Impact of EHR on Quality of Care Farrokh Alemi, Ph.D.
Impact of EHR on Quality of CareFarrokh Alemi, Ph.D.
Objectives
Review the literature Discuss the mechanism Outline future directions
0
20
40
60
80
100
2002 2004 2006 2008 2010 2012 2014
Year
% H
os
pit
als
wit
h C
PO
E
Use of Computerized Physician Order Entry in Hospitals
Studies cited in Health Information Technology in the United States: The Information Base for Progress
Goal
0
20
40
60
80
100
2002 2004 2006 2008 2010 2012 2014
Year
% H
os
pit
als
wit
h C
PO
E
Use of Computerized Physician Order Entry in Hospitals
Studies cited in Health Information Technology in the United States: The Information Base for Progress
Goal
In some areas,80% of Urban Hospitals
Use CPOE
Where is the beef?Why are hospitals implementing CPOE?
OperationalEfficiency
Four Eras of IT in Health Care
CQI / TQM
Efficacyof Care
PatientSafety
Patient Financial SystemsDepartmental Clinical Systems
Process IntegrationWorkflow Transformation
Data Integration: Patient-Centric ViewClinical Decision Support – CPOE
1980 1990 2000 2010 2020TODAY
ANALYTICS &CONTINUOUS IMPROVEMENT
Institute of Medicine reports
Technology Infusionfrom Other Industries
From John Cuddeback MD
Develop improved practiceDeploy improved practiceRETROSPECTIVEReal Time
InformationInformation Knowledge
DataDataData
ANALYTICALSYSTEMS
Population Level
Analytical systems are essential for integration and transformation.
Analytical models, risk adjustment Ad hoc query tools—exploratory analysis,
hypothesis generation/testing Comparative data, “best” practices Support for quality improvement teams Practice profile reports for clinicians
POINT- OF - CARESYSTEMS
Patient Level Administrative systems (scheduling, ADT) Clinical observations, assessment, plan Orders—tied to protocols, w/ decision support Tests, results, documentation of care (eMAR) Capture outcomes, key process variables Error / near-miss reporting
External DataDATA WAREHOUSES
TRANSACTION SYSTEMSCLINICAL DATA REPOSITORY
ImprovedPractice
Concept or reality?
From John Cuddeback MD
Rate of Drug Events
4031 adult admissions 11 medical and surgical units 2 tertiary care hospitals Over a 6-month period
Detected by self-report Classified as ADEs or potential ADEs
Bates et al., JAMA 1995;274:29-34
Rate of Drug Events
247 ADEs & 194 potential ADEs In 100 non-obstetrical admissions
6.5 ADEs 5.5 potential ADEs
Bates et al., JAMA 1995;274:29-34
Fatal; 1%Life
threaten-ing; 12%
Serious, 30%Significant; 57%
Rate of Drug Events
Ordering49%
Transcription11%
Dispensing14%
Administration26%
48% of errors intercepted
No errors intercepted !
23% of errors intercepted
37% of errors intercepted
Bates et al., JAMA 1995;274:29-34
Can CPOE Reduce ADE? How exactly it will do so?
Rate of Errors after CPOE
Before and after study Baseline & 3 subsequent years
One hospital All patients Three medical units Seven to ten-week periods
Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma'Luf N, Boyle D, Leape L. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999 Jul-Aug;6(4):313-21.
Rate of Errors after CPOE
Medication errors fell 81% From 142 to 27 per 1,000 patient-days
Non-intercepted serious medication errors fell 86%
Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma'Luf N, Boyle D, Leape L. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999 Jul-Aug;6(4):313-21.
Rates of Errors after CPOE
Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma'Luf N, Boyle D, Leape L. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999 Jul-Aug;6(4):313-21.
5 months after CPOE
Allergy warning system
Drug-drug interaction system
Rates of Errors after CPOE
Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma'Luf N, Boyle D, Leape L. The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999 Jul-Aug;6(4):313-21.
5 months after CPOE
Allergy warning system
Drug-drug interaction system
Just storing and retrieving data is not much of a benefit.
Analyze and use the data.
Medication Errors in an ICU
22-bed general ICU Sampled before and after
28 weeks before Hand Written Prescribing 2, 10, 25 and 37 weeks after CPOE
Unit pharmacist recorded details of errors
Shulman R, Singer M, Goldstone J, Bellingan G. Medication errors: a prospective cohort study of hand-written and computerized physician order entry in the intensive care unit. Crit Care. 2005 Oct 5;9(5):R516-21. Epub 2005 Aug 8.
Less Medication Errors
Shulman R, Singer M, Goldstone J, Bellingan G. Medication errors: a prospective cohort study of hand-written and computerized physician order entry in the intensive care unit. Crit Care. 2005 Oct 5;9(5):R516-21. Epub 2005 Aug 8.
