Clinical Decision Support Systems

Clinical Decision Support Systems

Nawanan Theera-Ampornpunt, M.D., Ph.D.June 14, 2014http://www.SlideShare.net/Nawanan

2

Outline

• What is a Decision?• Clinical Decision Making• Roles of IT in Decision Making• Clinical Decision Support Systems

– Definitions– Types & examples– Architecture

• Issues Related to CDS Implementation• Summary

3

WHAT IS A DECISION?

4

Wisdom

Knowledge

Information

Data

Data-Information-Knowledge-Wisdom (DIKW) Pyramid

5

Wisdom

Knowledge

Information

DataContextualization/

Interpretation

Processing/Synthesis/

Organization

Judgment

Data-Information-Knowledge-Wisdom (DIKW) Pyramid

6

Wisdom

Knowledge

Information

DataContextualization/

Interpretation

Processing/Synthesis/

Organization

Judgment

100,000,000

I have 100,000,000 baht in my bank

account

I am rich!!!!!

I should buy a luxury car(and a BIG house)!

Example

7

Example: Problem A

• Patient A has a blood pressure reading of 170/100 mmHg

• Data: 170/100

• Information: BP of Patient A = 170/100 mmHg

• Knowledge: Patient A has high blood pressure

• Wisdom (or Decision):– Patient A needs to be investigated for cause of HT– Patient A needs to be treated with anti-hypertensives– Patient A needs to be referred to a cardiologist

8

Example: Problem B

• Patient B is allergic to penicillin. He was recently prescribed amoxicillin for his sore throat.

• Data: Penicillin, amoxicillin, sore throat

• Information:– Patient B has penicillin allergy– Patient B was prescribed amoxicillin for his sore throat

• Knowledge:– Patient B may have allergic reaction to his prescription

• Wisdom (or Decision):– Patient B should not take amoxicillin!!!

9

Decision & Decision Making

• Decision– “A choice that you make about something

after thinking about it : the result of deciding” (Merriam-Webster Dictionary)

• Decision making– “The cognitive process resulting in the

selection of a course of action among several alternative scenarios.” (Wikipedia)

10

LET’S TAKE A LOOK AT PATIENT CARE PROCESS

11

Patient Care

Image Sources: (Left) Faculty of Medicine Ramathibodi Hospital (Right) /en.wikipedia.org/wiki/File:Newborn_Examination_1967.jpg (Nevit Dilmen)

12

EXERCISE 1Provide some examples of

“decisions” health care providers make

13

Clinical Decisions

• Patient Care– What patient history to ask?– What physical examinations to do?– What investigations to order?

• Lab tests• Radiologic studies (X-rays, CTs, MRIs, etc.)• Other special investigations (EKG, etc.)

– What diagnosis (or possible diagnosis) to make?

14

Clinical Decisions

• Patient Care– What treatment to order/perform?

• Medications• Surgery/Procedures/Nursing Interventions• Patient Education/Advice for Self-Care• Admission

– How should patient be followed-up?– With good or poor response to treatment, what

to do next?– With new information, what to do next?

15

Clinical Decisions

• Management– How to improve quality of care and clinical

operations?– How to allocate limited budget & resources?– What strategies should the hospital pursue &

what actions/projects should be done?

16

Clinical Decisions

• Public Health– How to improve health of population?– How to investigate/control/prevent disease

outbreak?– How to allocate limited budget & resources?– What areas of the country’s public health need

attention & what to do with it?

17

CLINICAL DECISION MAKING

18

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN

Elson, Faughnan & Connelly (1997)

Clinical Decision Making

19

PROBLEMS WITH HUMAN’S

DECISION MAKING

20Image Source: (Left) http://docwhisperer.wordpress.com/2007/05/31/sleepy-heads/ (Right) http://graphics8.nytimes.com/images/2008/12/05/health/chen_600.jpg

To Err is Human 1: Attention

21

Image Source: Suthan Srisangkaew, Department of Pathology, Facutly of Medicine Ramathibodi Hospital, Mahidol University

To Err is Human 2: Memory

22

To Err is Human 3: Cognition

• Cognitive Errors - Example: Decoy Pricing

The Economist Purchase Options

• Economist.com subscription $59• Print subscription $125• Print & web subscription $125

Ariely (2008)

16084

The Economist Purchase Options

• Economist.com subscription $59• Print & web subscription $125

6832

# of People

# of People

23

• Medical Errors–Drug allergies–Drug interactions

• Missing Abnormal Lab Findings• Clinical Practice Guidelines• Bias in Judgment & Decision-Making

Common Errors in Healthcare

24

ROLES OF INFORMATION TECHNOLOGY

IN DECISION MAKING

25

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN


Clinical Decision Making

26

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN


Possible Human Errors

Possibility of Human Errors

27

CLINICAL DECISION SUPPORT SYSTEMS

(CDS)

28

• Clinical Decision Support (CDS) “is a process for enhancing health-related decisions and actions with pertinent, organized clinical knowledge and patient information to improve health and healthcare delivery” (Including both computer-based & non-computer-based CDS)

(Osheroff et al., 2012)

What Is A CDS?

