data from dropouts are ignored and only the observed pairs are us
ed to analyze the data. When the reason for dropping out is not random
, the result may be misleading. In this paper, we assume that (1) the
distribution of the measure of efficacy (i.e., the change between two
outcome measurements) is Gaussian, (2) dropouts would have worse effic
acy than the median if they were observed, and (3) the dropout rate is
less than 50%. We propose a median-based t-like statistic using the s
ample median in place of the sample mean. The variance of the median i
s estimated using only data from the complete half-sample, i.e., the h
alf-sample with better efficacy. Simulations under five patterns of dr
opouts are performed to compare the proposed statistic with the paired
t-test. The results show that the median-based statistic provides a c
onservative bound for the test of significance of the treatment. Ln co
ntrast, because the paired t-test does not preserve its level of signi
ficance, except when the dropout mechanism is uniform, the paired t-te
st should not be used for trials in which dropouts tend to have poorer
efficacy than completers. (C) Elsevier Science Inc. 1997.