Transcript lecture09_eHealth

CS5038 The Electronic Society
Lecture 9: eHealth
Lecture Outline
•
•
•
•
•
•
•
•
•
•
•
•
•
•
The Future of Healthcare - The banking metaphor
Existing Health on the Web
eHealth - terminology
Transmural Care
Electronic Medical Records (EMR)
Medical Records - Access
Clinical Decision Support Systems
Telemedicine - Case Studies
eHealth Standards
eHealth / eScience : Cancer Diagnosis
Benefits of eHealth
Medical Errors
Why is eHealth Adopted Slowly?
New sources of "health"
1(#total)
eHealth - The Future of Healthcare
The banking metaphor
Most transactions carried out by
the customer
Centralisation of specialist
services
Decentralisation of nonspecialist services
2(#total)
Existing Health on the Web
Estimated to be ~20,000 health websites
Used by 98 million adults
 75% of people who have web access
 average of 3.3 times per month
More than consult doctors each day
 7M e-patients/day on the net; 2-3M patients see a doctor
1
2
3
3(#total)
Existing Health on the Web
Access to accurate information can lead to
more knowledgable, empowered, less anxious
patients
more participatory health decisions
better care as patient and doctor become partners
Mis-information can lead to
confused and angry patients
bad decisions, mis-placed hope, worse care,
harm
Privacy violations can cause emotional and economic
damage
4(#total)
eHealth
“Healthcare which is supported by electronic
processes”
Other terms:
Healthcare informatics or Health Information
Technology (HIT)
Medical Information Systems (MIS)
Biomedical informatics (also includes Bioinformatics:
gene sequencing etc.)
5(#total)
eHealth
“Healthcare which is supported by electronic processes”
eHealth includes:
 Electronic Medical Records: easy communication of patient data between different
healthcare professionals (GPs, specialists, care team, pharmacy)
 Telemedicine: do not require a patient and specialist in same physical location.
 Decision support systems in healthcare




 Data can be analysed to provide alerts, reminders and real-time decision aids
Evidence Based Medicine:
 The application of the scientific method to medical practice
 Check if diagnosis is in line with scientific research.
 Data can be kept up-to-date.
Citizen-oriented Information Provision: for both healthy individuals and patients
Specialist-oriented Information Provision: best practice guidelines from latest
medical journals.
Virtual healthcare teams: collaborate and share information on patients through
digital equipment (for transmural care).
6(#total)
Transmural Care
Transmural: Care should not stop at the walls of the hospital
 Both intra- and extra-mural, thus ‘transmural care’.
 Care before, during and after the hospital stay.
 Cooperation and coordination among local practitioner,
hospital, home care and rehabilitation centres
 Patient part of an agreed programme - protocols and
standards.
7(#total)
Electronic Medical Records (EMR)
(also called Electronic Health Record (EHR))
 Access of patient data by clinical staff at any given location
 Accurate and complete claims processing by insurance
companies
 Building automated checks for drug and allergy interactions
 Clinical notes
 Prescriptions
 Scheduling
 Sending and viewing labs
8(#total)
Electronic Medical Records (EMR)
(also called Electronic Health Record (EHR))
NHS project in UK: One of the largest projects in the world for a
countrywide EMR
 Goal: 60,000,000 patients with centralised EMR by 2010.
Two types of record:
 “Born digital" record : information originally entered in
electronic format
 “Digital format” record : originally produced in a hardcopy form
(x-ray film, photographs, etc.), scanned or imaged and
converted to a digital form.
Also: Personal Health Record (PHR) - stored and maintained by
the patient.
 Issue: Home computer vulnerable to attack
9(#total)
Electronic Medical Records (EMR)
Maintaining Records
 May be required many years after a patient’s death
 Insurance claims or murder investigation
 Investigate illnesses within a community
• industrial or environmental disease
• doctors committing murders
 need for periodic conversion and migration to ensure the
formats they were captured in remain accessible
 Media degrades
 Media becomes obsolete
 protection of privacy is a major concern - need privacy and
security policies
10(#total)
Electronic Medical Records (EMR)
Abuses
 Outsourcing of transcription and storage: Could violate
patient-physician confidentiality
 Governments may refuse to disclose records of military
personnel used as experimental subjects.
 http://www.guardian.co.uk/military/story/0,11816,1331784,0
0.html
11(#total)
Medical Records - Access
Laws governing access to medical records:
In the UK:
 Data Protection Acts & Freedom of Information Act 2000
 Patients or their representatives have the right to a copy of
their record. Except:
• where information breaches confidentiality
(e.g. information from another family member or where a patient
has asked for information not to be disclosed to third parties)
• where information would be harmful to patient's well-being
(e.g. some psychiatric assessments).
