Clinical trial results on the first proposed Alzheimer's disease drug were mixed, with some measures showing improvement in patients and others producing no significant statistical findings.

At a hearing before a Food and Drug Administration Advisory Committee, Warner-Lambert Co. said it believed clinical trials at 16 university hospitals of the drug Tacrine showed clear improvement in patients.But Ronald M. Cresswell, Warner-Lambert vice president, acknowledged that the results were unclear for one of two scales used to measure the drug.

Researchers assessed the effects of the drug by a test called the Alzheimer's Disease Assessment Scale, and by an objective evaluation by doctors called the Clinical Global Indication of Change.

Cresswell said that under the ADAS test, which examined patients' ability to recall words, there was an "overwhelmingly statistically significant" finding in favor of the drug.

However, the CGIC test produced no statistically significant results, he said.

"The study is going to be criticized, we know that."