The Role of Health Information Technology in Medication Safety

Wednesday, April 25, 200712:00 – 1:00 p.m. EDT

Moderator: Paul Sharek, MD, MPH, FAAPAssistant Professor of Pediatrics, Stanford School of

MedicineMedical Director of Quality ManagementChief Clinical Patient Safety OfficerLucile Packard Children’s HospitalPalo Alto, California

This activity was funded through an educational grant from the Physicians’

Foundation for Health Systems Excellence.

Disclosure of Financial Relationships and Resolution of Conflicts of Interest for AAP CME Activities

Grid

The AAP CME program aims to develop, maintain, and increase the competency, skills, and professional performance of pediatric healthcare professionals by providing high quality, relevant, accessible and cost-effective educational experiences. The AAP CME program provides activities to meet the participants’ identified education needs and to support their lifelong learning towards a goal of improving care for children and families (AAP CME Program Mission Statement, August 2004).

The AAP recognizes that there are a variety of financial relationships between individuals and commercial interests that require review to identify possible conflicts of interest in a CME activity. The “AAP Policy on Disclosure of Financial Relationships and Resolution of Conflicts of Interest for AAP CME Activities” is designed to ensure quality, objective, balanced, and scientifically rigorous AAP CME activities by identifying and resolving all potential conflicts of interest prior to the confirmation of service of those in a position to influence and/or control CME content. The AAP has taken steps to resolve any potential conflicts of interest.

All AAP CME activities will strictly adhere to the 2004 Updated Accreditation Council for Continuing Medical Education (ACCME) Standards for Commercial Support: Standards to Ensure the Independence of CME Activities. In accordance with these Standards, the following decisions will be made free of the control of a commercial interest: identification of CME needs, determination of educational objectives, selection and presentation of content, selection of all persons and organizations that will be in a position to control the content, selection of educational methods, and evaluation of the CME activity.

The purpose of this policy is to ensure all potential conflicts of interest are identified and mechanisms to resolve them prior to the CME activity are implemented in ways that are consistent with the public good. The AAP is committed to providing learners with commercially unbiased CME activities.

DISCLOSURES Activity Title: Safer Health Care for Kids - Webinar The Role of Health Information Technology in Medication Safety Activity Date: April 25, 2007

DISCLOSURE OF FINANCIAL RELATIONSHIPS All individuals in a position to influence and/or control the content of AAP CME activities are required to disclose to the AAP and subsequently to learners that the individual either has no relevant financial relationships or any financial relationships with the manufacturer(s) of any commercial product(s) and/or provider of commercial services discussed in CME activities.

Name Name of Commercial Interest(s)*

(*Entity producing

health care goods

or services)

Nature of Relevant Financial

Relationship(s) (If yes, please list: Research Grant,

Speaker’s Bureau, Stock/Bonds

excluding mutual funds, Consultant,

Other - identify)

CME Content Will Include

Discussion/ Reference to Commercial

Products/Services

Disclosure of Off-Label (Unapproved)/Investigational Uses of Products

AAP CME faculty are required to disclose to the AAP and to learners when they plan to discuss or

demonstrate pharmaceuticals and/or medical devices that are not approved

Rainu Kaushal, MD, MPH

No No No No

DISCLOSURESSAFER HEALTH CARE FOR KIDS - PROJECT ADVISORY COMMITTEE AND STAFF DISCLOSURE OF FINANCIAL RELATIONSHIPS All individuals in a position to influence and/or control the content of AAP CME activities are required to disclose to the AAP and subsequently to learners that the individual either has no relevant financial relationships or any financial relationships with the manufacturer(s) of any commercial product(s) and/or provider of commercial services discussed in CME activities.