Types of Errors HWP CPOE
Drug prescribed on incorrect drug chart section (e.g. continuous IV infusion prescribed on 'when required' part of drug chart)
2 (2.8%)
1 (0.9%)
Drug needed but not given as not prescribed properly
3 (4.2%)
5 (4.3%)
Inappropriate/inadequate additional information on prescription to adequately administer the drug appropriately
8 (11.3%
)
12 (10.3%
)
Dose/units/frequency omitted on prescription
22 (31%)
1 (0.9%)
Prescription not signed or change not signed/dated
10 (14.1%
)
39 (33.3%
)
Types of Errors HWP CPOE
Still wrong next day after pharmacist recommended appropriate correction that was agreed with doctor
0 (0%)3
(2.6%)
Dose error12
(16.9%)31
(26.5%)
Wrong drug prescribed 3 (4.2%)6
(5.1%)
Incorrect route/unit 5 (7%)8
(6.8%)Formulary not followed without reason
3 (4.2%)1
(0.9%)Administration not in accordance with prescription
3 (4.2%)3
(2.6%)Required drug not prescribed 0 (0%) 7 (6%)
Severity of Errors
Error category Minor Moderate Major
HWP non-intercepted errors 43 0 0CPOE non-intercepted errors 93 4 0HWP intercepted errors 7 19 0CPOE intercepted errors 2 15 3
21
10
Rise in Mortality after CPOE
CPOE 13 months before 6 days implementation 5 months after
Subjects Children Admitted via inter-facility transport Regional, academic, tertiary-care level
children’s hospital
Han, Y. Y. et al. Paediatrics 2005;116:1506-1512
Copyright ©2005 American Academy of Pediatrics Han, Y. Y. et al. Paediatrics 2005;116:1506-1512
Rise in Mortality after CPOE
Rise in Mortality after CPOE
Inability to "preregister" patients (resolved)
Time needed to enter orders Need for a second physician Nurses away from the bedside
Changes to health care team dynamics Delays from centralization of pharmacy
Han, Y. Y. et al. Paediatrics 2005;116:1506-1512
EXAMPLE OF UNINTENDED CONSEQUENCES
With antibiotic administration, subsequent dosing schedules were not timed according to the time of initial dose administration but rather at predetermined default times
Han, Y. Y. et al. Paediatrics 2005;116:1506-1512
Same Mortality after CPOE
CPOE 13 months before 13 months after
Subjects Tertiary care PICU 20 beds 1100 annual admissions
Number of subjects 2533 children admitted 284 transported from other facilities
Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.
Same Mortality after CPOETotal
PatientsMortality,
%Relative
Risk 95% CI P
All patients 2533 3.83 0.82 0.55–1.21 .32
Before CPOE 1232 4.22
After CPOE 1301 3.46
Transfers 284 7.75 0.66 0.29–1.47 .30
Before CPOE 125 9.60
After CPOE 159 6.29
Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.
Same Mortality after CPOE
Pre-set orders 12 infant ICU-specific
16-PICU specific Extra- corporeal life support Renal replacement therapy Complex cardiac Transplant surgery Frequent orders preset
Active involvement
Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.
Same Mortality after CPOE
Visiting prior implementation Active Involvement in design
More order sets Pre-set completed sentences Code-set filtering
Process redesign Emergency medication dispensing Pre-registering transports
Continuous improvement
Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.
EXAMPLE OF INTENDED CONSEQUENCES
The first infant transported into the ICU: the resident was able to place an entire set of orders in <5 minutes without errors in a highly stressed environment
Campbell EM, Sittig DF, Ash JS, Guappone KP, Dykstra RH. J Am Med Inform Assoc. 2006 Sep-Oct;13(5):547-56.
What could go wrong?
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
Hospital Size (beds)
CPOE System
Up Since
Percent Orders Entered
Wishard Memorial, Indianapolis, IN
340 Homegrown
1973 100%
Massachusetts General Hospital, Boston, MA
893 Homegrown
1994 100%
Faulkner Hospital, Boston, MA
150 Meditech 2003 95%
Brigham & Women's Hospital, Boston, MA
725 Homegrown
1991 90%
Alamance Regional Medical Center, Burlington, NC
238 Eclipsys 1998 95%
What could go wrong?
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
Unintended Consequence Frequency (%) n = 324
More/new work for clinicians 19.8Workflow issues 17.6Never ending system demands 14.8Paper persistence 10.8Changes in communication patterns and practices
10.1
Emotions 7.7New kinds of errors 7.1Changes in the power structure 6.8Overdependence on technology 5.2Total 100
What could go wrong?1. More work for clinicians
Slows speed of clinical documentation Recovers over time Learning to use CPOE takes time Excessive clinical alerts Research not related to care Poor integration of multiple systems
2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns and
practices6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?
1. More work for clinicians2. Unfavorable workflow issues
Rigid modeling of work processes Fails to support all actors Simultaneous multiple orders
3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns
and practices6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands
More users require more access time More software updates Overhead in maintenance Single user exceptions for order sets Changes in practice
4. Problems related to paper persistence5. Untoward changes in communication patterns and
practices6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?
1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence
Integration with other clinical systems Temporary, handwritten data storage Paper reminders
5. Untoward changes in communication patterns and practices
6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?
1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns
and practices Replaces the nexus of previously interpersonal
conversations Order entry may precede or remotely follow rounds
6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?
1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns
and practices6. Negative emotions
“At first we hated every second of it.” “This is how everyone should work.”
7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns and
practices6. Negative emotions7. Generation of new kinds of errors
Problematic electronic data presentations Confusing order option presentations Inappropriate text entries Misunderstandings related to production versions Workflow process mismatches
8. Unexpected changes in the power structure9. Overdependence on the technology
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns and
practices6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure
Controls on who may do what and when Physicians report loss of professional autonomy Tends to encourage centralization IT department gains in power
9. Overdependence on the technologyAsh JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
What could go wrong?
1. More work for clinicians2. Unfavorable workflow issues3. Never ending system demands4. Problems related to paper persistence5. Untoward changes in communication patterns
and practices6. Negative emotions7. Generation of new kinds of errors8. Unexpected changes in the power structure9. Overdependence on the technology
System failures increasingly wreak havoc May increase access to protocols & educational materials
Ash JS, Berg M, Coiera E. J Am Med Inform Assoc. 2004 Mar-Apr;11(2):104-12.
Future
CPOE systems increase
Knowledge about use of system increases
Lessons learned are not
lost
Take Home LessonImpact on Quality is Complex