Assessment Scales for Disorders of Consciousness: Evidence-Based Recommendations for Clinical Practice and Research

Abstract

Report of the American Congress of Rehabilitation Medicine, Brain Injury-Interdisciplinary Special Interest Group, Disorders of Consciousness Task Force: Seel RT, Task Force Chair, Sherer M, Whyte J, Katz DI, Giacino JT, Rosenbaum AM, Hammond FM, Kalmar K, Pape TL, Zafonte R, Biester RC, Kaelin D, Kean J, Zasler N. Assessment scales for disorders of consciousness: evidence-based recommendations for clinical practice and research. Objectives To conduct a systematic review of behavioral assessment scales for disorders of consciousness (DOC); provide evidence-based recommendations for clinical use based on their content validity, reliability, diagnostic validity, and ability to predict functional outcomes; and provide research recommendations on DOC scale development and validation. Data Sources Articles published through March 31, 2009, using MEDLINE, CINAHL, Psychology and Behavioral Sciences Collection, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Cochrane Central Register of Controlled Trials, Biomedical Reference Collection, and PsycINFO. Thirteen primary terms that defined DOC were paired with 30 secondary terms that defined aspects of measurement. Scale names, abbreviations, and authors were also used as search terms. Task force members identified additional articles by using personal knowledge and examination of references in reviewed articles. Study Selection Primary criteria included the following: (1) provided reliability, diagnostic validity, and/or prognostic validity data; (2) examined a cohort, case control, or case series sample of persons with DOC who were age older than or equal to 18 years; and (3) assessed in an acute care or rehabilitation setting. Articles were excluded if peer review was not conducted, original data were not reported, or an English language article was not available. The initial search yielded 580 articles. After paired rater review of study abstracts, guideline development was based on 37 articles representing 13 DOC scales. Data Extraction Rater pairs classified studies addressing diagnostic and prognostic validity by using the American Academy of Neurology 4-tier level of evidence scheme, and reliability by using a task force–developed 3-tier evidence scheme. An independent quality review of ratings was conducted, and corrections were made. Data Synthesis The Coma Recovery Scale-Revised (CRS-R), Sensory Stimulation Assessment Measure (SSAM), Wessex Head Injury Matrix (WHIM), Western Neuro Sensory Stimulation Profile (WNSSP), Sensory Modality Assessment Technique (SMART), Disorders of Consciousness Scale (DOCS), and Coma/Near-Coma Scale (CNC) have acceptable standardized administration and scoring procedures. The CRS-R has excellent content validity and is the only scale to address all Aspen Workgroup criteria. The SMART, SSAM, WHIM, and WNSSP demonstrate good content validity, containing items that could distinguish persons who are in a vegetative state, are in a minimally conscious state (MCS), or have emerged from MCS. The Full Outline of UnResponsiveness Score (FOUR), WNSSP, CRS-R, Comprehensive Levels of Consciousness Scale (CLOCS), and Innsbruck Coma Scale (INNS) showed substantial evidence of internal consistency. The FOUR and the CRS-R showed substantial evidence of good interrater reliability. Evidence of diagnostic validity and prognostic validity in brain injury survivor samples had very high levels of potential bias because of methodologic issues such as lack of rater masking. Conclusions The CRS-R may be used to assess DOC with minor reservations, and the SMART, WNSSP, SSAM, WHIM, and DOCS may be used to assess DOC with moderate reservations. The CNC may be used to assess DOC with major reservations. The FOUR, INNS, Glasgow-Liege Coma Scale, Swedish Reaction Level Scale-1985, Loewenstein Communication Scale, and CLOCS are not recommended at this time for bedside behavioral assessment of DOC because of a lack of content validity, lack of standardization, and/or unproven reliability.