 Patients have the right to check for any errors in their
record and insist that amendments be made if required.
12(#total)
Medical Records - Access
Laws governing access to medical records:
In the US:
 Only patient and health care providers directly involved in delivering care
have the right to view the record.
 Patient may grant consent for any person to evaluate the record.
 Special Situations.
 Capacity
• When a patient does not have capacity, a legal guardian is designated
 Medical emergency involving a non-communicative patient
• Consent is assumed unless written documentation has been drafted
previously
 Research, auditing, and evaluation
• Individuals involved in research have access. They are not allowed access
to identifying information.
 Risk of death or harm
• Information can be shared without permission if failure to do so would
result in death or harm.
13(#total)
Clinical Decision Support Systems
Software to aid clinical decision-making; characteristics of patient
are matched to knowledge base, recommendations are
presented to the clinician/patient
(Sim et al, JAMIA, 2001)
Objectives:
 Diagnostic support
 Drug dosing
 Preventive care reminders
 Disease management (diabetes, hypertension, AIDS, asthma)
 Test ordering, drug prescription
14(#total)
Clinical Decision Support Systems
Methods:
rule-based, bayesian network, neural network, fuzzy
logic, genetic algorithms, case-based reasoning, etc.
Forward reasoning (data-driven) use if sparse data
 start with data, execute applicable rules, see if new
conclusions trigger other rules:
 if high WBC AND cough AND fever AND etc. =>
pneumonia
 if pneumonia => give antibiotics, etc.
Backward reasoning (goal-driven) use if lots of data
 start with “goal rule,” determine whether goal rule is true by
evaluating the truth of each necessary premise
 patient with lots of findings and symptoms
15(#total)
 is this lupus? => are 4 or more relevant criteria satisfied?
Telemedicine
“The delivery of medicine at a distance.”
Two basic forms:
 Live telemedicine - videoconference link
 Store-and-forward telemedicine - transmit for assessment
offline
Typical Telemedicine interaction: store and forward followed by
live interaction.
Data types
 text (e.g. patient's notes)
 image (e.g. x-ray) Telemedicine often relies on images (still or
moving)
Equipment
 general purpose (e.g. PCs)
 specialist (e.g. electronic stethoscope)
16(#total)
Telemedicine (contd.)
“The delivery of medicine at a distance.”
Telemedicine most useful when
 Specialist services are in very high demand or
 Patients are extremely isolated (Antarctica or remote
communities in Australia, Africa and Alaska)
Home care is often delivered by telemedicine
 Automatic monitoring and pill dispensing etc.
Telesurgery may also be considered as a subset of
telemedicine.
 Patient operated on by remotely controlled robotic arms etc.
17(#total)
Telemedicine - Case Study
8 Cornwall hospitals - Minor Injury Units:
 replacing "unviable" accident & emergency departments
 nurse led
Before:
 No access to specialist consultants
 Long wait for GP to be called in
After:
 Linked to specialist consultant in Royal Cornwall Hospital
 Recorded sessions useful for training nurses and for audit
 Cornwall Healthcare Trust expects to save £100,000 a year
http://www.bthealth.com/casestdy/cstudy/case11.htm
18(#total)
Telemedicine - Case Study
NHS Direct
http://www.nhsdirect.nhs.uk/
Biggest telemedicine project in the world
Mainly telephone service
Expanding to:
 Web
 Online diagnosis for common conditions
 Health encyclopaedia
 My NHS healthspace (personal info portal): news,
reminders, knowledge
 Digital TV
More information on Telemedicine: www.teis.nhs.uk
19(#total)
eHealth Standards
Not-for-profit organisations involved in the standardization
process
 American Society for Testing and Materials (ASTM),
 Health Level 7 (HL7)
 Healthcare Information and Management Systems Society
(HIMSS)
 CEN (European Committee for Standardisation)
 ISO TC215
Controlled medical vocabularies:
 Standardized Nomenclature of Medicine, Clinical Terms
(SNOMED-CT)
 Logical Observation Identifiers Names and Codes (LOINC)
 OpenGALEN Common Reference Model
20(#total)
eHealth / eScience : Cancer Diagnosis
Telemedicine on the Grid
 Multi-site videoconferencing
 Real-time delivery of microscope imagery
 Communication and archiving of radiological
images
 Supports multi-disciplinary meetings for the
review of cancer diagnoses and treatment.
