When assessing a correlation between two exposure or biological marker vari
ables, one sometimes encounters the problem of indeterminate values for one
of the variables due to an assay detection limit. In this event, investiga
tors often report correlation coefficients computed after removing the pair
s involving non-detectable values, or after substituting some small constan
t for those values. These ad hoc practices can lead to bias in both point a
nd confidence interval estimates of the true correlation coefficient. To ad
dress this issue, we consider two parametric techniques for estimating the
correlation in the presence of left censoring for one of the variables. The
first is a maximum likelihood approach, and the second is an adaptation of
multiple imputation motivated primarily by potential benefits in confidenc
e interval coverage. Both of the estimators studied reduce to the standard
Pearson's correlation coefficient in the event of no censoring, and hence a
re valid in cases where this measure would be appropriate for the complete
data. We assess these approaches empirically and contrast them with ad hoc
methods for estimating the correlation between cervicovaginal human immunod
eficiency virus (HIV) viral load measurements and CD4+ lymphocyte courts fr
om HIV positive women enrolled in a clinical trial conducted in Bangkok, Th
ailand. Copyright (C) 2001 John Wiley & Sons, Ltd.