2 Dr. Shahram Yazdani© 2002 ATGCI
Medical Informatics
Medical informatics is the application of computers, communications and information technology and systems to all fields of medicine-medical care, medical education and medical research
3 Dr. Shahram Yazdani© 2002 ATGCI
Biomedical Information
Care Providers
Patients
DatabasesSearch enginesValue added Pub.EB Pub.POEM collections
Information masteryCritical appraisal
Data acquisition
Data storage
Vocabularies
Organization of data
Machine interfaces
Data retrieval
Data
Dx/T
x
Research
Information
Electronic medical records
Laboratory IS
Digital imaging and
Radiological IS
Patient monitoring systems
Medical information
directed to:
–a specific need
–at the right time
–in the right place
–to the right person
Decision-Support Systems
Diagnosis / Interpretation
Therapy / Management
Automated reminders and alert systems
Electronic prescription and order entry
Drug IS and Automated dispensing
Telemedicine
National health databank
National health code
National health card
Computer Assisted InstructionWeb Based Learning
Clinical JudgmentDiagnostic ReasoningTherapeutic Planning
4 Dr. Shahram Yazdani© 2002 ATGCI
Medical Informatics is Multidisciplinary
Medicine/ Biology Mathematics Information Systems Computer Science Statistics Decision Analysis Economics/Health Care Policy Psychology
5
Medical Records
6 Dr. Shahram Yazdani© 2002 ATGCI
Investment in medical records
In UK, the cost of medical records units run between 2% and 6% of NHS turnover
Healthcare providers spend 20-75% of their time reading, writing, sorting and searching through the notes
7 Dr. Shahram Yazdani© 2002 ATGCI
Shortcomings of the paper medical record
Find the record (lost, being used elsewhere) Find data within the record (poorly organized,
missing) Read data (legibility) Update data Record fragmentation Moving records Redundancy (re-enter data in multiple forms) Statistics and Research (can not search across
patients) Passive (no automated decision support)
8 Dr. Shahram Yazdani© 2002 ATGCI
Benefits of electronic medical record
Access, Availability, transfer and retrieval Legibility, Abstraction, reporting Saves time: find data 4 times faster Reduced data entry (reuse data) Better organization by imposing structure Storage space Allow multiple views including aggregation Automated checks on data entry (spelling
checks,k=50, sum of WBC, pregnant man) Data quality and standards Automated decision support Statistics and research
9
Images
10 Dr. Shahram Yazdani© 2002 ATGCI
Picture Archiving and Communications System
11 Dr. Shahram Yazdani© 2002 ATGCI
Image management Traditional Film libraries
– Expensive – Inefficient – Film can only be in one place– Problem in interpretation (20% on detection, 10-
50% on diagnosis)
Digital image library – Less expensive– More efficient– Tele-radiology– Automated interpretation
13
Medical Errors and
Point of Care Delivery of Information
14 Dr. Shahram Yazdani© 2002 ATGCI
Lots of lives to save Medical error is the 3rd largest cause of death in
the U.S. exceeded only by heart disease and cancer. (JAMA, July 26, 2000, p 483)
– 12,000 deaths/year from unnecessary surgery– 7,000 deaths/year from medication errors in hospitals– 20,000 deaths/year from other errors in hospitals– 80,000 deaths/year from nosocomial infections in
hospitals– 106,000 deaths/year from adverse effects of
medications (4th leading cause of death) 50-90% of medication errors can be eliminated
at time of installation of automation at the point of care.
15 Dr. Shahram Yazdani© 2002 ATGCI
Source: Wireless and Mobile Computing. First Consulting Group, Oct 2001.
16 Dr. Shahram Yazdani© 2002 ATGCI
Point of Care Automation Results
Malcolm, BEliminating Medication Errors Through Point of Care Devices. Proceedings of HIMSS 2000, Session 73, Dallas.
17 Dr. Shahram Yazdani© 2002 ATGCI
PDA Medical Record All essential medical data on a
Palm Pilot or PocketPC. Designed to integrate with
global medical data system.
