Summarization of clinical information: A conceptual model

  • Authors:

  • Joshua C. Feblowitz;Adam Wright;Hardeep Singh;Lipika Samal;Dean F. Sittig

  • Affiliations:

  • Brigham & Women's Hospital, Boston, MA, USA and Partners HealthCare, Boston, MA, USA;Brigham & Women's Hospital, Boston, MA, USA and Partners HealthCare, Boston, MA, USA and Harvard Medical School, Boston, MA, USA;Houston VA Health Services Research and Development Center of Excellence, Houston, TX, USA and Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX, USA and Baylor College of Medicine, ...;Brigham & Women's Hospital, Boston, MA, USA and Partners HealthCare, Boston, MA, USA and Harvard Medical School, Boston, MA, USA;Memorial Hermann Health System, Houston, TX, USA and University of Texas Health Science Center, Houston, TX, USA and National Center for Cognitive Informatics and Decision Making, Houston, TX, USA

  • Venue:
  • Journal of Biomedical Informatics
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Background: To provide high-quality and safe care, clinicians must be able to optimally collect, distill, and interpret patient information. Despite advances in text summarization, only limited research exists on clinical summarization, the complex and heterogeneous process of gathering, organizing and presenting patient data in various forms. Objective: To develop a conceptual model for describing and understanding clinical summarization in both computer-independent and computer-supported clinical tasks. Design: Based on extensive literature review and clinical input, we developed a conceptual model of clinical summarization to lay the foundation for future research on clinician workflow and automated summarization using electronic health records (EHRs). Results: Our model identifies five distinct stages of clinical summarization: (1) Aggregation, (2) Organization, (3) Reduction and/or Transformation, (4) Interpretation and (5) Synthesis (AORTIS). The AORTIS model describes the creation of complex, task-specific clinical summaries and provides a framework for clinical workflow analysis and directed research on test results review, clinical documentation and medical decision-making. We describe a hypothetical case study to illustrate the application of this model in the primary care setting. Conclusion: Both practicing physicians and clinical informaticians need a structured method of developing, studying and evaluating clinical summaries in support of a wide range of clinical tasks. Our proposed model of clinical summarization provides a potential pathway to advance knowledge in this area and highlights directions for further research.