Predictive Value of the Disorders of Consciousness Scale (DOCS)
Objective
To examine the predictive validity of measures of neurobehavioral change derived from the Disorders of Consciousness Scale (DOCS) for predicting return to consciousness 4, 8, and 12 months after severe brain injury (BI).
Design
Prospective observational predictive validity study
Setting
Inpatient rehabilitation hospitals and postrehabilitation residence
Participants
A total of 113 persons with a mean age of 38 ± 17.8 years who were unconscious for >28 days consecutively after severe BI; 73% (83/113) with traumatic BI and 27% (30/113) with other BI.
Independent Variables
Baseline DOCS, DOCS average, change from baseline DOCS to subsequent DOCS (DOCS2, DOCS3, DOCS4, DOCS5, DOCS6), and injury type (traumatic BI vs. other BI)
Main Outcome Measure
Time to consciousness at 4, 8, and 12 months after injury
Results
When controlling for injury type, the DOCS average as well as DOCS change between the first and second DOCS (DOCS1-2), first and fifth DOCS (DOCS1-5) and first and last DOCS (DOCStotalchg) significantly (P ≤ .05) contributed to predicting recovery and lack of recovery of consciousness at 4, 8, and/or 12 months after injury. DOCS1-5 manifested the most balanced accuracy in predictions, where predicting recovery of consciousness is accurate 87% of the time and predicting lack of recovery of consciousness is accurate 88% of the time.
Conclusion
For persons with prolonged disorders of consciousness, the findings indicate that evidence-based prognostication for individual patients is possible. The implications for research are that the DOCS can be used as a meaningful, reliable, and valid primary outcome to measure treatment effects in clinical trials. The evidence indicates further that DOCS measures merit inclusion in future research that aims to develop multivariate prognostication models.
To access this article, please choose from the options below
Disclosure Key can be found on the Table of Contents and at www.pmrjournal.org
This material is based upon work supported by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development Rehabilitation Research and Development Service and Health Services Research and Development Service through a HSR&D merit grant (Grant No. CCN-07-133); a RR&D advanced research career development award (Grant No. B4951-R); and a RR&D career development transition award (Grant No. B4949N). Funding also was provided by the Nick Kot Charity, nfp. In kind-contributions from Marianjoy Rehabilitation Hospital, the Minneapolis VA Medical Center, the Hines VA Hospital, and the Rehabilitation Institute of Chicago also support the work described in this article. The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.
PII: S1934-1482(08)00029-4
doi:10.1016/j.pmrj.2008.11.002
© 2009 American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved.
