TY  - JOUR
PY  - 2015//
TI  - Improving identification of fall-related injuries in ambulatory care using statistical text mining
JO  - American journal of public health
A1  - Luther, Stephen L.
A1  - McCart, James A.
A1  - Berndt, Donald J.
A1  - Hahm, Bridget
A1  - Finch, Dezon
A1  - Jarman, Jay
A1  - Foulis, Philip R.
A1  - Lapcevic, William A.
A1  - Campbell, Robert R.
A1  - Shorr, Ronald I.
A1  - Valencia, Keryl Motta
A1  - Powell-Cope, Gail
SP  - 1168
EP  - 1173
VL  - 105
IS  - 6
N2  - OBJECTIVES: We determined whether statistical text mining (STM) can identify fall-related injuries in electronic health record (EHR) documents and the impact on STM models of training on documents from a single or multiple facilities. <br><br>METHODS: We obtained fiscal year 2007 records for Veterans Health Administration (VHA) ambulatory care clinics in the southeastern United States and Puerto Rico, resulting in a total of 26 010 documents for 1652 veterans treated for fall-related injury and 1341 matched controls. We used the results of an STM model to predict fall-related injuries at the visit and patient levels and compared them with a reference standard based on chart review. <br><br>RESULTS: STM models based on training data from a single facility resulted in accuracy of 87.5% and 87.1%, F-measure of 87.0% and 90.9%, sensitivity of 92.1% and 94.1%, and specificity of 83.6% and 77.8% at the visit and patient levels, respectively. <br><br>RESULTS from training data from multiple facilities were almost identical. <br><br>CONCLUSIONS: STM has the potential to improve identification of fall-related injuries in the VHA, providing a model for wider application in the evolving national EHR system. (Am J Public Health. Published online ahead of print April 16, 2015: e1-e6. doi:10.2105/AJPH.2014.302440).<p />  <p>Language: en</p>
LA  - en
SN  - 0090-0036
UR  - http://dx.doi.org/10.2105/AJPH.2014.302440
ID  - ref1
ER  -