Managing Technological Change in Informatics

1
Managing Technological Change in Informatics

• Joan S. Ash, Ph.D., M.L.S., M.B.A.
• Associate Professor and Vice-Chair, Department of
  Medical Informatics and Clinical Epidemiology
• School of Medicine
• Oregon Health Science University

2
Are They Really Against Change?
3
Change needs to be managed

• Models of change
• Managing informatics projects
• Success factors for change
• Mercer Medical Center Case
• Unintended consequences of change
• Highland Medical Center Case

4
About Change
5
Change in health care is different from other
areas because of its complexity

• Life and death issues
• Personnel structure
• Ownership characteristics
• Stakeholders
• Exploding technologies
• (from Lorenzi and Riley, Managing Technological
  Change, 2nd ed., 2003)

6
Health care information systems mirror this
complexity

• Hardware
• Software
• Peopleware
• People skills
• Project management skills
• Technical skills
• (from Lorenzi and Riley)

7
The success of a project is 80 dependent on
the social/political skills of the developer
and 20 or less on the hardware and
software. Reed Gardner
8
(No Transcript)
9
Why do we resist change?

• Fear of unknown
• Habit
• Security
• Economic factors
• For clinicians, time

10
How can we overcome resistance?

• Education and communication
• Participation
• Facilitation and support
• Negotiation
• Manipulation
• Coercion

11
Models of change and change theories can be useful

• Lewins Three Stage Model
• Bridges Three Stage Transitions Model
• Diffusion of Innovations Theory

12
(No Transcript)
13
Changes and transitions are slightly different

• Change is a new situation
• Transition is a psychological reorientation
  people go through when coming to terms with change

14
The Alamance story is a remarkable one

• Successful implementation of CPOE in a rural
  community hospital in North Carolina

15
Physicians really use CPOE voluntarily there

• Whats their secret?
• Bridges!

16
Bridges outlines three phases of transition
ENDING
NEUTRAL ZONE
BEGINNING
From Bridges,W , Managing Transitions Making
the Most of Change, 1991
17
The Bridges Transition Model makes sense

• Begins with the ENDING
• Planning, partnerships, identify whos losing
  what, give people information
• NEUTRAL ZONE
• Transition monitoring team, strengthen intragroup
  connections, communicate, train
• BEGINNING
• Clarify purpose, create picture and plan, give
  people a part

18
Communication Channels
Adoption over Time
The Innovation
Consequences
Social System
Diffusion of Innovations Theory
19
There are some frightening stories about the
unintended consequences of CPOE

• Story 1 University of Virginia
• Story 2 Pediatric mortality
• Story 3 Viagra

20
A classic CPOE story told by Massaro was about
the University of Virginia

• House officers threatened to strike
• CPOE was dropped, then reintroduced
• Human consequences are avoidable

21
The Virginia story had a happy ending

• The University of Virginia story told by Massaro
• Interns look and feel like this, and they
  revolted
• The administration, clinical leaders, and house
  officers collaborated in developing order sets

22
A story about one pediatric hospital made
headlines

• Han et al. paper about increased pediatric
  mortality
• Han et al. Unexpected increased mortality after
  implementation of a commercially sold CPOE
  system. Pediatrics 2005.

23
The story of another pediatric hospital did not
make headlines

• Study by Del Beccaro et al. showed no increased
  mortality
• They listened to users and implemented with care
• -Del Beccaro et al. Pediatrics 2006

24

• Another unintended consequence of CPOE
• More about unintended consequences later

25
Managing Informatics Projects
26
Informatics Projects (adapted from Lorenzi
Riley 2003)

• Discontent

Planning
Develop/modify
Acceptance
Rollout
Test/pilot
27
Collaborative project management is key

• Project management that includes various
  stakeholders
• Three legged stool of
• Resources
• Schedule
• Performance expectations

28
What is informatics project success?

• Meets budget, schedule, and quality criteria
• Within scope
• Positive long term impact
• Meets expectations of all parties

29
What causes informatics project failure
(suboptimization)?

• Does not satisfy specifications
• Produces undesirable results
• Failure is in the eye of the beholder
• Failure often leads to success

30
The Mercer Case
31
The Mercer Medical Center case is illustrative of
the tough decisions needed

• Dr. Reeds dilemma
• The CEOs demand
• Dr. James efforts
• Clinical skeptics views

32
For our studies, we used qualitative methods

• At five hospitals for success factors study
• Five hospitals for unintended consequences study
• Thanks to the National Library
• of Medicine, NIH, for funding
• Grants LM06942 and ASMM10031

33
We used multiple researchers, methods, sites, and
types of subjects to assure trustworthiness

• Observation 784 person-hours
• Interviews, focus groups 87
• Data 4,069 pages
• Analysis meetings 86

34
Success Factors for Implementation
35
Success is related to time issues

• Three patterns
• Speed of order entry
• Speed of full order process
• Life cycle of implementation

36
Success means meeting information needs

• Three patterns
• Ways of doing things
• Organization of information
• System flexibility
• Customization

37
Success needs multidimensional integration

• Human integration
• Workflow integration
• Systems integration

38
Management is related to the existence of special
people

• Administrative leaders
• Clinical leaders, champions, curmudgeons
• Bridgers / support staff, help at the elbow
• Training
• Vendor
• Ash JS, Stavri PZ, Dykstra R, Fournier L.
  Implementing
• computerized physician orderentry The
  importance of
• special people. International Journal of
  Medical
• Informatics 200369235-250.

