Clinical Content Tracking System – An Efficient Request Tracking system via a Graphical User Interface.

1. Clinical Content Tracking System – An Efficient Request Tracking system via a Graphical User Interface. Presented by: Saif S. Khairat Advisor: Dr. Chi-Ren Shyu Master’s Defense

2. Overview • Background • CPOE systems • CCTS • Database Management System • System Architecture • Conclusion, Future work and Limitations • System Demonstration

3. Background • Between 44,000 to 98,000 Americans die each year due to medical errors, and about 1 million people are injured. (Institute of Medicine, 1999) • Despite the desirability of implementing an efficient Computerized Physicians Order Entry (CPOE) system, only 9.6% of U.S. hospitals presently have CPOE completely available. (Ash, Gorman, 2002) • Paper-based systems, especially in the Health Informatics field, tend to be error prone and inefficient.

4. What is a CPOE system in healthcare? • A Computerized Physician Order Entry (CPOE) system is a process of electronic order entries for the treatment of patients under physicians care. • Benefits: • Decreases delay in order completion • Reduces errors related to handwriting • Allows order entry at off site locations

5. Current CPOE systems • The CPOE system has not yet proven its ability to fully eliminate medication errors. • Current CPOE systems do not use programmable decision-support infrastructures, and most health-care institutions do not have centralized and encoded clinical databases. (Guilherme, Roberto, 2005) • The cost of CPOE systems ranges from $3 million to $10 million, depending of hospital size and how well-built the IT infrastructure is. (Advisory Board Company, 2004)

6. What is CCTS and its role in the CPOE implementation process? • CCTS is a tool that helps physicians build clinical contents, in an electronic format, that will be then used in the CPOE system. • CCTS develops new electronic order sets by modify existing paper order sets through a governed structure. An order set is the grouping of patient orders for a specific diagnosis or condition. • The system delivers order sets that are less error prone to a CPOE system.

7. Why CCTS? • The main motivation behind this research is to develop an adaptive system that can be configured or customized to meet the needs of various health institute. • Our research emphasized the development a system that can be used by a large number of hospitals, regardless of their size or funds. • CCTS has a decision-support system built within to ensure that the workflow is compatible with the original workflow in the hospitals. • The system minimizes the use of free text by users to help eliminate errors.

8. System Information • Average turn around time depends on the size of the hospital. • Usually several months up to one year. • Hospitals currently using the system: • Winnipeg Regional Health Authority • Valley Baptist Medical Center • About 500-600 Clinical Contents (CC) to load in the system at developing stage. Then, 3-5 CC per month (in a 350 bed hospital).

9. What our CCTS does: • Stores images of current-state paper order sets (in .PDF, .JPEG or .DOC form) • Tracks various review-group decisions as these groups review all the existingorder sets for electronic conversion and implementation into the Computerized Physician Order Entry system • Allows requests for neworder sets to be made by members in the system, which then go through the review process, too. • Provides a versioning system for electronic order sets.

10. What our CCTS does: (Cont) • Creates electronic templates for existingorder set review/modification requests and for newrequests. Templates based on uploaded “paper” examples and review input. • Tracks the decision of each group involved in reviewing and approving (or failing) the electronic templates, which are then handed off to the Clinical Information System (CIS) implementation team.

11. Sample of a template:

12. Database Management System • CCTS is a database-driven system. • Our research focuses on developing an adaptive Database Management System (DBMS) that allows for the grooming, reviewing and tracking of decision for clinical contents. • The database design has been optimized and tuned several times in order to make the system as reliable and flexible as possible.

13. Old Entity-Relationship Diagram (1-2)

14. Old Entity-Relationship Diagram (2-2)

15. The ERD

16. Database Optimization and Tuning • Several iterations of design were done to refine our database schema from 60 tables in the original design to roughly one third of the number of tables in the final design. • The optimization process includes consolidations of contents from various tables and restructuring the design to a start-like architecture where Table Request plays as a heart of the entire database activities . • The new design provides version tracking of templates for a certain request using Table E_template which gives an extra dimension of advantages over the paper-based process.

17. Request People RequestID ActsOn PeopleID Name FirstName Description Actor attribute of relationship Database Optimization and Tuning (Cont.) • To deal with M-N relationship, we add an extra table in between so that each table will have a one-to-many (1-M) relationship.

18. People Request PeopleID RequestID FirstName Name Request-Acts Acts People-Acts ActorID PeopleID RequestID Associative entity type Database Optimization and Tuning (Cont.)