18
Decision Support Systems
19 Dr. Shahram Yazdani© 2002 ATGCI
Levels of Automated Support
(Van Bemmel and Musen, 1997)
20 Dr. Shahram Yazdani© 2002 ATGCI
Computerized reminders
Computerized reminders can improve compliance with recommended guidelines– Increase preventive services– Increase use of appropriate medications – Increase use of other interventions
Hunt et al. JAMA. 1998; Shea et al. J Am Med Inform Assoc. 1996
21 Dr. Shahram Yazdani© 2002 ATGCI
Computerized decision support
Computerized decision support can improve quality – Prevention of venous thromboembolism– Use of antibiotics
Durieux et al. JAMA. 2000; Teich et al. Arch Intern Med. 2000;
Evans et al. N Engl J Med. 1998
22 Dr. Shahram Yazdani© 2002 ATGCI
EMR with Decision Support Electronic medical records with decision
support can reduce costs – Reducing medical errors and adverse events– Recommending equally effective but less costly
alternative interventions– Reducing the use of inappropriate tests– Reducing the ordering of redundant tests
Teich et al. Arch Intern Med. 2000; Bates et al. JAMA. 1998; Glaser et al. Proc Healthcare Information and Management Systems Society Annual Conf. 1996
23 Dr. Shahram Yazdani© 2002 ATGCI
de Dombal’s System Domain: Acute abdominal pain Input: Signs and symptoms of patient Output: Probability distribution of diagnoses Method: Bayesian Evaluation: an eight-center study involving 250 physicians
and 16,737 patients Results:
– Diagnostic accuracy rose from 46 to 65%– The negative laparotomy rate fell by almost half– Perforation rate among patients with appendicitis fell by half– Mortality rate fell by 22%
Results using survey data consistently better than the clinicians’ opinions and even the results using human probability estimates!
24
Electronic Prescription
25 Dr. Shahram Yazdani© 2002 ATGCI
Prescription warning messages generated by computerised system from October 1998 to August 1999
CategoryNo of warning
messagesNo (%) of prescriptions
completed
Disallowed 58 0 Contraindications 37 0 Interactions 21 0Password level warnings 749 322 (43) Contraindications 141 103 (73) Interactions 99 84 (85) Maximum recommended single dose
exceeded 206 89 (43)
Maximum recommended daily dose exceeded
303 46 (15)
Low level warnings 16 607 15 350 (92) Contraindications 793 677 (85) Interactions 15 743 14 635 (93) Maximum recommended single
dose exceeded 46 25 (54)
Maximum recommended daily dose exceeded
25 13 (52)
BMJ 2000;320:750–3
26 Dr. Shahram Yazdani© 2002 ATGCI
Computerized order entry Computerized order entry systems can
reduce medication errors – Improve drug prescribing– Improve drug dosing– Drug-drug interactions– Drug allergies
Bates et al. JAMA. 1998; Bates et al. J Am Med Inform Assoc. 1999; Teich et al. Arch Intern Med. 2000
27
Automated Dispensing
28
Card Technologies in
Health Care
29 Dr. Shahram Yazdani© 2002 ATGCI
Functions of health care cards
Identification Access control Data carrier (portable record) Information transfer Authentication Encryption/Decryption
30 Dr. Shahram Yazdani© 2002 ATGCI
Card technologiesCard technologies Unit cost($) Major benefits
Paper with barcode 0.01- 0.04 Inexpensive, bar code
Plastic, embossed 0.1- 0.15 Familiar, paper transfer
Serial memory card 1.5- 4.0 Additional storage
Computer chip card 3.5- 15.0 Additional data security, difficult to copy
Optical card 6.0- 8.0 Much more storage
IC optical card 10.0- 15.0 Large storage and data security
PC card 50.0- 100.0 More storage and computational capacity
31
Computer Literacyamong Physicians
32 Dr. Shahram Yazdani© 2002 ATGCI
Barriers are neither technology nor cost, Barriers are cultural: the doctors
Michael L.
