Ac. Lepri et al., A PRACTICAL APPROACH TO ADJUSTING FOR ATTRITION BIAS IN HIV CLINICAL-TRIALS WITH SERIAL MARKER RESPONSES, AIDS, 12(10), 1998, pp. 1155-1161
Objectives: To illustrate a simple approach to adjusting for bias due
to drop-outs (i.e., attrition bias) when evaluating the effect of a ce
rtain therapy in HIV clinical trials using the mean change in plasma v
iral load. To evaluate its validity and to compare its performance wit
h that of another simple method for handling dropouts: the last observ
ation carried forward (LOCF) method. Design: Data from a notional trea
ted group of 100 patients followed up to 52 weeks were generated. Attr
ition bias was introduced by mimicking selective patient dropout (i.e.
, more likely in patients doing badly). Methods: The difference betwee
n the true mean change in HIV RNA levels at 52 weeks and the observed
mean change because of drop-outs was calculated (attrition bias). The
reduction in bias obtained by using the proposed approach was then cal
culated and compared with that obtained by using the LOCF method. To a
ssess the performance of the methods over the entire follow-up, the me
an areas under the curves were considered. Results: Our method reduced
the bias by a clinically relevant amount in a variety of different se
ttings. In most of our simulations, bias was reduced by a larger amoun
t than that obtainable from using the LOCF method. Conclusions: The cu
rrent situation is that results from trials in HIV infection are invar
iably presented with no associated attempt to quantify the attrition b
ias present. Attrition-adjusted plots of mean change in HIV RNA should
, we believe, be presented alongside usual plots as a form of sensitiv
ity analysis. (C) 1998 Lippincott-Raven Publishers.