|
BACKGROUND/RATIONALE:
In April 2007 the VA issued VHA Directive 2007-013 that established a policy for screening and evaluation of possible traumatic brain injury (TBI) in OEF and OIF veterans the TBI Clinical Reminder. Hoge and colleagues recently found that returning soldiers who suffered a concussion while deployed in Iraq have a
higher number of physical and postconcussive symptoms than soldiers with other injuries. However, after adjusting for PTSD and depression, concussion (mild TBI) was no longer associated with any of these symptoms with the exception of headache. In the discussion they point out that there has been no empirical
validation of the TBI Clinical Reminder questions, and based on their overall findings, they hypothesized that such population screening for TBI Deployment-related health problems represent a complex clinical phenomenon with unique and comorbid conditions often involving mTBI, PTSD, depression, pain, and other
conditions. These unique yet often overlapping conditions present challenges to health care providers in determining what condition or conditions account for the clinical presentation. Studying these factors in retrospective data is hampered by the fact that important clinical information is not routinely coded in the
electronic medical record (EMR). It is likely however that much of this nformation is available in the free text of the EMR. Automated techniques that search the text-based EMR clinical notes to extract information about OEF veterans could potentially make these data available for clinicians and researchers.
OBJECTIVE(S):
The objectives for this six-month pilot study are to: 1) Determine if text mining of clinical notes from the VistA EMR can reliably distinguish between those with PTSD and a positive TBI Clinical Reminder and those with PTSD and a negative TBI Clinical Reminder. 2) Determine whether any specific type(s) of text notes
(e.g. in-patient, outpatient, physician, nursing, psychology, neuropsychology, etc.) contribute more to the ability to classify PTSD patients with and without a reported history of mild TBI. 3) Conduct preliminary analyses to investigate whether the results of the text mining can identify clinical concepts, symptoms, signs, or findings that may improve the accuracy of the TBI Clinical Reminder.
METHODS:
A list of patients treated for mTBI in Tampa (n = 200) will be identified though the VistA EMR. In addition, text-notes for a comparison group, based on age and gender, of non-mTBI patients will be identified from the Tampa VA. All text-based clinical notes for the TBI sample (cases and controls) will be extracted, including both inpatient and outpatient notes for the first six months after the patients referral to the PCR. We will use SAS text mining software for our analysis. SAS Text Miner provides an environment to conduct file processing, text parsing, and transformation/dimension reduction and document analysis. Text mining approaches count occurrences of words in documents. A term-by-document frequency matrix is built and serves as the foundation for analysis of the document collection. Similar to factor analysis, the frequency matrix is decomposed of eigenvalues and eigenvectors that create linearly independent components of the data. The smaller components can be ignored and the similarity between two documents can be determined by the values of the remaining factors.
FINDINGS/RESULTS:
Enter text here.
IMPACT:
Automated techniques that could search the text-based EMR clinical notes to identify mTBI patients and extract information about their condition represent a potentially important resource to VA clinicians and HSR&D researchers. This tool could augment the process of identifying patients with this complex and subtle
condition. This study represent an initial step towards developing such a resource.
PUBLICATIONS:
None at this time.
DRA:
none
DRE:
none
Keywords:
none
MeSH Terms:
none
|