19. System Architecture The system has eight modules. Each module has its own members and functionalities.

20. Work flow

21. Groups/roles (of users) involved -- 1 of 4 CCTS has 8 typical groups/roles within hospitals that are involved in these processes. These are the main participant-users of this system: • Stakeholder – anyone within the hospital that is allowed to make requests for order set reviews, amendments and new order sets. Who is allowed to do this would depend on hospital policy. Likely this would include head physicians and nurse managers within each program or service area. • Content Board – a small group of senior decision-makers from various programs and service areas who oversee all clinical content reviews, revision and updates for their hospital. If the Content Board approves a request, they assign a Facilitator and a Content Specialist.

22. Groups/roles (of users) involved -- 2 of 4 • Facilitator – a higher level administrative assistant who assists the Content Board by creating templates • Content Specialist – every clinical program or service area would have at least one content specialist, if not more. These would be knowledgeable, experienced clinicians assigned by their program or service directors to participate in the ongoing content review and updating process. • Peer Specialist – every clinical program or service area would have a least one peer specialist, if not more. These would also be knowledgeable, experienced clinicians with whom the Content Specialist would confer on clinical content questions.

23. Groups/roles (of users) involved -- 3 of 4 • Clinical Program - this group comprises one person, likely the director or the director’s designate, from each clinical program or service area. This is the senior in charge for that area who makes the final review decision on Templates for input into the CPOE-CIS and the final decision on the Working Product – i.e. the computerized physician order entry order set – within the new CPOE prior to CPOE go-live date. • Clinical Content Design Team - this group comprises a small number of hospital IT people and informatics-savvy clinicians who will work with a consultant and the CIS implementation team to oversee the CIS implementation within their hospital.

24. Groups/roles (of users) involved -- 4 of 4 • System Analyst – this is the person from the hospital IT department who is the primary liaison between the CCDT and the CIS Team. This person could be a member of the CCDT or it could be the hospital IT manager responsible for the CIS implementation. • CIS Team – this is not an entity or user within our system. This is a separate entity consisting of members of an implementation team from the CIS vendor for the CIS that the hospital has purchased. The CIS Team will build the CPOE system based on the Working Product that the System Analyst delivers. • Administrator– this is a designated hospital IT person responsible for the operation and maintenance of this CPOE system.

25. Outline of the work flow within CCTS - 1 of 3 • Stakeholdermakes a request • Content Board meets regularly as group to review requests - approves a request, assigns Facilitator, assigns Content Specialist - or fails request • Content Specialistreviews the request - gathers and uploads scholarly articles to support her recommendations for clinical content changes - consults with Peer Specialist about recommendations - uploads .PDFs, JPEGs or .docs containing instructions on changes

26. Outline of the work flow within CCTS- 2 of 3 • Facilitatorcreates an electronic template from the materials the Content Specialist sent. • Clinical Program Director (the director or a designate) from the related specialty (e.g. pediatrics, radiology) reviews the template - approves template - or fails template • Clinical Content Design Team (CCDT)reviews the template - approves template - or fails template

27. Outline of the work flow within CCTS - 3 of 3 • Systems Analyst - works outside of this system with the CIS implementation team - creates a working product based on the approved template within the CPOE system

28. Screen shots from CCTS

29. Screen shots from CCTS

30. Screen shots from CCTS

31. Conclusion • The need for electronic order sets in health institutes is growing rapidly for many reasons. • This research proposed a system that will develop, review and track electronic order sets for CPOE systems to use. • CCTS has a built-in request tracking feature and a versioning tool. • It is essential to analyze our system against similar paper-based systems in order to evaluate our system.

32. Future work • It is our goal in the future to create a library of all order set names, modules and line items entered in to the system through users. • Create a Template via an Extensible Markup Language (XML) form with all the information in the library will be displayed to the user to choose from. • The system will allow the storage, review decision-tracking and grooming of clinical content other than order sets: e.g. medical logic modules and documentation templates.

33. Limitations • Currently the system does not support searching through the literature collected in the system. • Limitations of creating an XML library is that items inserted into the library might include errors.

34. Acknowledgments • Thank you to Dr. Chi-Ren Shyu for all his efforts and support during my course of study. • Thanks to Dr. Duan Ye and Dr. Guilherme DeSouza for their service as my thesis committee members . • Thanks to everyone who gave up some of their time to attend my defense.

35. System Demonstration Click here

36. Questions? Thank you!