1 / 84

Modeling in Healthcare: Requirements and Clinical Processes

Explore the challenges in healthcare, requirements engineering, and the role of modeling in optimizing clinical processes for improved outcomes. Discover real-time decision support, interdisciplinary team dynamics, and research opportunities in this dynamic field.

bohannan
Download Presentation

Modeling in Healthcare: Requirements and Clinical Processes

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Daniel Amyotdamyot@uottawa.ca EBC8101March 21, 2019 Modeling in Healthcare: Requirements and Clinical Processes

  2. Overview • Some important healthcare challenges • Domain requirements • Requirements Engineering in healthcare • Where can modeling help in healthcare? • Clinical processes • Real-time decision support • Interdisciplinary healthcare teams • Architectures • Predictive analytics • Privacy compliance • Research opportunities Modeling in Healthcare - EBC8101 p.2

  3. Partners in Collaborative Projects Telfer Health Transformation Exchange (THTEX) Modeling in Healthcare - EBC8101 p.3

  4. TOH Projects Adverse Event Management System BPM, DW, BI, and privacy compliance Goal-driven methodology for developing healthcare quality metrics Patient-device association and disassociation with a Real-Time Location System Real-time, location-based hand hygiene monitoring and notification system Real-time pathology dashboard Care Ami: a mobile patient navigator for lung cancer assessment A Quality Assurance Framework for Care Process Management Valuing extraordinary patient experience and safe care for multiple pregnancy Modeling in Healthcare - EBC8101 p.4

  5. Projects with Other Hospitals • Location-Aware Business Process Management for Real-time Monitoring of Patient Care (Osler) • Real-Time Simulation of Patient Care Processes in Healthcare (Osler) • BI-Driven Management of Patient Flows (Osler) • Privacy-preserving (K-anonymity) algorithm for healthcare information sharing (Privacy Analytics, CHEO) • Privacy-Preserving Patient Tracking for Phase 1 Clinical Trials (Privacy Analytics, CHEO) • Supporting Interdisciplinary Healthcare Team Dynamics with Business Process Management (Bruyère) • Real-Time Demand Capacity Tool (QCH, HealthNow) • A Mobile Self-Triage Application (Montfort) • Activity-based Process Integration Framework to Improve User Satisfaction and Decision Support in Healthcare (Montfort) • User-Centered Change Management Process for CPOE Implementation (Montfort) • Process Mining in Healthcare Processes (Montfort) • Evidence-based Newborn Screening Process Improvement (NSO) Modeling in Healthcare - EBC8101 p.5

  6. Background Concepts • A requirement is: • A statement about the proposed system that all stakeholders agree must be made true in order for the customer’s problem to be adequately solved. • Requirements Engineering (RE) is: • The activity of development, elicitation, specification, analysis, validation, and management of the stakeholder requirements, which are to be met by a new or evolving system. • A model is: • A reduced representation (simplified, abstract) of (one aspect of) a system used to help understand complex problems/solutions, communicate information about the problem/solution, and direct implementation (especially in software). Modeling in Healthcare - EBC8101 p.6

  7. Some important healthcare challenges Modeling in Healthcare - EBC8101 p.7

  8. Healthcare in Canada, 2018 https://www.cihi.ca/sites/default/files/document/nhex-trends-narrative-report-2018-en-web.pdf Modeling in Healthcare - EBC8101 p.8

  9. Healthcare Spending Trends in Canada https://www.cihi.ca/sites/default/files/document/nhex-trends-narrative-report-2018-en-web.pdf Modeling in Healthcare - EBC8101 p.9

  10. How Does Canada Compare? https://www.cihi.ca/sites/default/files/document/nhex-trends-narrative-report-2018-en-web.pdf Modeling in Healthcare - EBC8101 p.10

  11. Wait Times in Ottawa, Ontario http://www.ontariowaittimes.com/ Most large community and teaching hospitals operate near congestion levels, with long wait times Emergency rooms near the University of Ottawa (“Total Time” columns indicate that 1 patient our of 10 will stay longer than this) Modeling in Healthcare - EBC8101 p.11

  12. https://www.hqontario.ca/System-Performance/Time-Spent-in-Emergency-Departmentshttps://www.hqontario.ca/System-Performance/Time-Spent-in-Emergency-Departments Modeling in Healthcare - EBC8101 p.12

  13. Where Does Outaouais Stands? [leDroit.com, June 2, 2016] [http://www.cbc.ca/news/canada/ottawa/gatineau-hospital-coroners-report-1.3736927] Modeling in Healthcare - EBC8101 p.13

