| Optimizing the Hachinski Ischemic Scale. | |
| | |
MedLine Citation:
|
PMID: 21987392 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
BACKGROUND: Vascular causes and factors remain the most significant preventable component of cognitive disorders of elderly individuals. The Hachinski Ischemic Score (HIS) is the questionnaire most commonly used for diagnosis of vascular dementia. OBJECTIVE: To consolidate and further validate the HIS. DESIGN: The Canadian Study for Health and Aging was used for this study. It was a cohort study conducted in 3 waves in 1991, 1996-1997, and 2001-2002. The HIS containing 13 items was subjected to correspondence analysis to identify its optimal scaling of item scores and minimal set of items while maximizing the explainable variance. SETTING: A community-based cohort study. PATIENTS: For this analysis, we used 2968 of 3054 well-characterized and well-diagnosed cases with complete HIS data (86 cases had ≥1 item missing) from Canadian Study for Health and Aging phases 2 (1996-1997; n = 2431) and 3 (2001-2002; n = 623). RESULTS: Two optimized HIS versions were identified that classify patients with vascular dementia vs those with nonvascular dementia as well as or more accurately than the original HIS instrument. Assuming the HIS instrument measures only a single dimension, correspondence analysis identified the 7 most discriminative HIS items. Binary scoring (0, 1) of these items led to a 7-item HIS model that classified as well as the original 13-item HIS instrument. By merging highly similar HIS items and applying correspondence analysis, a 5-item composite HIS model was created that measures 2 meaningful dimensions of information and classified vascular vs nonvascular dementia better than the original HIS instrument. Each HIS version developed has specific advantages and disadvantages in terms of simplicity, scoring, generalizability, and accuracy. CONCLUSION: Depending on the specific setting, 2 reduced HIS versions consisting of 5 composite-question items or 7 single-question items classify as well as or better than the original HIS instrument. |
| | |
Authors:
|
Vladimir Hachinski; Shahram Oveisgharan; A Kimball Romney; William R Shankle |
Publication Detail:
|
Type: Journal Article; Research Support, Non-U.S. Gov't; Validation Studies Date: 2011-10-10 |
Journal Detail:
|
Title: Archives of neurology Volume: 69 ISSN: 1538-3687 ISO Abbreviation: Arch. Neurol. Publication Date: 2012 Feb |
Date Detail:
|
Created Date: 2012-02-14 Completed Date: 2012-04-09 Revised Date: 2012-10-11 |
Medline Journal Info:
|
Nlm Unique ID: 0372436 Medline TA: Arch Neurol Country: United States |
Other Details:
|
Languages: eng Pagination: 169-75 Citation Subset: AIM; IM |
Affiliation:
|
Department of Clinical Neurological Sciences, University Hospital, University of Western Ontario, London, Ontario, Canada. vladimir.hachinski@lhsc.on.ca |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Aged Aged, 80 and over Aging / physiology Brain Ischemia / diagnosis*, psychology Canada / epidemiology Cognition Disorders / etiology, psychology Cohort Studies Female Humans Image Processing, Computer-Assisted Longitudinal Studies Male Middle Aged Models, Statistical Neuropsychological Tests / standards* Reproducibility of Results |
| Comments/Corrections | |
Comment In:
|
Arch Neurol. 2012 Aug;69(8):1084-5; author reply 1085
[PMID:
22892672
]
|
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Cause-specific mortality among a cohort of U.S. flight attendants.
Next Document: Association Between Immediate Initiation of Intramuscular Interferon Beta-1a at the Time of a Clinic...