 Remote access to computational medical
simulations of tumours and other cancer-related
problems
 Data-mining of patient record databases
 Improved clinical decision making.
 Currently clinicians travel large distances
 Grid technology can provide access to
appropriate clinical information and images
across the network.
More Information:
http://www.escience.cam.ac.uk/projects/telemed/
http://www.rcuk.ac.uk/escience/examplesh.asp
21(#total)
Benefits of eHealth
Reduced record keeping expenses
More accurate data
 No poor handwriting problems
Automated sharing among patients and provider
 Empower the patient to manage their own health via Internet information and decision support tools
Reduced office visits to get results
Avoidance of duplicating tests
Automatic summarisation/graphical displays of
context-relevant information to the physician
22(#total)
Benefits of eHealth (contd.)
 Decision Support Tools -> Improved decisions
 Remote access to data - e.g. ill while travelling
 Improved workflows
 Decreased risk of malpractice suits
 Ability to mine large record databases
 Research causes of disease
 Assess effectiveness of treatment programmes/drugs
 Monitor outbreaks of diseases
 Easier to conduct clinical trials and rapidly incorporate
research results in decision support tools
23(#total)
Medical Errors
Human Errors:
IOM Report, 1999
44,000 to 98,000 die in US annually from medical
errors
 at 44,000, would rank as 8th leading cause of
death
• car accidents: 43,458
• breast cancer: 42,297
• AIDS: 16,516
 7000 deaths from medication errors alone
24(#total)
Medical Errors
Does Software Help?
Therac-25
Radiation therapy machine
At least six known accidents between 1985 and 1987
Patients were given massive overdoses of radiation
At least five patients died of the overdoses
Causes: both organisational and technical
100 medical software recalls annually, 40 for software
errors
25(#total)
Why is eHealth Adopted Slowly?
Integration of IT
Public Services
(Health…)
2005 - WRONG!
Business Services
(Banks)
Manufacturing
1980
1990
2000
Jean-Claude Healy
May 2000
IT as a gadget
Trojan horse: networks, …
Full Integration of IT into Business (Organisational, Legal) Re-engineering of the system 26(#total)
Why is eHealth Adopted Slowly?
 Lags behind other industries by 10-15 years
 Complex regulations - e.g.
 Patient records
 Privacy laws
 Lack of interoperability/standards
 Doctors reject IT systems
Risks
 Potential for errors due to software bugs
 Highly coupled systems - greater risk of catastrophe
 Decision support systems could lead to mass produced
mistakes
 Privacy - data vulnerable to attack
27(#total)
Why is eHealth Adopted Slowly?
Many projects have been abandoned after massive
investment
Report by Standish Group (‘95):
 $250 billion each year on 175,000 IT projects
 31% projects cancelled before completion = $81
billion for canceled projects
 53% of projects cost 190% of estimated costs
28(#total)
New sources of "health" 1
29(#total)
New sources of "health" 2
30(#total)
New sources of "health" 3
31(#total)
eHealth - Busan, S Korea
Medical Tourism
 2 hours by air for 2 billion people
 1% with disposable income = 20 million
 Cardiac - Cancer - Mental Health
 Costs can be competitive
 Popular tourist resort for families
32(#total)
Summary
The Future of Healthcare - The banking metaphor
Existing Health on the Web
eHealth - terminology
Transmural Care
Electronic Medical Records (EMR)
Medical Records - Access
Clinical Decision Support Systems
Telemedicine - Case Studies
eHealth Standards
eHealth / eScience : Cancer Diagnosis
Benefits of eHealth
Medical Errors
Why is eHealth Adopted Slowly?
New sources of "health"
33(#total)