SIB developed for evaluation of cognitive function in participants, who demented to a degree that they cannot complete conventional neuropsychological testing. Test items consisted of simple, one-step commands presented with gestural cues and instructions that were repeated if necessary. SIB test consisted of 51-item scale, divided into 9 subscales: social interaction (0-6), memory (0-14), orientation (0-6), language (0-46), attention (0-6), praxis(0-8), visuospatial ability(0-8), construction(0-4), orienting to name(0-2). Total possible score:0-100; lower score = greater cognitive impairment.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

ADCS-ADLsev: 19-item scale measures basic and instrumental abilities in participant population and had good metric properties and reliability in detecting change. Individual score range: 0 to 5 for telephone, 0 to 4 for dressing, watch television, get around outside home, 0 to 3 for eating, walking, toilet, bathing, grooming, conversation/small talk, clear dishes, find personal belongings, obtain beverages, dispose of garbage, left on own, 0 to 1 for run water from and turn off faucet to wash hands, turn on and off light. Total score range: 0 to 54 lower scores=greater functional impairment.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

NPI:12-domain caregiver assessment of behavioral disturbances occurring in dementia: delusions, hallucinations, agitation/aggression, depression/dysphoria, anxiety, elation/euphoria, apathy/indifference, disinhibition, irritability/lability, motor disturbance, appetite/eating, nighttime behavior. Severity(1=Mild to 3=Severe),frequency(1=occasionally to 4=very frequently) scales recorded for each domain; frequency*severity=each domain score(range 0-12). Total score=sum of each domain score(range 0-144);higher score=greater behavioral disturbances;negative change score from baseline=improvement.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

NPI is a 12-domain caregiver assessment of behavioral disturbances occurring in dementia. Severity (1=Mild to 3=Severe) and frequency (1=occasionally to 4=very frequently) scales were recorded separately for each domain and their product gives individual domain score (range 0-12). Sum of delusions and hallucinations sub-domain scores of NPI was calculated as a measure of Alzheimer's Disease (AD) related psychosis. Total possible score range: 0-24 with higher score indicating greater behavioral disturbances.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

MMSE measured general cognitive functioning: orientation, memory, attention, concentration, naming, repetition, comprehension, and ability to create a sentence and to copy two intersecting polygons. Total score derived from sub-scores; total ranged from 0 to 30, higher score indicates better cognitive state.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

Alzheimer's Disease Cooperative Study-Clinical Global Impression of Change (ADCS-CGIC) version of CIBIC-Plus was used to assess participant's overall disease severity. The corresponding baseline assessment rated participant on 7-point scale: (1) extremely severe AD to (7) no symptoms of AD. Overall impression of change from baseline was rated on a 7-point scale: 1=marked improvement; 2=moderate improvement; 3=minimal improvement; 4=no change; 5=minimal worsening; 6=moderate worsening; 7=marked worsening; all assessments are relative to baseline. Higher score = worsening of global function.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

RUD Lite: instrument used to assess amount of both formal and informal resources used by demented participants and primary caregiver. It was completed by caregivers and compiles data on following resources: use of social services, frequency and duration of hospitalizations, all contacts with health care professionals, participant living accommodations, amount of time the caregiver spends giving care and the impact of care giving on the caregiver's job. Overall cost of care was evaluated to quantify resources utilized.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

EQ-5D: participant rated questionnaire to assess health-related quality of life in terms of a single utility score. Health State Profile component assesses level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, and anxiety and depression; 1 indicates better health state (no problems); 3 indicates worst health state. Total possible score ranged from 5 to 15; lower score indicated a better health state.

Data for this outcome measure was collected and reported in individual participant listings but not statistically summarized for analysis due to early termination of the study.

Data for this Outcome Measure are not reported here because the analysis population includes participants who were not enrolled in this study. ClinicalTrials.gov is designed for reporting results from only those participants who were enrolled in the study and described in the Participant Flow and Baseline Characteristics modules.

Any untoward medical occurrence (including clinially important changes in clinical safety laboratory assessments, electrocardiograms and vital signs) in a participant who received study treatment was considered an AE without regard to possibility of causal relationship.

Safety analysis set included all participants who received at least 1 dose of study medication, including partial doses.