To the Editor.
—I disagree with the statement in the article on tacrine by Dr Knapp and colleagues1 that there was no evidence that the large number of patient withdrawals biased the overall conclusions of their study.The number of patient withdrawals was progressively higher for the higher doses (ie, 36% for group 1 [placebo], 53% for group 2, 68% for group 3, and 73% for group 4). These were very high rates. However, they did correspond with the higher doses and supported the dose-related toxicity. Table 2 of their article suggested that duration of therapy was also a factor.To me, their data suggested an important contribution from patient withdrawal. The variances for the indexes recorded in their Table 1 suggested significant variability and/or skewedness. But the most concerning observation is related to the changes in duration of Alzheimer's disease at the initiation of the study and at