Name Name of Commercial Interest(s)*

(*Entity producing health care goods

or services)

Nature of Relevant Financial Relationship(s)

(If yes, please list: Research Grant, Speaker’s

Bureau, Stock/Bonds excluding mutual funds,

Consultant, Other - identify)

CME Content Will Include Discussion/

Reference to Commercial Products/Services

Disclosure of Off-Label (Unapproved)/Investigational Uses

of Products AAP CME faculty are required to

disclose to the AAP and to learners when they plan to discuss or

demonstrate pharmaceuticals and/or medical devices that are not approved

Karen Frush, MD, FAAP (PAC Member)

No No No No

Uma Kotagal, MD, MBBS, MSc, FAAP (PAC Member)

No No No No

Christopher Landrigan, MD, MPH, FAAP (PAC Member)

No No No No

Marlene R. Miller, MD, MSc, FAAP (PAC Chair)

No No No No

Paul Sharek, MD, MPH. FAAP (PAC Member)

No No No No

Erin Stucky, MD, FAAP (PAC Member)

No No Not sure No

Nancy Nelson (AAP Staff) No No No No

Melissa Singleton, MEd (Project Manager – AAP Consultant)

No No No No

Junelle Speller (AAP Staff) No No No No

Linda Walsh, MAB (AAP Staff)

No No No No

DISCLOSURESAAP COMMITTEE ON CONTINUING MEDICAL EDUCATION (COCME) DISCLOSURE OF FINANCIAL RELATIONSHIPS All individuals in a position to influence and/or control the content of AAP CME activities are required to disclose to the AAP and subsequently to learners that the individual either has no relevant financial relationships or any financial relationships with the manufacturer(s) of any commercial product(s) and/or provider of commercial services discussed in CME activities.

Name Name of Commercial Interest(s)*

(*Entity producing health care goods

or services)

Nature of Relevant Financial Relationship(s)

(If yes, please list: Research Grant, Speaker’s

Bureau, Stock/Bonds excluding mutual funds,

Consultant, Other - identify)

CME Content Will Include Discussion/

Reference to Commercial Products/Services

Disclosure of Off-Label (Unapproved)/Investigational Uses

of Products AAP CME faculty are required to

disclose to the AAP and to learners when they plan to discuss or

demonstrate pharmaceuticals and/or medical devices that are not approved

Ellen Buerk, MD, FAAP

No No No No

Meg Fisher, MD, FAAP

No No No No

Robert A. Wiebe, MD, FAAP

No No Not sure No

Jack Dolcourt, MD, FAAP

No No No No

Thomas W. Pendergrass, MD, FAAP

No No No No

Beverly P. Wood, MD, FAAP No No No No

CME CREDIT

The American Academy of Pediatrics (AAP) is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

 The AAP designates this educational activity for a

maximum of 1.0 AMA PRA Category 1 Credit™. Physicians should only claim credit commensurate with the extent of their participation in the activity.

 This activity is acceptable for up to 1.0 AAP credit.

This credit can be applied toward the AAP CME/CPD Award available to Fellows and Candidate Fellows of the American Academy of Pediatrics.

OTHER CREDIT

This webinar is approved by the National Association of Pediatric Nurse Practitioners (NAPNAP) for 1.2 NAPNAP contact hours of which 0.3 contain pharmacology (Rx) content. The AAP is designated as Agency #17. Upon completion of the program, each participant desiring NAPNAP contact hours should send a completed certificate of attendance, along with the required recording fee ($10 for NAPNAP members, $15 for nonmembers), to the NAPNAP National Office at 20 Brace Road, Suite 200, Cherry Hill, NJ 08034-2633.

 The American Academy of Physician Assistants

accepts AMA PRA Category 1 Credit(s)TM from organizations accredited by the ACCME .

Rainu Kaushal, MD, MPHAssociate Professor of Pediatrics and Public Health,Weill Medical College of Cornell UniversityDirector of Pediatric Quality and Safety, KCCH at NYPHNew York, New York

11

The Role of Health Information Technology

in Medication Safety

April 25, 2007

12:00 – 1:00 p.m. EDT

12

Learning Objectives

Upon completion of this activity, you will be able to:

Review the epidemiology of ambulatory medication safety in pediatrics.

