General Information
Conflict of Interest
Request Clinical Trials
Associated Trial(s):- NCT00249158 - Risperidone in the Treatment of Behavioural and Psychological Signs and Symptoms in Dementia (BPSSD): a Multicentre, Double-blind, Placebo-controlled Parallel-group Trial
- NCT00253123 - A Randomized, Double-Blind, Placebo-Controlled Study of Risperidone for Treatment of Behavioral Disturbances in Subjects With Dementia
- NCT00249145 - Risperidone in the Treatment of Behavioral Disturbances in Demented Patients: an International, Multicenter, Placebo-controlled, Double-blind, Parallel-group Trial Using Haloperidol as Internal Reference
Request Clinical Trials
Data Request Status
Status: OngoingResearch Proposal
Project Title: Comparison of a clinically meaningful benefit for the 29-item vs 19-item Cohen-Mansfield Agitation Inventory
Scientific Abstract:
BACKGROUND
The original Cohen-Mansfield Agitation Inventory (CMAI) is a caregiver-rated questionnaire that quantifies the frequency of 29 agitated behaviors. Following the consensus definition for agitation in patients with cognitive disorders developed by the International Psychogeriatric Association (IPA) Agitation Definition Work Group, the CMAI-IPA version was developed. CMAI-IPA includes 19 items that are mapped to IPA agitation domains.
OBJECTIVE
To understand the clinical meaningfulness of symptom score changes in scales like CMAI (to support non-pharmacological and pharmacological management of agitation and aggression in people with Alzheimer's disease).
STUDY DESIGN & PARTICIPANTS
We will use individual participant data from the requested risperidone studies (all treatment arms will be included). We will include all participants with a predefined minimal severity of agitation at baseline and for whom a predefined minimal set of data is available. We will extract outcome data for baseline, Week 6 and Week 12 visits.
OUTCOME MEASURES
29-item CMAI; 19-item CMAI-IPA, Clinical Global Impression Severity (CGI-S), Clinical Global Impression Change (CGI-C)
ANALYSIS
We will perform anchor - based analysis of minimal clinically important differences (MCIDs) for CMAI and CMAI-IPA at two time points - 6 and 12 weeks. As anchors, we will use CGI-S, and CGI-C. There are no statistical hypotheses pre-specified and there will be no statistical comparisons performed between MCIDs estimated from three requested studies.
Brief Project Background and Statement of Project Significance:
Agitation is highly prevalent in and persistently associated with patients with Alzheimer's disease (AD). It is estimated that more than 70% of people with AD. It has long been recognized that the occurrence of NPS in dementia patients can accelerate disease progression, lead to early institutionalization, and interfere with treatment effects and prognosis. There are various efforts underway to develop safe and effective approaches for management of agitation (including aggression) in people with AD dementia.
Our ability to interpret agitation treatment trial outcomes is limited by insufficient evidence on the clinical meaningfulness of score changes on the most commonly used primary outcome scales employed in these trials.
The original Cohen-Mansfield Agitation Inventory (CMAI) is a caregiver-rated questionnaire that quantifies the frequency of 29 agitated behaviors. CMAI has been widely used in clinical research in Alzheimer's disease since more than four decades.
Following the recent consensus definition for agitation in patients with cognitive disorders developed by the International Psychogeriatric Association (IPA) Agitation Definition Work Group (Sano et al., 2024), the CMAI-IPA version was developed.
CMAI-IPA includes 19 items that are mapped to IPA agitation domains. It is this shorter version of the CMAI that is likely to be used in the future clinical studies.
Therefore, there is an urgent need to understand the clinical meaningfulness of symptom score changes in the 19-item CMAI (to support non pharmacological and pharmacological management of agitation and aggression in people with Alzheimer's disease) (de Mauleon et al., 2021).
For the original 29-item CMAI, minimal clinically important differences (MCIDs) are reasonably well established or at least there are several publications on this subject (including publications that used data from studies accessed via the YODA project). There is no possibility to derive MCIDs for the 19-item CMAI from the MCIDs established for the 29-item CMAI without repeating the analysis. Ideally, such analysis should be performed using several anchors scales and using data from several independent studies. For such analyses, participant-level data are needed.
We would like to repeat this exercise to estimate MCIDs for the 19-item CMAI-IPA scale.
Specific Aims of the Project:
OBJECTIVE:
To understand the clinical meaningfulness of symptom score changes in scales like CMAI (to support non pharmacological and pharmacological management of agitation and aggression in people with Alzheimer's disease).
HYPOTHESES:
There are no statistical or scientific hypotheses pre-specified.
For the 29-item CMAI, prior research established minimal clinically important symptom improvement over periods of either 4 weeks or 12 weeks using various approaches and using datasets from several clinical trials as well as the interviews from caregivers and physicians.
Our objective to estimate minimal clinically important differences (MCIDs) for a shorter 19-item CMAI-IPA using conventional anchor-based analysis. We will use data from three different studies using two different anchors at two different time points (6 weeks and 12 weeks). However, there will be no statistical comparisons performed between obtained MCID estimates.
