Premium
Does the evidence say a 4‐point change in ADAS‐cog score is clinically significant?
Author(s) -
Oremus Mark
Publication year - 2014
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.2013.12.017
Subject(s) - biostatistics , citation , epidemiology , library science , medicine , cog , clinical epidemiology , computer science , artificial intelligence , pathology
The cognitive subscale of the Alzheimer’s Disease Assessment Scale (ADAS-cog) [1] is the primary measure of cognitive change in Alzheimer’s disease (AD) drug trials. Conventional wisdom says at least a 4-point change in ADAS-cog score is clinically significant. Research articles (e.g., Matthews et al. [2], Farlow et al. [3]) and clinical practice guidelines [4] accept this wisdomwithout citing seminal publications that establish the validity of the 4-point threshold. Analysis of data from one trial [5] suggests that no ADAS-cog threshold can classify persons with AD according to clinical response. The validity of attributing clinical significance to a 4point change in ADAS-cog score is important to address from a policy perspective. Research has shown that establishing clinically important thresholds for AD medications can affect public insurance reimbursement recommendations for these drugs [6]. This is because statistically significant changes in score are not always clinically meaningful, but the extra step of assigning clinical significance to a certain minimum change in the score could enhance the credibility of medications that meet or exceed the minimum. The current group of accepted AD medications (i.e., cholinesterase inhibitors and memantine) has been reimbursed in many jurisdictions regardless of the demonstrated change in score on the ADAS-cog, largely because of the lack of alternative treatments for the disease. However, policy makers may require newer AD medications to demonstrate greater efficacy as a precondition of reimbursement, especially in jurisdictions with limited resources. Therefore, the context within which evidence is considered “clinically significant” could become much more important for policy making. To identify English language articles that might explain why a 4-point change in ADAS-cog score is clinically significant, I searched MEDLINE, EMBASE, and PsycINFO for articles published between January 1985 and June 2013. The following keyword string governed the search: Alzheimer AND (ADAS OR ADAS-cog OR Alzheimer’s Disease Assessment Scale). In total, the search yielded 719 MEDLINE citations, 1253 EMBASE citations, and 294 PsycINFO citations. I examined the titles and abstracts of these citations as a first step to identifying relevant articles. If I was