Sepsis Treatment Management Tool

Edited on 2018-08-22

Sepsis Treatment Enhanced through Electronic Protocolization (STEEP)

Project Summary

Challenges of sepsis management

  • Sepsis is a common and lethal illness. Due to its nature, it is frequently managed in an intensive care unit (ICU) setting.
  • To achieve expected outcomes, sepsis management requires standardizing the care of patients using the readily available evidence-based guidelines.
  • Coordination of care teams during treatment requires precise documentation of the state of the patient and treatment. (Knowledge Transfer)
  • Changes to the evidence needs to be disseminated to both clinicians and clinical decision support (CDS) tools. (Knowledge Dissemination)

Goals: Support the overall clinical process by generating individualized care plans from evidence-based clinical protocols.

  • Provide subject matter experts (SMEs) with a modeling environment for capturing best practice (Surviving Sepsis Campaign (SSC) guidelines) in a formal manner.
  • Use the resulting protocol models to implement actionable and customizable clinical decision support tools that aid the clinical treatment processes by tracking patient state and performed clinical actions.
  • Enrich the solution with visual dashboards to show the status of the running guideline instances.

Results

Developed jointly by TRUST and VUMC research teams, a fully model-integrated clinical decision support and patient management system, called Sepsis Treatment Enhanced through Electronic Protocolization (STEEP), was implemented. The work was funded by NSF-TRUST, NIH and VUMC.

Modeling: STEEP represented the sepsis management protocol using a custom domain-specific language.

Model-driven UI

Protocol model-driven UI

Implementation: The implementation included builing prototype tools for the detection and management of sepsis. These tools integrated with the current EHR systems and were configured through the protocol models.

STEEP Architecture

STEEP Architecture

Evaluation: VUMC performed randomized trials for both detection (real-time alerting of modified SIRS criteria) and management of sepsis (using the sepsis protocol). However, due to various suspected causes neither solution influenced guideline compliance or clinical outcomes.

Contribution

  • Domain analysis, documentation
  • Solution architecture design
  • Domain model development
  • Domain modeling tool configuration and user education
  • Model translator creation for analysis, verification and visualization
  • Simulation and verification engine implementation
  • Contributed to UX (user experience) design

Publications

Semler, M.W.; Weavind, L.; Hooper, M.H.; Rice, T.W.; Gowda, S.S.; Nadas, A.; Song, Y.; Martin, J.B.; Bernard, G.R.; Wheeler, A.P. An Electronic Tool for the Evaluation and Treatment of Sepsis in the ICU: A Randomized Controlled Trial. Crit Care Med 43, 1595–1602 2015.

Mathe, J.L. The Precise Construction of Patient Protocols: Modeling, Simulation and Analysis of Computer Interpretable Guidelines. Vanderbilt University, Nashville, TN 2012.

Hooper, M.H.; Weavind, L.; Wheeler, A.P.; Martin, J.B.; Gowda, S.S.; Semler, M.W.; Hayes, R.M.; Albert, D.W.; Deane, N.B.; Nian, H.; Mathe, J.L.; Nadas, A.; Sztipanovits, J.; Miller, A.; Bernard, G.R.; Rice, T.W. Randomized Trial of Automated, Electronic Monitoring to Facilitate Early Detection of Sepsis in the Intensive Care Unit. Crit. Care Med. In press, 2096–2101 2012.

Miller, A. Alerts and Reminders: Is This All There is to Clinical Decision Support?. Presented at the 28th Human-Computer Interaction Lab Symposium (HCIL 2011), University of Maryland, Maryland, USA 2011.

Mathe, J.L.; Werner, J.; Sztipanovits, J. Model-Based Design of Trustworthy Health Information Systems, in: Homeland Security Facets: Threats. Countermeasures, and the Privacy Issue. Artech House, London, UK, pp. 119–134 2011.

Mathe, J.L.; Nadas, A.; Sztipanovits, J. A Model-Integrated, Guideline-Driven, Clinical Decision-Support System. Presented at the TRUST Review Meeting, Stanford, CA, USA 2010.

Hooper, M.H.; Martin, J.B.; Weavind, L.M.; Semler, M.W.; Albert, D.W.; Deane, N.B.; Paulett, J.M.; Mathe, J.L.; Miller, A.; Wheeler, A.P.; Bernard, G.R.; Rice, T.W. Automated Surveillance Of Modified SIRS Criteria Is An Effective Tool For Detection Of Sepsis In The Medical Intensive Care Unit, in: D53. SEPSIS: MECHANISMS AND IMPLICATIONS FOR MANAGEMENT, American Journal of Respiratory and Critical Care Medicine 2010. Presented at the American Thoracic Society 2010 International Conference, American Thoracic Society, New Orleans, pp. A6137–A6137 2010.

Mathe, J.L.; Martin, J.; Hooper, M. Towards an Adaptable Framework for Modeling, Verifying, and Executing Medical Guidelines. Presented at the TRUST Review Meeting, Washington, D.C. 2009a.

Mathe, J.L. Towards an Adaptable Framework for Modeling, Verifying, and Executing Medical Guidelines, in: Proceedings of the Doctoral Symposium at MODELS 2009. Presented at the ACM/IEEE 12th International Conference on Model Driven Engineering Languages and Systems (MODELS ’09), Denver, CO 2009.

Mathe, J.L.; Martin, J.; Miller, P.; Ledeczi, A.; Weavind, L.; Nadas, A.; Miller, A.; Maron, D.; Sztipanovits, J. A Model-Integrated, Guideline-Driven, Clinical Decision-Support System. IEEE Softw. 26, 54–61 2009b.

Salisbury, D.F. Vanderbilt doctors and software engineers pioneer an advanced sepsis detection and management system. VUCast Vanderbilt University’s News Network 2009.

Mathe, J.L.; Martin, J.B.; Miller, P.; Ledeczi, A.; Weavind, L.; Miller, A.; Maron, D.J.; Nadas, A.; Sztipanovits, J. STEEP – A Model-Integrated Clinical Information System Application. Presented at the TRUST Review Meeting, Berkeley, CA 2009c.

Mathe, J.L.; Martin, J.; Miller, P.; Ledeczi, A.; Weavind, L.; Miller, A.; Maron, D.; Nadas, A.; Sztipanovits, J. STEEP – A Model-Integrated Clinical Information System Application. Presented at the TRUST Review Meeting, Nashville, TN 2008a.

Mathe, J.L.; Martin, J.; Miller, P.; Weavind, L.; Maron, D.; Ledeczi, A.; Miller, A.; Nadas, A.; Sztipanovits, J. A Model-Integrated Approach to Implementing Individualized Patient Care Plans Based on Guideline-Driven Clinical Decision Support and Process Management. Presented at the TRUST Review Meeting, Nashville, TN 2008b.

Martin, J.; Mathe, J.L.; Miller, P.; Ledeczi, A.; Weavind, L.; Miller, A.; Maron, D.J.; Nadas, A.; Sztipanovits, J. A Model-Integrated Approach to Implementing Individualized Patient Care Plans Based on Guideline-Driven Clinical Decision Support and Process Management – A Progress Report, in: 2nd International Workshop on Model-Based Design of Trustworthy Health Information Systems (MOTHIS 2008). Presented at the MOTHIS 2008 Workshop @ MODELS, Toulouse, France 2008.

Back Home   

Contact Information

e | janos.l.mathe@vumc.org
p | +1 (615) 875-8778
a | Rm 515B, Ste 500, Bld 3401WE