Millenson
33 Dr. Shahram Yazdani© 2002 ATGCI
34 Dr. Shahram Yazdani© 2002 ATGCI
Do you personally use a computer?
Personal Computer Use % of the NOs who plan to do so in the next 12 months
Total YES NO No response
Overall 3128 78.6% 20.2% 1.2% 100.0% 37.2%
Females 938 73.0% 25.9% 1.1% 100.0% 42.0%
Males 2190 81.0% 17.7% 1.3% 100.0% 34.3%
< 35 320 76.3% 22.2% 1.6% 100.0% 43.7%
35-44 1033 84.2% 15.1% 0.7% 100.0% 44.2%
45-54 957 83.4% 15.3% 1.4% 100.0% 33.6%
55-64 585 72.1% 26.3% 1.5% 100.0% 39.0%
65+ 233 53.6% 44.6% 1.7% 100.0% 25.0%
GP/FP 1685 74.7% 24.0% 1.4% 100.0% 38.4%
Med Spec 1035 84.6% 14.3% 1.1% 100.0% 38.5%
Surg Spec 408 79.7% 19.4% 1.0% 100.0% 29.1%
Rural 302 73.8% 24.2% 2.0% 100.0% 38.4%
Urban 2826 79.1% 19.7% 1.1% 100.0% 37.1%
CMAJ - October 19, 1999
35 Dr. Shahram Yazdani© 2002 ATGCI
Barriers Lack of technology infrastructure Lack of standards Cultural barriers
– Eminence-based medicine– Tradition– Resistance to change
Complexity of medicine Workflow issues Human factor issues
36
Web Based Learning
37 Dr. Shahram Yazdani© 2002 ATGCI
Transformed view of knowledge
Dynamic, open ended, multidimensional,
and public
Static, finite, linear, and private.
38 Dr. Shahram Yazdani© 2002 ATGCI
Defining web-based teaching & networked learning
materials + human interaction
materials + human interaction
39 Dr. Shahram Yazdani© 2002 ATGCI
Online learning, Teaching, Tutoring
Classroom teaching
Tutor enhanced online learning
Computer-enhanced classroom teaching
Independent online learning
40 Dr. Shahram Yazdani© 2002 ATGCI
Online learning, Teaching, Tutoring
Classroom teaching
Computer-enhanced classroom teaching
Tutor enhanced online learning
Collaborative and independent
online learning
Distance learning
41 Dr. Shahram Yazdani© 2002 ATGCI
Advantages of CAI and WBL
Self directed learning Interactivity Privacy Individualized to student:
– Pace– Weak area
24 hr access Social interaction and Group work
42 Dr. Shahram Yazdani© 2002 ATGCI
Immediate Feedback Permanent record Economy of scale Promote active engagement Multimedia Access to global resources Information retrieval
Advantages of CAI and WBL
43 Dr. Shahram Yazdani© 2002 ATGCI
Uses of CAI in clinical medicine
Focus on diagnosis and therapy rather than on facts
Encourage experimentation and exploration Greater scope-see outpatient cases Prototypic cases that are not complex Include rare cases Better measure of competency than multiple
choice test
44 Dr. Shahram Yazdani© 2002 ATGCI
Can experiment without danger to patient
Follow course of patient over time Uses physiologic models, production
rules, …
Uses of CAI in clinical medicine
45
Virtual Reality
46 Dr. Shahram Yazdani© 2002 ATGCI
Using the MIST system (Mentice Medical Simulation AB, Gothenburg, Sweden) for training and assessment of psychomotor skills for minimally invasive surgery
BMJ VOLUME 323 20 OCTOBER 2001
47 Dr. Shahram Yazdani© 2002 ATGCI
The da Vinci Surgical System (Intuitive Surgical, California, USA) for performing minimally invasive surgery. The surgeon sits at a control console with 3D visualization of the surgical field and the robotic surgical instruments
BMJ VOLUME 323 20 OCTOBER 2001
Goals of Collaborative eHealth:
Saving Time, Saving Money,
Saving Lives
Turning the Promise of Mobile Computing into a Reality