29

• Computer-based clinical decision support (CDS): “Use of the computer [ICT] to bring relevant knowledge to bear on the health care and well being of a patient.”

(Greenes, 2007)

What Is A CDS?

30

• The real place where most of the values of health IT can be achieved

• There are a variety of forms and nature of CDS

Clinical Decision Support Systems (CDS)

31

• Expert systems– Based on artificial

intelligence, machine learning, rules, or statistics

– Examples: differential diagnoses, treatment options

CDS Examples

Shortliffe (1976)

32

• Alerts & reminders– Based on specified logical conditions

• Drug-allergy checks• Drug-drug interaction checks• Drug-lab interaction checks• Drug-formulary checks• Reminders for preventive services or certain actions

(e.g. smoking cessation)• Clinical practice guideline integration (e.g. best

practices for chronic disease patients)

CDS Examples

33

Example of “Reminders”

34

• Reference information or evidence-based knowledge sources–Drug reference databases–Textbooks & journals–Online literature (e.g. PubMed)–Tools that help users easily access

references (e.g. Infobuttons)

CDS Examples

35

Infobuttons

Image Source: https://webcis.nyp.org/webcisdocs/what-are-infobuttons.html

36

• Pre-defined documents– Order sets, personalized “favorites”– Templates for clinical notes– Checklists– Forms

• Can be either computer-based or paper-based

CDS Examples

37

Order Sets

Image Source: http://www.hospitalmedicine.org/ResourceRoomRedesign/CSSSIS/html/06Reliable/SSI/Order.cfm

38

• Simple UI designed to help clinical decision making–Abnormal lab highlights–Graphs/visualizations for lab results–Filters & sorting functions

CDS Examples

39

Abnormal Lab Highlights

Image Source: http://geekdoctor.blogspot.com/2008/04/designing-ideal-electronic-health.html

40

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN


How CDS Supports Decision Making

Abnormal lab highlights

41

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Order Sets

42

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Drug-Allergy Checks

43

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Drug-Drug Interaction

Checks

44

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Drug-Drug Interaction

Checks

45

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Clinical Practice Guideline

Alerts/Reminders

46

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Integration of Evidence-Based Resources (e.g. drug databases,

literature)

47

External Memory

Knowledge Data

Long Term Memory

Knowledge Data

Inference

DECISION

PATIENT

Perception

Attention

WorkingMemory

CLINICIAN



Diagnostic/Treatment Expert Systems

48

User User Interface

Patient Data

Inference Engine

Knowledge BaseOther Data

• Rules & Parameters• Statistical data• Literature• Etc.

• System states• Epidemiological/surveillance data• Etc.

Example of CDS Architecture

Other Systems

49

ISSUES RELATED TO CDS IMPLEMENTATION

50

• How will CDS be implemented in real life?• Will it interfere with user workflow?• Will it be used by users? If not, why?• What user interface design is best?• What are most common user complaints?• Who is responsible if something bad

happens?• How to balance reliance on machines &

humans

Human Factor Issues of CDS

51

IBM’s Watson

Image Source: socialmediab2b.com

52

Image Source: englishmoviez.com

Rise of the Machines?

53

Issues• CDSS as a supplement or replacement of clinicians?

– The demise of the “Greek Oracle” model (Miller & Masarie, 1990)

The “Greek Oracle” Model

The “Fundamental Theorem”

Friedman (2009)


Wrong Assumption

Correct Assumption

54

• Features with improved clinical practice (Kawamoto et al., 2005)

– Automatic provision of decision support as part of clinician workflow

– Provision of recommendations rather than just assessments

– Provision of decision support at the time and location of decision making

– Computer based decision support

• Usability & impact on productivity


55

Issues• Alert sensitivity & alert fatigue

Alert Fatigue

56

• Liabilities– Clinicians as “learned intermediaries”

• Prohibition of certain transactions vs. Professional autonomy (see Strom et al., 2010)

Ethical-Legal Issues of CDS

57

Workarounds

58

• “Unanticipated and unwanted effect of health IT implementation” (www.ucguide.org)

• Resources– www.ucguide.org– Ash et al. (2004)– Campbell et al. (2006)– Koppel et al. (2005)

Unintended Consequences of CDS & Health IT

59

Ash et al. (2004)


60

• Errors in the process of entering and retrieving information– A human-computer interface that is not

suitable for a highly interruptive use context– Causing cognitive overload by

overemphasizing structured and “complete” information entry or retrieval

• Structure• Fragmentation• Overcompleteness

Ash et al. (2004)


61

• Errors in communication & coordination– Misrepresenting collective, interactive work as

a linear, clearcut, and predictable workflow• Inflexibility• Urgency• Workarounds• Transfers of patients

– Misrepresenting communication as information transfer

• Loss of communication• Loss of feedback• Decision support overload• Catching errors

Ash et al. (2004)


62

• Which type of CDS should be chosen?• What algorithms should be used?• How to “represent” knowledge in the system?• How to update/maintain knowledge base in

the system?• How to standardize data/knowledge?• How to implement CDS with good system

performance?