Describe the role of health information technology in improving medication safety.

Utilize specific health information technology applications.

13

Historical Perspective

In 1925, 4 main types of adverse events identified for hospitalized patients: Burns due to hot water Delirious patients jumping from hospital

windows Accidents connected with hospital elevators Mistakes in the use of drugs

Aikens C. Study in the Ethics for Nurses. Philadelphia: Saunders; 1925

14

Patient Safety and Medical Errors Become a National Concern

November 29, 1999

15

16

IOM Report: Preventing Medication Errors

One medication error per day per inpatient Variation across institutions

At least 25% of injuries from medications are preventable

Annual preventable injuries from medications 380,000-450,000 in hospitals for $3.5 billion 800,000 in long-term care 530,000 in Medicare ambulatory patients for $887 million

17

Preventing Medication Errors: Recommendations Patient’s rights and enhancing consumer

information Utilizing HIT

Prescribing and transmission of all prescriptions electronically by 2010

Appropriate clinical decision support Adopt other appropriate technology (eMAR, bar

coding, smart iv pumps) Monitor for medication errors Standards for HIT

More research Pediatrics a prime focus area

18

Overview

Why medication errors occur in children Pediatric medication error epidemiology

Inpatient Outpatient

Prevention strategies HIT

Safety and qualityFinancial

19

Why medication errors occur in children

Weight based dosing Stock medicine dilution Ten fold errors Decreased communication abilities Inability to self-administer medications Increased vulnerability of young, critically ill

children Immature renal and hepatic systems

20

Definitions

Near Misses

ADEs

Medication Errors

21

Comparisons of Adult and Pediatric Inpatients

Pediatrics Adults**

Orders reviewed 10,778 10,070

Medication errors 616 (5.7%) 530 (5.3%)

Near Misses 115 (1.1%) 35 (0.35%) *

ADEs 26 (0.24%) 25 (0.25%)

Preventable ADEs 5 (0.05%) 5 (0.05%)

*p value <0.05

**Study performed at Brigham and Women’s Hospital in 1992 using similar methods Kaushal et al, JAMA 2001

22

Error Stage for Medication Errors

Ordering74%

Administer2%

Monitor1%

Dispense13%

Transcribe10%

23

Near Misses in the NICU per 100 orders

2.8

1.3

0.77

0.35

0

0.5

1

1.5

2

2.5

3

NICU PICU Med/Surg Adult

**

* P<0.001JAMA 2001;285;2114-20

*

24

Ambulatory Setting: Medication Errors

2952 medication 2952 medication errorserrors 1.6 errors per 1.6 errors per

patientpatient 1.3 errors per 1.3 errors per

prescriptionprescription 521 (12%) rx 521 (12%) rx

inappropriate inappropriate abbreviationabbreviation

1389 (64%) rx 1389 (64%) rx partially illegiblepartially illegible

WeightOtherDoseRouteAmountStrength

25

Preliminary Results For Six Office Practices

Rate N 95% CI

Preventable ADEs 3% 57 3-4

Non-preventable ADEs

13% 226 11-15

Ameliorable ADEs 9% 152 7-10

Non-intercepted Near Misses

25% 455 22-29

26

Stages

Administering

Ordering

Dispensing

Transmitting

Administering

Ordering

Dispensing

Transmitting

Preventable ADEs Near Misses

27

Why Do Errors Occur?

Physician writes an order Nursing, pharmacist, and clerical staff

mechanisms are in place to carry out orders What occurs in reality?

28

We deliver medications in hospitals in a manner that essentially hasn’t changed in 60 years.

physician writes order

secretary transcribes

secretary faxes

nurse double checks

pharmacist checksdrug interactions

pharmacist checksorder/allergies

pharmacist enter order

pharmacist receives fax

nurse checks drugagainst med sheets

nurse obtains drugfrom delivery system

pharmacy tech placesdrawer in delivery system

pharmacy tech loadsdrawer

nurse checks patient,drug, dose, route, time

nurse administersdrug

Is a double checknecessary?

