Automated Problem List

The Automated Problem List system was developed and evaluated by the PI and Peter Haug in 2003-2005. The goal of this system was to automatically enhance the completeness and timeliness of the medical electronic problem list in the locally developed EHR at Intermountain Healthcare. This system used NLP to automatically extract a selection of potential medical problems from all narrative text notes in the EHR, and stored the extracted medical problems back into the EHR (in a structured and coded format), along with an HL7 CDA (Clinical Document Architecture) version of the analyzed notes. These structured and coded problems could then be managed in the electronic problem list. The complete system was implemented and evaluated in an Intensive Care unit and a Cardiovascular Surgery unit at the LDS Hospital (Salt Lake City, Utah). The completeness of the problem list was significantly improved, growing from 6% to 41% in the Intensive Care Unit, and even to 77.4% if all automatically added problems were included. This study was the first published application of NLP to analyze clinical text in a real-time clinical setting, and gave us significant experience about the performance and possible uses of NLP for information extraction from clinical text.

ScreenShot

Publications:

  • Meystre, S. M., & Haug, P. J. (2008). Randomized controlled trial of an automated problem list with improved sensitivity. International Journal of Medical Informatics, 77(9), 602–612.
  • Meystre, S., & Haug, P. J. (2006). Natural language processing to extract medical problems from electronic clinical documents: performance evaluation. Journal of Biomedical Informatics, 39(6), 589–599.
  • Meystre, S. M., & Haug, P. J. (2006). Using Natural Language Processing to Convert Documents to the HL7 Clinical Document Architecture, Release Two, with Embedded Coded Medical Problems (pp. 1–5). SSMI Annual Symposium.
  • Meystre, S., & Haug, P. (2006). Improving the sensitivity of the problem list in an intensive care unit by using natural language processing. AMIA Annu Symp Proc, 554–558.
  • Meystre, S., & Haug, P. J. (2005). Evaluation of Medical Problem Extraction from Electronic Clinical Documents Using MetaMap Transfer (MMTx). Studies in Health Technology and Informatics, 116, 823–828.
  • Meystre, S. M., & Haug, P. J. (2005). Comparing natural language processing tools to extract medical problems from narrative text. AMIA Annu Symp Proc, 525–529.
  • Meystre, S., & Haug, P. J. (2005). Automation of a problem list using natural language processing. BMC Medical Informatics and Decision Making, 5, 30.