Premium
P2‐346: INNOVATIONS IN CLINICAL TRIALS: IMPROVING STUDY START UP (SSU)
Author(s) -
Mohs Richard,
Bork Jason,
Goldfeder Gabe,
Dwyer John,
Beauregard Doug
Publication year - 2018
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2018.06.1036
Subject(s) - commit , benchmark (surveying) , clinical trial , gap analysis (conservation) , site selection , business , computer science , operations management , medicine , engineering , political science , geography , database , cartography , ecology , biodiversity , pathology , law , biology
Background: Empirical evidence regarding minimal clinically important difference (MCID) estimates for clinical outcome assessments across the Alzheimer’s disease (AD) spectrum is limited. Using anchor-based and distribution-based approaches, this study estimated the MCID for the Mini Mental State Examination (MMSE), Clinical Dementia Rating (CDR) Scale global and sum of boxes (SB) scores, and the Functional Assessment Questionnaire (FAQ) for patients stratified by cognitive impairment stage.Methods:Patients with 2 visits to Alzheimer’s Disease Centers (ADCs) were identified from the National Alzheimer’s Coordinating Center Uniform Data Set (9/2005-9/2016), and stratified into the following cohorts (Table 1): 1) Normal cognition, 2) Mild cognitive impairment (MCI)-AD, 3) Mild AD dementia, 4) Moderate-severe AD dementia. For each cohort, mean changes from previous visit, effect size (ES), standardized response mean (SRM), and half of the standard deviation at baseline (1⁄2 SD) for MMSE, CDR, and FAQ scores were estimated and stratified by whether, in the clinician’s assessment, there was a meaningful decline in cognitive, behavioral, or functional attributes from the previous visit. Analyses were performed at a visit level. Results:Using a clinician’s assessment of meaningful decline since the previous visit as an anchor, on average, a 2-3 point decrease in MMSE, 0.2-0.3 point increase in global CDR, 1-2 point increase in CDR-SB, and 3-5 point increase in FAQ were indicative of a meaningful decline (Table 2; overall). The proportion of visits with meaningful decline increased with more advanced stages of cognitive impairment (Table 2). TheMCID values also generally increased by stage, and the observed ES/SRM for those with meaningful decline were consistently in the acceptable ranges for MCID (ES: 0.2-0.5; SRM: 0.4-0.8), whereas the ES/SRM among visits with no meaningful decline were very small. Results from the distribution-based method (1⁄2 SD) supported the anchor-based approach. Conclusions: These findings provide estimates of MCID based on clinical practice in ADCs which may be used to inform the design and interpretation of future trials. Application of these MCID estimates as responder definitions in contemporary clinical trials warrants further exploration. The findings also suggest that it is important to account for variation in MCID by disease severity.