Is drug administeredvia pump

If order incorrect:multiple other steps

29

We deliver medications in hospitals in a manner that essentially hasn’t changed in 60 years.

physician writes order

secretary transcribes

secretary faxes

nurse double checks

pharmacist checksdrug interactions

pharmacist checksorder/allergies

pharmacist enter order

pharmacist receives fax

nurse checks drugagainst med sheets

nurse obtains drugfrom meditrol

pharm. tech placesdrawer in meditrol

pharm. tech loadsdrawer

nurse checks patient,drug, dose, route, time

nurse administersdrug

Is a double checknecessary?

Is drug administeredvia pump

If order incorrect:multiple other steps

Consider: What’s the error rate in each one of

Consider: What’s the error rate in each one of

these steps - 5%, 1%, 0.1%, 0.01%

these steps - 5%, 1%, 0.1%, 0.01%

30

The Role of Complexity inPreventing Errors

Probability of Performing Perfectly:Probability of Performing Perfectly:

# of

ele

men

ts

100

50

1

25

0.006

0.08

0.95

0.28

0.37

0.61

0.99

0.78

0.90

0.95

0.999

0.96

0.99

0.995

0.9999

0.998

0.95 0.99 0.999 0.9999

Probability of success, each element:

31

No one makes an error on purpose.

Lucian LeapeLucian Leape

32

Everyone makes dumb mistakes every day.

33

No one admits an error if you punish them for it.

34

Error Reduction: Systems Approach Culture of Safety

Non-punitive systems Multidisciplinary error prevention

Involve parents and families Executive walk rounds Medication Safety Initiative

Strong, clear, visible attention to safety Avoid fatigue Minimize distractions Independent double checks

2nd checker will detect ~ 90% of errors made by first checker

HIT

35

How will HIT help?

Improve communication What has been done and by whom

Improve accessibility Paper records unavailable 1/3 of the time Physicians spend 20-30% of their time searching

for and organizing information Require key pieces of information Improve information retrieval

Impossible to store all needed clinical information in a physician’s head

36

How will HIT help? (cont)

“Just in time” decision support Assist with calculations Make the right thing the easiest to do Perform checks in real-time

Assist with monitoring Advance the quality agenda

Quality measurement Low cost way to diffuse evidence-based best

practices

37

Impact of HIT on Quality, Efficiency and Costs

4 Benchmark institutions4 Benchmark institutions QualityQuality

Increased adherence to guidelinesIncreased adherence to guidelines Enhanced surveillance and monitoringEnhanced surveillance and monitoring Medication errorsMedication errors

EfficiencyEfficiency Decreased utilization of careDecreased utilization of care

FinancialFinancial

Chaudhry B et al, Annals 2006

38

Types of HIT

EHR, Electronic health record CPOE, Computerized physician order entry Robots Smart IV systems Bar coding Telemedicine Automated drug delivery systems

39

CPOE: Low hanging fruit

Illegible handwriting Incomplete information Unacceptable Abbreviations Lack of leading zeros Inclusion of trailing zeros

40

Key Areas of Decision Support

Requiring complete ordersRequiring complete orders Default dosesDefault doses Drug-allergy checkingDrug-allergy checking DDI checkingDDI checking Renal dosingRenal dosing Geriatric dosingGeriatric dosing Drug-lab checkingDrug-lab checking Dose ceilingsDose ceilings

41

Handwriting example

42

“Corollary” order reminders reduce errors of omission

Target - corollary order pairs (n=87) NSAID – creatinine level Aminoglycoside – drug levels, creatinine Warfarin – routine protimes

