1. Medical informatics Lecture 1
Introduction to Medical Informatics Definition and scope of HI, medical research to clinical practice lifecycle, electronic patient records

2. The big picture Understanding diseases and their treatment Develop
Standards based
formalisation of clinical data and research results
Understanding
diseases and
their treatment
Develop
and test treatments
Patient-specific
Decision-making to optimise and personalise treatment
Clinical engagement, post-marketing surveillance, data mining
Service delivery,
performance assessment
Ensure right
Patients receive
right intervention
Manage safe workflow, professional communication, security

3. Course objectives
Provide an overview of the main development areas in health informatics.
Understand the role of informatics in translating medical research into clinical practice
Look at 4 key topics in more depth
Electronic patient records
Formal representation of clinical data and medical knowledge
Clinical decision making and decision support
Care pathways and workflow management

4. Recommended texts
Guide to Health Informatics - Enrico Coiera 2nd edition 2003
From Patient data to Medical Knowledge - Paul Taylor 2006
Other useful resources at

5. Biomedical informatics (1): Bio-informatics
Rapidly developing branch of biology: highly interdisciplinary, using techniques and concepts from IT, statistics, mathematics, chemistry, biochemistry, physics, and linguistics!
Seeks knowledge from computer analysis of
biological data (e.g. genomics, proteomics)
experimental results
patient statistics
scientific literature.
Research in bioinformatics includes development of methods for storage, retrieval, and analysis of data, modeling and simulation of cellular/molecular systems.

6. Biomedical informatics (2): Health-informatics
Also known as medical or clinical informatics
It is applied to primary and specialist patient care, nursing, dentistry, pharmacy, public health etc.
Deals with the resources, devices, and methods required to optimize the acquisition, storage, retrieval, and use of information in delivery of healthcare services
A particular focus is on services at the point of care and emphasis is increasingly being placed on informatics for patients and carers as well as professionals.

7. Topics in health informatics (1): traditional perspective
Architectures for electronic medical records and other health information systems used for billing, scheduling, and research
Standards (e.g. DICOM, HL7) … to facilitate the exchange of information between healthcare information systems - these specifically define the means to exchange data, not the content
Controlled vocabularies … used to allow a standard, accurate exchange of data content between systems and providers
Software for specialist services and devices

8. Topics in health informatics (2): new drivers
Quality and safety
US Institute of Medicine
“To err is human”
“Crossing the quality chasm”
McGlynn data on service delivery
Fineberg lecture on YouTube
NHS
Emergence of clinical decision support and workflow management systems

9. Topics in health informatics (3): Contemporary multidisciplinary view
Traditional “engineering” topics
Hardware and software service architectures
Specialist technical services
Digital signal processing
Human and organisational factors in quality and safety
User interface design (Tang lecture on YouTube)
Organisational memory
Learning from experience
Change management
Formal representation of data and knowledge
Controlled vocabularies, “ontologies”
Applying knowledge to data: logic and description logics, decision theory, guidelines and workflows

10. The key challenges (adapted from Coiera p 104)
How do we apply knowledge to achieve a particular clinical objective?
How do we decide how to achieve a particular clinical objective?
How do we improve our ability to deliver clinical services?

11. Medical research, clinical practice
Understanding
diseases and
their treatment
Develop
and test treatments
Health
Records
Service delivery,
performance assessment
Ensure right
Patients receive
right intervention

12. First … Health Records Capture your data, accurately, completely
Make the data readily accessible
Health
Records

13. The paper record, pros Portable Familiar and easy to use
Exploits everyday skills of visual search, browsing etc
Natural: “direct” access to clinical data
Handwriting
Charts, graphs
Drawings, images…

14. The paper record: cons Can only be used for one task at a time
If 2 people need notes one must wait
Can lead to long waits (unavailable up to 30% of time in some studies)
Records can get lost
Consume space
Large individual records are hard to use
Fragile and susceptible to damage
Environmental cost

15. Electronic health records
An electronic health record is a repository of information about a single person in a medical setting, including clinical, demographic and other data.
The repository resides in a system specifically designed to support users by
providing accessibility to complete and accurate data
may include services to provide alerts, reminders, links to medical knowledge and other aids to clinical practice.

18. The electronic medical record

19. Examples

20. Functions of the EHR (1)
Supports structured data collection using a defined vocabulary.
Accessible at any or all times by authorized individuals.
Contains a problem list - patient’s clinical problems and current status
Supports systematic measurement and recording of data to promote precise and routine assessment of the outcomes of patient care
States the logical basis for all diagnoses or conclusions as a means of documenting the clinical rationale for decisions about the management of the patient’s care.

21. Functions of the EHR (2)
Can be linked with other clinical records of a patient—from various settings and time periods—to provide a longitudinal (i.e. lifelong) record of events that may have influenced a person’s health.
Can assist the process of clinical problem solving by providing clinicians with decision analysis tools, clinical reminders, prognostic risk assessment and other clinical aids.
Can be linked to both local and remote databases of knowledge, literature and bibliography or administrative databases and systems so that such information is readily available to assist practitioners in decision making.
Addresses patient data confidentiality.
Can help practitioners and health care institutions manage the quality and costs of care.

22. Electronic health records: pros
Compact
Simultaneous use
Easily copied/archived
Portable (handheld and wireless devices)
Secure
Supports many other services
Decision support
Workflow management
Performance audits
Research

23. Electronic health records: cons
High capital investment
Hardware, software, operational costs
Transition from paper to computer
Training requirements
Power outs – the whole system goes down!
Continuing security debate
Stealing one paper record is easy, 20 is harder, 10,000 effectively impossible – the security risks are very different for electronic data

24. Views of record systems

25. Ad hoc view

26. User view

27. Service architecture view
Chronic
care
services
Acute
services
Primary
Care
services
Communication &
Coordination
services
Point of
care
services
Search and
analysis
services
Clinical data
“Organisational
Memory”
Federated
EHR
Terminologies
Ontologies
Clinical
guideline
repository
Clinical
trials
repository

28. Functional view

29. Medical record structures (1)
Integrated record
Data are recorded and presented chronologically around episodes of care, following the sequence of events, encounters and actions associated with the patient’s medical needs.
Source oriented record
organized around the organization of the healthcare service, with separate sections for medical notes, nursing notes, laboratory data, radiological results etc. No concept of a clinical task or process in this form of data recording.

30. Medical record structures (2)
Protocol-oriented record
Often used when a patient is being treated according to a standard treatment plan or pathway. Highly task- oriented, providing useful guidance for what needs to be done at any point in treatment, but providing little overview of the patient’s needs.
Problem-oriented record
Organised around a list of the patient’s medical problems, which is used to index the whole record, and an integrated treatment plan. The plan describes what is to be done for each problem, with all associated progress notes, lab tests, medications etc linked to the initiating problem.

31. Point of care services Clinical documentation Patient administration
Search services
Decision support
Workflow management
Communication and coordination

32. “Grand Challenge” The NHS of the future?