  14. http://ici.radio-canada.ca/nouvelle/1064909/temps-attente-urgence-civieres-outaouaishttp://ici.radio-canada.ca/nouvelle/1064909/temps-attente-urgence-civieres-outaouais Modeling in Healthcare - EBC8101 p.14

  15. Wait Times in Outaouais, Québec • Some stats on ER patients waiting on stretchers on March 17, 2019 (https://www.indexsante.ca/urgences/): Modeling in Healthcare - EBC8101 p.15

  16. Resource Management • Too often, resources are waisted • For example, in 2017, Gatineau used its IRM equipment only 56% of the recommended 16h/day • Partly explains long wait times • Too many attempts to optimize locally • Only move bottlenecks to other steps of the process • Need for global optimizations and real-time monitoring • Domain requirement • Healthcare systems shall consider the impact of designs on the end-to-end journey of patients Modeling in Healthcare - EBC8101 p.16

  17. Physicians and Nurses are Busy • Survey of 20,000 U.S. Physicians Shows 80% of Doctors are Over-Extended or at Full Capacity • The Physicians Foundation, September 2014 • Physicians and nurses have to make way too many important decisions every day! • Decision fatigue (e.g., with antibiotic prescriptions) • Domain requirements • Healthcare systems shall not add to the workload of health practitioners • Healthcare systems shall help reduce the number of daily decisions made by physicians Modeling in Healthcare - EBC8101 p.17

  18. Making Sense of the Data • Healthcare generates HUGE quantities of data • Extracting useful knowledge from that data is still difficult • Sharing and aggregating data is difficult • Privacy legislation gets in the way… • Data often becomes available too late to help • Domain requirement • Healthcare systems shall help providing the right information to the right people in the right format at the right time Modeling in Healthcare - EBC8101 p.18

  19. Clinical/Healthcare Processes:Patients are not Products • Healthcare processes and clinical pathways deal with human beings, not industrial products • Many patients with a similar health problem need to be treated slightly differently • Somewhere between conventional Business Process Management and Case Management • Clinical pathways for many types of morbidities are well defined, but many patients have multiple morbidities • Domain requirement • Healthcare processes shall adapt to the unicity of patients (patient centricity) in a way that enables treating multiple morbidities Modeling in Healthcare - EBC8101 p.19

  20. Funding Model and Legislation • Private vs public funding • In Canada, physicians are not hospital employees • Payment based on acts or on value • Traditionally: funding based on the volume of care delivered to patients. • Trend: funding based on the outcomes of the care and its end-to-end value for the patient • A procedure in Ontario is now reimbursed according to the provincial average cost • Competition! Modeling in Healthcare - EBC8101 p.20

  21. Funding Model and Legislation • HITEC and Meaningful Use in the USA • Force the harmonization and evolution of systems • In Canada, little harmonization; hospitals use different systems, augmenting the need and effort for integration (although this is changing) • Laws and regulations in constant evolution • Healthcare undergoing major transformations! • Domain requirement • Healthcare systems shall be designed according to the relevant legislation and funding model Modeling in Healthcare - EBC8101 p.21

  22. Increasing U.S. Healthcare Regulation CCHIT MU Stage 1 HIPAA HITECH (ARRA) MU Stage 2 Meta-Standards Rule 2007 2008 1996 2009 2010 2011 2012 HIPAA Omnibus Regs NwHIN Governance DEA EPCS Data Breach Notification Various State Laws … and each of these laws evolves too (multiple versions!) Requirements Management J.C. Maxwell, A.I. Antón, P. Swire: Managing Changing Compliance Requirements by Predicting Regulatory Evolution. Intl. Conf. on Requirements Engineering, IEEE CS, 2012.

  23. Almoaber, B., Amyot, D.: Barriers to Successful Health Information Exchange Systems in Canada and the USA – A Systematic Review. Int. J. of Healthcare Information Systems and Informatics, 12(1), 2017 Modeling in Healthcare - EBC8101 p.23

  24. Requirements Engineering in healthcare Modeling in Healthcare - EBC8101 p.24

  25. RE for Digital Health (2015) Modeling in Healthcare - EBC8101 p.25

  26. What is Actually Used (419 Projects, 8% Healthcare)? Modeling in Healthcare - EBC8101 p.26 S. Fricker, R. Grau, A. Zwingli (2014). Requirements Engineering: Best Practice. Requirements Engineering for Digital Health. Springer.

