Some patients with Alzheimer's disease who receive cholinesterase inhibitors experience cognitive or behavioral symptomatic improvements. Others show no improvement at all. Mega and colleagues attempted to determine which baseline behavioral characteristics made patients with Alzheimer's disease more likely to respond with positive behavioral outcomes to donepezil.

Exclusion criteria included delirium, history of alcohol or substance abuse, history of psychiatric disorder preceding the onset of symptoms of dementia and other systemic or neurologic conditions that might affect cognitive function. All patients were administered a Mini-Mental State Examination to assess cognitive response. Patients were included if they agreed to take donepezil and if their caregivers were available for interviews. Additional inclusionary factors were an acquired persistent decline in at least three of the following areas: language, memory, visuospatial skills, cognition (e.g., calculation, judgment), emotion or personality. Patients took 5 mg of donepezil daily for four weeks, with an increase in dosage to 10 mg daily for the next four weeks. Behaviors were reviewed with the caregiver, and the caregiver was asked if the patient's behavior was different than it had been before the onset of the dementia and if that behavior had been present during the previous month. The behaviors measured by the Neuropsychiatric Inventory (NPI) included agitation, anxiety, apathy, abnormal motor output, depression, delusions, disinhibition, euphoria, hallucinations and irritability. After week 8 of therapy, patients who experienced a significant behavioral improvement were classified as responders, those who worsened were classified as nonresponders and all other patients were classified as “unchanged behaviorally.”

Of the 86 patients evaluated, almost one half (41 percent) showed an improved behavioral response (responders), 28 percent worsened (nonresponders) and 31 percent were unchanged. No significant cognitive improvement with donepezil therapy was exhibited within any group, and no significant differences in the concomitant psychotropic medications were observed during the eight weeks of treatment. Patients who responded to donepezil treatment had more behavioral disturbances across all NPI areas, except hallucinations. Patients who worsened behaviorally (nonresponders) showed a lower percentage of all behaviors, except hallucinations. Anxiety, delusions, depression and irritability were the specific behaviors that showed the largest change as compared with baseline for responders and nonresponders. The unchanged group was not included in this evaluation because they exhibited the fewest behavioral abnormalities at baseline. When nonresponders who were symptomatic at baseline were evaluated, significant deterioration of agitation, anxiety, delusions, depression, disinhibition and irritability was evident.

The authors suggest that patients with Alzheimer's disease who exhibit pretreatment delusions, agitation, depression, anxiety, apathy, disinhibition and irritability are more likely to improve with cholinesterase inhibitor (specifically, donepezil) treatment. The efficacy of cholinesterase inhibitors should be judged on the basis of pretreatment behavioral changes as well as cognitive changes. These changes help predict patient responses to therapy.