Study Design: Individual trial analysis
What is the purpose of the analysis being proposed? Please select all that apply.: Develop or refine statistical methods Research on clinical trial methods
Software Used: R, RStudio
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
Exclusion criteria:
None
Inclusion criteria:
- CGI-S >= 4
- positive CMAI Factor 1 agitation (ie, >= 1 aggressive behavior occurring several times per week, or >= 2 aggressive behaviors occurring once or twice per week, or >= 3 aggressive behaviors occurring less than once per week) at Baseline.
- The participant exhibits at least one of the following behaviors that are associated with observed or inferred evidence of emotional distress (eg, rapid changes in mood, irritability, outbursts):
--- Excessive motor activity (eg, pacing, rocking, gesturing, pointing fingers, restlessness, performing repetitious mannerisms).
--- Verbal aggression (eg, yelling, speaking in an excessively loud voice, using profanity, screaming, shouting).
--- Physical aggression (eg, grabbing, shoving, pushing, resisting, hitting others, kicking objects or people, scratching, biting, throwing objects, hitting self, slamming doors, tearing things, and destroying property).
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
NOTE:
- For our analysis, we do not have primary / secondary outcomes. We only define the scale for which MCIDs are to be established and the anchor scales
- In the studies to which we request access, primary outcomes were different from those in which we are interested
- Therefore, the primary / secondary outcome definitions below are based on the use of the these scales in future clinical trials
Primary outcome measure:
- Cohen-Mansfield Agitation Inventory (CMAI)
Secondary outcome measure (anchor scales):
- Clinical Global Impression Severity
- Clinical Global Impression Change
Main Predictor/Independent Variable and how it will be categorized/defined for your study:
N/A
What we are planning to perform is essentially a regression analysis to relate changes in the primary scale of interest (CMAI) with those in the anchor scales.
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study: N/A
Statistical Analysis Plan:
As a moderate correlation of >=0.3 between anchor and a clinical outcome assessment (COA) scale has been recommended for the appropriate estimation of an MCID, we will calculate the combined and per visit Spearman's rank correlation coefficient, which assesses the cross-sectional monotonic relationship between two ordinal/continuous variables, for the following anchor-based MCID estimates: 1) between CMAI and CGI-S, and 2) CMAI change and CGI-C from baseline. We perform all analyses for both the 29-item and 19-item versions of CMAI.
We then estimate the change in CMAI score corresponding to a minimal improvement, defined as, 1) a 1-category improvement on the CGI-S scale, and 2) 'Minimal improvement' from baseline category on the CGI-C scale. Earlier studies of MCIDs in AD have used similar approaches. The 'minimal worsening' category of the CGI-C from baseline ratings served as clinician-rated anchors for minimal worsening on cognitive scales in individuals with mild cognitive impairment. For the data, which include at least two within-individual CMAI score change measures, CMAI change from baseline in a model that allows different intercepts for each participant is expected a better fit than one that assumed a single overall mean change in total score without accounting for grouping effects, based on the Akaike Information Criterion (AIC). Thus, linear mixed-effects regression models will be used for anchor-based MCID estimations between CMAI change and CGI-C from baseline scores (but not CMAI and CGI-S change scores between consecutive visits). For the CGI-C anchor, a binary indicator variable will be created to distinguish individuals who experienced a minimal improvement versus those who did not, and the estimated marginal means from a linear mixed-effects model was obtained to derive the mean CMAI change score (adjusted for random effects) for the minimal improvement group in the model. MCID estimates for the CGI-S anchor are calculated using the simple overall mean change in score corresponding to a one-unit improvement on the scales using all available measures.
Narrative Summary:
The original Cohen-Mansfield Agitation Inventory (CMAI) is a caregiver-rated questionnaire that quantifies the frequency of 29 agitated behaviors. Following the consensus definition for agitation in patients with cognitive disorders developed by the International Psychogeriatric Association (IPA) Agitation Work Group, the CMAI-IPA version was developed. CMAI-IPA includes 19 items that are mapped to IPA agitation domains. There is a need to understand the clinical meaningfulness of symptom score changes in scales like CMAI. Minimal clinically important differences (MCIDs) are established for the 29-item CMAI scale. We would like to perform anchor-based assessment of MCIDs for the 19-item CMAI-IPA scale.
Project Timeline:
Project start date: February 1, 2026
Analysis completion date: July 1, 2026
Manuscript submitted for publication: November 2026
Results reported back to the YODA Project: November 2026
Dissemination Plan: We are going to present the results at an Alzheimer's disease conference (e.g., AAIC) and will submit a full-length manuscript for publication at a journal like Alzheimer's Dementia
Bibliography:
De Mauleon, A., Delrieu, J., Cantet, C., et al. (2021). “Longitudinal Course of Agitation and Aggression in Patients with Alzheimer’s Disease in a Cohort Study: Methods, Baseline and
Longitudinal Results of the A3C Study.” J Prev Alzheimers Dis 8(2): 199--209.
Liu KY, Ivenso C, Howard R, et al. (2025) “Three approaches to determining clinically meaningful benefit on the Cohen-Mansfield Agitation Inventory in dementia clinical trials for agitation.” Alzheimer's Dement. 11:e70099. https://doi.org/10.1002/trc2.70099
Sano, M., Cummings, J., Auer, S., et al. (2024). “Agitation in cognitive disorders: Progress in the International Psychogeriatric Association consensus clinical and research definition.” Int
Psychogeriatr 36(4): 238--250