The MMSE was a brief test that assessed the cognitive status of the participant. The 30-point test included items that evaluate orientation to time and place, immediate and delayed recall, attention, language, and construction. The total number of correct responses was obtained. The scores ranged from 0 to 30, with higher scores representing better performance. Summaries and analyses in the Intent-to-Treat (ITT) population were carried out using observed cases at each visit. A Study Endpoint evaluation was performed using the last observation carried forward (LOCF) from the open label treatment phase for each participant. The outcome of the study was based on analyses of the primary efficacy variable at Study Endpoint, which was defined as end of study assessment, using the ITT population with LOCF.

ITT population with LOCF included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable.

The CAS was a validated tool that measured the time caregivers spent aiding Alzheimer's participants with their day-to-day activities. The CAS recorded time spent on six activities of daily living, communicating with the person, using transportation, dressing, eating, looking after one's appearance, and supervising the person. Caregivers were asked to report the amount of time spent on each activity during a 'typical' caregiving day. Total time for the CAS was calculated as the sum of the sub-item times.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The CAS was a validated tool that measured the time caregivers spent aiding Alzheimer's participants with their day-to-day activities. The CAS recorded time spent on six activities of daily living, communicating with the person, using transportation, dressing, eating, looking after one's appearance, and supervising the person. Caregivers were asked to report the amount of time spent on each activity during a 'typical' caregiving day. Total time for the CAS was calculated as the sum of the sub-item times. Change from Baseline in CAS Total Time at Week 6 (Visit 3), Week 12 (Visit 4) or ET Visit, Week 12 LOCF (Study Endpoint) has been reported.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at Week 12 included ITT population with LOCF.

The MMSE was a brief test that assessed the cognitive status of the participant. The 30-point test included items that evaluated orientation to time and place, immediate and delayed recall, attention, language, and construction. The total number of correct responses was obtained. The scores ranged from 0 to 30, with higher scores representing better performance. Summaries and analyses in the ITT population were carried out using observed cases at each visit. A Study Endpoint evaluation was performed using the LOCF from the open label treatment phase for each participant. The outcome of the study was based on analyses of the primary efficacy variable at Study Endpoint, which was defined as end of study assessment, using the ITT population with LOCF. Change from Baseline in MMSE Total Score at Week 6 (Visit 3), Week 12 (Visit 4) or ET Visit, Week 12 LOCF (Study Endpoint) has been reported.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The NPI-8 was an 8-item scale that assessed eight behavioral domains: delusions, hallucinations, agitation, depression, anxiety, apathy, irritability, and aberrant motor behavior. The frequency (0 to 4) and severity (0 to 3) of each domain were assessed; the sub-score for each domain was calculated as the product of the frequency and severity rating. The total score for the NPI was calculated as the sum of the domain sub-score, range from 0 to 96, with higher scores indicating greater behavior disturbances.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The NPI-8 was an 8-item scale that assessed eight behavioral domains: delusions, hallucinations, agitation, depression, anxiety, apathy, irritability, and aberrant motor behavior. The frequency (0 to 4) and severity (0 to 3) of each domain were assessed; the sub-score for each domain was calculated as the product of the frequency and severity rating. The total score for the NPI was calculated as the sum of the domain sub-score, range from 0 to 96, with higher scores indicating greater behavior disturbances. Change from Baseline in NPI-8 Total Score at Week 12 (Visit 4) or ET Visit, Week 12 LOCF (Study Endpoint) has been reported.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The ADRQL was an observer-rated quality of life instrument that measured the following domains: social interaction, awareness of self, feelings and mood, enjoyment of activities, and response to surroundings. The ADRQL was a 47-item questionnaire with five domains: relating to and being around other people (ADRQL-A; 12 items), a person's special identity and important relationships (ADRQL-B; 8 items), different types of behavior (ADRQL-C; 15 items), usual activities (ADRQL-D; 5 items), and behavior in a person's living environment (ADRQL-E; 7 items). Domain sub-scores were summed to yield a total raw score, which was divided by the total possible score and multiplied by 100 to produce the final score. Each subscale score could be calculated in the similar approach. Total score ranges from 0-100 where higher scores reflected a better quality of life.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The ADRQL was an observer-rated quality of life instrument that measured the following domains: social interaction, awareness of self, feelings and mood, enjoyment of activities, and response to surroundings. The ADRQL was a 47-item questionnaire with five domains: relating to and being around other people (ADRQL-A; 12 items), a person's special identity and important relationships (ADRQL-B; 8 items), different types of behavior (ADRQL-C; 15 items), usual activities (ADRQL-D; 5 items), and behavior in a person's living environment (ADRQL-E; 7 items). Domain sub-scores were summed to yield a total raw score, which was divided by the total possible score and multiplied by 100 to produce the final score. Each subscale score could be calculated in the similar approach. Higher scores reflected a better quality of life. Change from Baseline in ADRQL Total Score at Week 12 (Visit 4) or ET Visit, Week 12 LOCF (Study Endpoint) has been reported.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The DAD was a 10 domain 40-item scale that measured a participant's ability to initiate, plan, organize, and perform both basic and instrumental activities of daily living. These domains were: hygiene (7 items), dressing (5 items), continence (2 items), eating (3 items), meal preparation (3 items), telephoning (4 items), going on an outing (5 items), finance (4 items), medication (2 items), and leisure (5 items). The three responses to the DAD items were "No, Yes, and N/A". A "No" answer scored 0 and a "Yes" answer scored 1. The scoring range was 0-40, with a higher score indicating less disability (better quality of life). If N/A was selected then it was treated as missing. When there were items with missing values, the domain sub-scores and the total score were imputed. Domain sub-scores were summed to yield a total raw score. This was divided by the total possible score and multiplied by 100 to produce the final score. Each subscale score can be calculated in the similar approach.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.

The DAD was a 10 domain 40-item scale that measured a participant's ability to initiate, plan, organize, and perform both basic and instrumental activities of daily living. These domains were: hygiene (7 items), dressing (5 items), continence (2 items), eating (3 items), meal preparation (3 items), telephoning (4 items), going on an outing (5 items), finance (4 items), medication (2 items), and leisure (5 items). The three responses to the DAD items were "No, Yes, and N/A". A "No" answer scored 0 and a "Yes" answer scored 1. The scoring range was 0-40, with a higher score indicating less disability (better quality of life). If N/A was selected then it was treated as missing. When there were items with missing values, the domain sub-scores and the total score were imputed. Domain sub-scores were summed to yield a total raw score. This was divided by the total possible score and multiplied by 100 to produce the final score. Each subscale score can be calculated in the similar approach.

ITT population included all enrolled participants who took at least one dose of study medication, and had baseline and at least one post-baseline assessment of at least one efficacy variable. Study endpoint evaluation at week 12 included ITT population with LOCF.