39
Curmudgeons can be helpful

• Provide feedback
• Furnish leadership
• convince the curmudgeon

40
There are important foundations for success

• Organizational culture
• Administrative commitment, vision
• Trust
• Leadership open to feedback
• Stable organization and vendor
• History of innovation and clinical systems

41
We must provide value to users and tradeoffs

• Value remote entry, legibility, decision
  support
• Tradeoffs time, rigidity, adapting to upgrades

42
Success is related to adequate financial
resources
43
Success requires improvement through evaluation
and learning

• Patterns
• Pilots
• Continuous modification
• Involvement

44
A meta success factor is continuous involvement

• The secret to clinician involvement is an
  adequate (and never ending) pizza budget

45
The unintended consequences of change
46
What does unintended mean?

• Unanticipated and not specifically a goal of the
  project
• Unintended most often connotes consequences
  that are unanticipated and undesirable
• They are not uniformly errors or mistakes they
  are simply surprises

47
We identified nine types of unintended adverse
consequences

• More/new work for clinicians
• Workflow issues
• Never ending system demands
• Paper persistence
• Changes in communication patterns
• Emotions
• New kinds of errors
• Changes in the power structure
• Overdependence on the technology
• Campbell E, Sittig DF, Ash JS, Guappone K,
  Dykstra R. Types of
• unintended consequences related to computerized
  provider order entry.
• JAMIA 2006 13547-556.
• Ash JS, Sittig DF, Poon EG, Guappone K, Campbell
  E, Dykstra RH. The extent and
• importance of unintended consequences related to
  computerized provider order entry.
• JAMIA 200714415-23.

48
We interviewed 176 hospital representatives

• Talked to staff at 299 of the 561 acute care
  hospitals
• Discovered that 89 listed as having CPOE did not,
  
• 34 hospitals had policies against doing surveys.
• Response rate (based on 176 valid interviews) was
  47 (using ISER calculation)
• Ash JS, Sittig DF, Poon EG, Guappone K, Campbell
  E, Dykstra RH. The extent and importance of
  unintended
• consequences related to computerized provider
  order entry. Journal of the American Medical
  Informatics Association 2008

49
We found that these systems cause more work, new
work, and never ending system demands

• More work for users and others
• Demand for hardware software purchase,
  implementation, and maintenance
• Personal order sets are difficult to
  standardize, update, or maintain over time
• Prevention strategies integration and value to
  users

50
Workflow issues are huge

• Mismatch between how work processes are intended
  to function and how they actually operate
• Clinical roles, procedures, and policies are
  overly standardized
• Confusion, duplication, misunderstanding across
  clinical role boundaries result
• Prevention strategies include time, integration,
  meeting information needs, providing value

51
CPOE alters communication among providers,
ancillary services, and clinical departments

• Causes reductions in face-to-face communication
• Causes illusion of communication, belief that
  the proper people will see it and act upon it
• Causes depersonalization
• Dykstra R. Computerized physician order
  entry and communication Reciprocal impacts.
  Proceedings AMIA 2002230-4.

52
Emotions run high

• CPOE evokes strong emotional responses
• strongly negative
• highly positive emotions
• Strong positive correlation between time system
  is in place and positive emotions
• Sittig DF, Krall M, Kaalaas-Sittig J, Ash JS.
  Emotional aspects of
• computer-based provider order entry A
  qualitative study. Journal
• of the American Medical Informatics Association
  2005 12(5)561-7.

53
CPOE can cause insidious silent errors

• Problems with data presentation, selection,
  entry
• Pick lists for data entry promote juxtaposition
  errors
• If the correct data entry location is not found,
  busy providers tend to place data where they
  might fit
• Prevention calls for all success factors

54
Paper persistence Your hospital will be
paperless, the same day my bathroom is
Michael Shabot, M.D.
55
CPOE causes changes in the power structure

• Loss of clinician autonomy
• Administration and I.T. gain power
• Clinical decision support can tell doctors how
  to practice
• Coalitions
• Ash JS, Sittig DF, Campbell E, Guappone K,
  Dykstra R.
• An unintended consequence of CPOE
  implementation
• Shifts in power, control, and autonomy.
  Proceedings
• AMIA 200611-15.

56
Clinical care becomes overdependent on the
computing infrastructure

• System failures wreak havoc unless good downtime
  procedures exist
• Reliance on clinical decision support may reduce
  learning
• If its in the computer it must be right!

57
System downtime, regardless of cause, can "create
chaos" for users and organizations
System Unavailability System Downtime
Not enough hardware
"it can be a real fight at times to get work
done, because people are always in need of a
computer."
"They use a white board screen saver in the ER
that keeps track of people in the ER. When the
hospital registry goes down, it can't provide the
patient ID number, so we can't enter or find any
information."
Single system component failure
Whole system is down
"It's funny now. When the computer goes down,
we don't remember how to document on paper."
58
Be prepared for and practice working during
system downtimes
Track system performance
Develop contingency plans for continued operation
during downtimes
Create robust backup systems and test them with
scheduled drills
59
Lets talk about the Highland Medical Center case

• Problems when patient transferred from emergency
  department to floor
• Issues related to the pharmacy system
• Barcode scanner problems
• Hands-free device issues