Technical Issues of CDS

63

• Choosing the right CDSS strategies• Expertise required for proper CDSS design &

implementation• Everybody agreeing on the “rules” to be enforced• Evaluation of effectiveness

Other Issues

64

• Speed is Everything• Anticipate Needs and Deliver in Real Time• Fit into the User’s Workflow• Little Things (like Usability) Can Make a Big Difference• Recognize that Physicians Will Strongly Resist Stopping• Changing Direction Is Easier than Stopping• Simple Interventions Work Best• Ask for Additional Information Only When You Really Need It• Monitor Impact, Get Feedback, and Respond• Manage and Maintain Your Knowledge-based Systems

Bates et al. (2003)

“Ten Commandments” for Effective CDS

65

• There are several decisions made in a clinical patient care process

• Data leads to information, knowledge, and ultimately, decision & actions

• Human clinicians are not perfect and can make mistakes

• A clinical decision support systems (CDS) provides support for clinical decision making (to prevent mistakes & provide best patient care)

• A CDS can be computer-based or paper-based

Key Points

66

• CDS comes in various forms, designs, and architecture

• There are many issues related to design, implementation and use of CDS– Technical Issues– Human Factor Issues– Ethical-Legal Issues

Key Points

67

• Current mindset: CDS should be used to help, not replace, human providers

• Be attentive to workarounds, alert fatigues, and other unintended consequences of CDS– They can cause more danger to patients!!– They may lead users to abandon using CDS (a

failure)• There are recommendations on how to best

design & implement CDS

Key Points

68

Intelligent & helpful machines

Machines with a human touch

Machines that replace humans

HAL 9000 Data David NS-5

Dangerous killer machines

What Will The Future Be for Health Care?

69

References

• Ash JS, Berg M, Coiera E. Some unintended consequences of information technology in health care: the nature of patient care information system-related errors. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.

• Ariely D. Predictably irrational: the hidden forces that shape our decisions. New York City (NY): HarperCollins; 2008. 304 p.

• Bates DW, Kuperman GJ, Wang S, Gandhi T, Kittler A, Volk L, Spurr C, Khorasani R, Tanasijevic M, Middleton B. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J Am Med Inform Assoc. 2003 Nov-Dec;10(6):523-30.

• Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. Types of unintended consequences related to computerized provider order entry. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.

• Elson RB, Faughnan JG, Connelly DP. An industrial process view of information delivery to support clinical decision making: implications for systems design and process measures. J Am Med Inform Assoc. 1997 Jul-Aug;4(4):266-78.

• Friedman CP. A "fundamental theorem" of biomedical informatics. J Am Med Inform Assoc. 2009 Apr;16(2):169-170.

70

References

• Greenes RA. Clinical decision support: the road ahead. Oxford (UK): Elsevier; 2007. 581 p.

• Institute of Medicine, Committee on Quality of Health Care in America. To err is human: building a safer health system. Kohn LT, Corrigan JM, Donaldson MS, editors. Washington, DC: National Academy Press; 2000. 287 p.

• Kawamoto K, Houlihan CA, Balas EA, Lobach DF. Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features critical to success. BMJ. 2005 Apr 2;330(7494):765.

• Koppel R, Metlay JP, Cohen A, Abaluck B, Localio AR, Kimmel SE, et al. Role of computerized physician order entry systems in facilitating medication errors.JAMA. 2005 Mar 9;293(10):1197-1203.

• Miller RA, Masarie FE. The demise of the "Greek Oracle" model for medical diagnostic systems. Methods Inf Med. 1990 Jan;29(1):1-2.

• Osheroff JA, Teich JM, Levick D, Saldana L, Velasco FT, Sittig DF, Rogers KM, Jenders RA. Improving outcomes with clinical decision support: an implementer’s guide. 2nd ed. Chicago (IL): Healthcare Information and Management Systems Society; 2012. 323 p.

71

References

• Shortliffe EH. Computer-based medical consultations: MYCIN. New York (NY): Elsevier; 1976. 264 p.

• Strom BL, Schinnar R, Aberra F, Bilker W, Hennessy S, Leonard CE, Pifer E. Unintended effects of a computerized physician order entry nearly hard-stop alert to prevent a drug interaction: a randomized controlled trial. Arch Intern Med. 2010 Sep 27;170(17):1578-83.

Clinical Decision Support Systems

Health & Medicine