Health Informatics: CDSS and EHRs

1
Health Informatics CDSS and EHRs

• Carolyn Green
• Doctoral Candidate (Health Informatics), UVic
• SLAIS presentation
• Mar 5th, 2003

2
Outline

• Health informatics
• Clinical decision support systems
• versus expert systems
• Electronic health records
• BC initiatives

3
Health (Medical) Informatics

• The study and application of methods to improve
  the management of patient data, medical
  knowledge, population data and other information
  relevant to patient care and community health
• progressive shift in focus from technology to
  information management

4
Clinical Decision Support Systems

• Any computer program designed to help health
  professionals make clinical decisions --
  Shortliffe (ex. PDAs)
• Active knowledge systems which use two or more
  items of patient data to generate case-specific
  advice -- Whatt and Spiegelhalter

5
Tan definition

• CDSS provide
• clinical databanks and algorithms
• analytic or pathophysiologic models
• clinical decision theoretical models
• statistical pattern recognition methods
• symbolic reasoning
• clinical expert knowledge bases
• to support and enhance the diagnostic and
  prognostic thinking and cognitive reasoning
  strategies of expert and nonexpert clinicians

6
Expert systems

• use specialized knowledge about a particular
  problem area rather than just general knowledge
• use symbolic reasoning (strategies and heuristics
  characteristic of expert reasoning) rather than
  only numerical calculations
• perform at a level of competence better than
  that of nonexpert humans

7
Comparison of goals

• CDSS - assist clinical decision making
• Medical ES - duplicate human experts in
  diagnostic and prognostic decisions (Greek Oracle
  approach)
• Expert DSS - assist clinician with improved
  intelligence and expert advice

8
Comparison of data sources

• CDSS - factual databases and model databases
• Expert DSS - data and model bases with knowledge
  elements

9
Comparison of strategies

• CDSS - decision models responsibility lies with
  user
• Expert DSS - incorporate flexible influence
  engines user responsible for decision-making
  strategy with assistance from system

10
Comparison of user interaction

• CDSS - interactive
• Medical ES - passive
• Expert DSS - active

11
Comparison of constraints

• CDSS - limited to model management no
  explanation facilities provided
• Expert DSS - flexibility in the management of
  data, models, and knowledge explanation
  facilities provided

12
Functions

• Information control storage, retrieval, and
  organization of data, information and knowledge
  ex. The knowledge cart
• Process models computational models that predict
  the behavor of real-world processes (ex. What
  if?) ex, Bed Allocation

13
Functions Continued

• Choice models - techniques involving the
  integration of individual criteria across
  different aspects or alternative choices - Drug A
  or B
• Representational aids visual, spatial, and
  matrix data and model representation methods -
  digital dashboards - report cards

14
Gold standard CDS system

• To provide the right information at the right
  time to the right person in order to facilitate
  sound and timely clinical judgements in an
  economically sustainable manner
• Boverman 99

15
The right information

• patient specific, comprehensive, accurate,
  evidence based, and relevant
• tools relating to diagnostic or treatment
  decisions are user-friendly, research-based,
  peer-reviewed and adequately tested in pilot
  evaluations

16
The right time

• timely
• up-to-date
• available 24 hours, 7 days a week
• quickly accessible

17
The right people

• health care providers making decisions regarding
  patient care
• patients participating in decisions regarding
  their care
• Authorized others with privacy, confidentiality
  and data security safeguards in place

18
The right place

• the point of care
• portable as the setting of care is variable
• may include telehealth initiatives
• may include non-traditional settings such as the
  home

19
Economically sustainable

• is cost effective, cost saving and at a cost that
  is feasible within available resources and HCO
  priorities and adaptable to changing needs

20
CDSS do they work?

• decades of research with few functioning systems
• credible research has emerged demonstrating that
  specific CDSS applications have resulted in
  improved patient outcomes, as well as cost
  reductions an elusive win - win combination

21
Computerized physician order entry on medication
error prevention

• The non-missed-dose medication error rate fell 81
  percent, from 142 per 1,000 patient-days in the
  baseline period to 26.6 per 1,000 patient-days in
  the final period (P lt 0.0001)
• Large differences were seen for all main types of
  medication errors dose errors, frequency errors,
  route errors, substitution errors, and allergies

Bates et al, 1999
22
Computer-assisted management of antibiotics/
antiinfective agents

• Significant reductions were found in the cost of
  antiinfective agents (average 102 vs. 427 and
  340), in total hospital costs (average 26,315
  vs. 44,865 and 35,283), and in the length of
  the hospital stay days (average 10.0 vs. 16.7 and
  12.9)
• Marked reductions in the mean number of days of
  excessive drug dosage and in adverse events
  caused by antiinfective agent
• Evans et al, 1998

23
www.theradoc.com
24
Automatic detection of acute bacterial pneumonia
from chest x-ray reports

• The applications include a decision support
  system called the antibiotic assistant, a
  computerized clinical protocol for pneumonia, and
  a quality assurance application in the radiology
  department
• In extracting pneumonia related concepts from
  chest x-ray reports, the performance of the
  natural language processing system was similar to
  that of physicians
• Fiszman et al, 2000

25
Patient interactive computer-generated behavioral
intervention systems

• Of 37 eligible trials, 92 reported either
  statistically significant or improved outcomes
• Of the 14 targeted intervention studies, 93
  reported improved outcomes
• Revere Dunbar, 2001

26
What is the catch?

• If CDSS are all they promise to be why dont we
  implement them tomorrow?

27
Electronic medical record

• and standardized information infrastructure

28
Paper system

• Illegible
• Incomplete
• Difficult to access in more than 1 place
• Insecure from unauthorized uses and users

29
An Electronic Health Record (EHR)

• provides each British Columbian with a secure
  and private lifetime record of their key health
  history and care within the health system. The
  record is available electronically to authorized
  health care providers and the individual
  anywhere, anytime,in support of high-quality
  care.

30
Information technology blueprint

• Framework for an HER for British Columbians
• Presented by the Council of Chief Information
  Officiers
• January, 2003

31
Building blocks
32
Strictly confidential...

• But readily accessible to those that need them

33
Privacy

• The right of an individual to control the
  circulation of information about himself/herself
  within social relationships freedom from
  interference in an individuals private life an
  individuals right to protection of data
  regarding him/her against misuse or unjustified
  publication

34
Responsibilities of HC system

• Confidentiality A third parties obligation to
  protect the information with which it has been
  entrusted

• Accountability The concept that individual
  persons or entities can be held responsible for
  specific actions such as obtaining informed
  consent or breaching confidentiality

35
Information management

• Security the degree to which data, databases, or
  other assets are protected from exposure to
  accidental or malicious disclosure unauthorized
  access, interruption, modification, removal or
  destruction.

36
Information management

• Data integrity the preservation of the original
  quality and accuracy of data, in any medium or
  form

37
Factors associated with success

• Organizational leadership, commitment and vision
• Improving clinical processes and patient care
• Involving clinicians in design and modification
  of system
• Maintaining or improving clinical productivity
• Building momentum and support among clinicians