  27. RE in Healthcare: The Literature ( requirements W/1 ( engineering OR elicitation OR gathering OR specification OR validation OR modeling OR modelling OR notation OR analysis ) AND ( medicine OR "health care" OR "healthcare" OR medical ) ) In 2017, Scopus returned 1345 publications Half coming from journals and half from conferences 75 (5.6%) involved Canadian authors Without the healthcare-oriented restriction, Scopus returned 34,185 documents Although healthcare is an essential and expensive sector (10.4% of GDP in Canada), only 4% of the RE literature addresses healthcare, with no observable change in trends Modeling in Healthcare - EBC8101 p.27

  28. RE in Healthcare: The Literature (2016) [M. Baslyman, thesis proposal, 2016] Modeling in Healthcare - EBC8101 p.28

  29. Informal Observations • One hospital with ~250 IT staff • Informal requirements, use cases, user stories • Business analysts between stakeholders and development team • Development and integration • One hospital with ~25 IT staff • Direct access to stakeholders (still infrequent) • Integration mainly • Requirements in Word on Sharepoint Modeling in Healthcare - EBC8101 p.29

  30. What About Technologies? • Software/systems modeling technologies • UML/SysML, Logic, Goals, URN, Simulations, Feature Models, BPMN, EvSEC, Simulation, ontologies, Aspects, … • Healthcare technologies • Health Information Systems, Electronic Health Records, Decision Support Systems, Imagery, … • Health Level Seven Int. (HL7) for system interoperability • ICD 10 and SNOWMED for clinical classification • Data modeling • Business Process Management (BPMN and others)… • Clinical Document Architecture (CDA), Cross-Enterprise Document Sharing (XDS) • Simulation, BI, (predictive) analytics, … • Domain requirement • Models shall enable interoperation with existing healthcare technologies Modeling in Healthcare - EBC8101 p.30

  31. WHERE CAN Modeling Help in Healthcare? Modeling in Healthcare - EBC8101 p.31

  32. Clinical Processes Business Process Model and Notation (BPMN) and similar models, together with management and execution tools, already used to support clinical pathways Modeling in Healthcare - EBC8101 p.33

  33. Cardiology Pathway (BPMN) Modeling in Healthcare - EBC8101 p.34

  34. IBM BPM: Service Implementation

  35. IBM BPM: “Coaches” (Forms)

  36. Nurse Test Order (Web) Modeling in Healthcare - EBC8101 p.37

  37. iPad Version Modeling in Healthcare - EBC8101 p.38

  38. Modeling in Healthcare - EBC8101 p.39

  39. Not So Fast... There are Modeling Challenges! • Major integration effort with health information systems (typically 100-150 different ones in a hospital) • Management of deviations and exceptions still a major problem • Getting realistic processes is also difficult • Many are investigating process miningin that context • Mans, R., van der Aalst, W.M.P., and Vanwersch, R. (2015) Process Mining in Healthcare: Evaluating and Exploiting Operational Healthcare Processes. SpringerBriefs in Business Process Management, Springer International Publishing, 2015. Modeling in Healthcare - EBC8101 p.41

  40. [M. Ghasemi, 2015] [M. Ghasemi, 2015]

  41. Process Mining in a Nutshell Modeling in Healthcare - EBC8101 p.43

  42. Basic Types of Process Mining van der Aalst, W.M.P. (2011) Process mining: discovery, conformance and enhancement of business processes. Springer Science & Business Media, 2011. Modeling in Healthcare - EBC8101 p.44

  43. Number of Process-Mining-Related Publications, Per Year (May 2016) Ghasemi, M., Amyot, D.: Process Mining in Healthcare — A Systematised Literature Review. Int. J. of Electronic Healthcare, 9(1), 20–35, 2016 Modeling in Healthcare - EBC8101 p.45

  44. Number of Healthcare Process-Mining-Related Publications, Per Year Modeling in Healthcare - EBC8101 p.46

  45. Challenges of Process Mining in Healthcare • Think of Process Mining as the intersection between data mining, AI, and modelling. • Data is one obvious issue! • Identifying, accessing, and integrating data sources (especially patient-related) • Data quality, granularity, and completeness (logs) • Visualization of process models, especially for complex and less-structured processes • Automated tools, but still a great amount of pre-processing and reliance on experts for applying process mining Modeling in Healthcare - EBC8101 p.47

  46. Ease his pain… Modeling in Healthcare - EBC8101 p.48

  47. Need for Real Time and Adaptation • Having valid process models that are executable is still not enough to ease the pains of physicians, nurses, and patients • Need to align resources with demand in real time • Simply adding more resources is unsustainable • Need to consider real-time location information • Some healthcare process instances are long lived • Days, months, years… • Need to adapt dynamically the composition of the care provider team Modeling in Healthcare - EBC8101 p.49

  48. Capacity Management at 15:00 Modeling in Healthcare - EBC8101 p.50

More Related