Intervention -- display reminder at time of ordering

Overhage JAMIA 1997;4:364-375

43

Effect of alerts on compliance

Overhage JAMIA 1997;4:364-375

44

Proportion of doses exceeding recommended maximum

2.10%

0.60%

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

Paper-based Computer ordering

Teich Archives Int Med 2000;160:2741

45

Dosing appropriateness in patients with renal impairment

67%

54%59%

35%

0%10%20%30%40%50%60%70%

Doseappropriate

Frequencyappropriate

Intervention Control

Both resultsP < 0.001

Chertow JAMA 2001;286:2839-44

46

Effect of an antibiotic advisor

35

146

0

20

40

60

80

100

120

140

160

Orders for drugs to whichpatient is allergic

87

405

0

100

200

300

400

500

Excess drug dosages

12

206

0

50

100

150

200

250

Antibiotic-susceptibilitymismatches

4

28

0

5

10

15

20

25

30

ADEs caused by antiinfectiveagent

Intervention Control

All results statistically significantCosts, LOS also reduced

Evans NEJM 1998;338:232-238

Effect of Changing Default Dosing Frequency for Ceftriaxone

0

10

20

30

40

50

60

Week

Ord

ers/

wee

k BIDQD

48

Computer suggests300 mg every 8 hours for this patient; based

on age and weight

NYPH -- Pediatric dosing decision support

Evaluation•32% of suggestions accepted exactly•Good reasons for not following•Subjectively, physicians like it

(Killela B, et al; Pediatrics, 2007)

Results of Two Studies on Medication Error Prevention

CPOE reduced medication errors by 80% CPOE reduced serious medication errors by

55%

Bates et al, JAMA 1998

50

Evidence of Reduction in Errors

CPOE Pittsburgh Children’s Harmful ADEs pre-CPOE 0.05/1000 doses Post-CPOE 0.03/1000 doses (p=0.05). CPOE prevents 1 ADE every 64 patient days

Upperman et al. J Pediatr Surg 2005;40:57-59

51

Evidence of Reduction in Errors

Mullett Stand alone anti-infective CDSS in PICU 59% decrease in the rate of pharmacy

interventions for wrong drug doses Potts

CPOE and medication ordering errors in PICU Medication errors reduced 96%

Near misses reduced 41%

Mullett CJ, et al. Pediatrics 2001;108:75-81

Potts AL, et al. Pediatrics 2004;113:59-63

52

So what’s the problem?

High federal and state policy interestHigh federal and state policy interest Leapfrog group identified as 1 of 3 patient safety Leapfrog group identified as 1 of 3 patient safety

‘leaps’‘leaps’ Only 10-15% of hospitals across the country have Only 10-15% of hospitals across the country have

active CPOE systemsactive CPOE systems Few primary care physicians have CPOEFew primary care physicians have CPOE High stakesHigh stakes

Enormous institutional investmentEnormous institutional investment Well-publicized ‘failures’Well-publicized ‘failures’

53

Unintended Consequences

““I have ordered the test that was right next I have ordered the test that was right next to the one I thought I ordered, you know, to the one I thought I ordered, you know, right below it that little thingie had come right below it that little thingie had come down and I clicked and I’m lookin’ at this down and I clicked and I’m lookin’ at this one but I in fact clicked on the thing before. one but I in fact clicked on the thing before. By that time I turned my head and I’m By that time I turned my head and I’m hitting return and typing my signature and hitting return and typing my signature and not seeing it.”not seeing it.”

Ash, JAMIA 2005

54

$$$,$$$,$$$ High cost and lack of capital

“The number one [barrier] is cost. I have been doing hospital software for 29 years, and this is the most expensive project I’ve ever done.”

Cost benefit mis-match “[CPOE] may save a lot of money [for] the health care

system overall, but [the money] is not being collected by the hospital.”

For ambulatory CPOE, assuming 11.6% capitation rate nationally 11% of ROI accrues to providers 89% to other stakeholders, primarily payers

55

Errors in AdministrationIV Infusions

Rule of 6’s intended for rapid dose calculation and drug preparation If:

conc of drug (mL/100 ml) = 6 x patient weight (kg) Then:

dose (µg/kg per min) = rate (mL/hour) Calculation intensive Numerous possible concentrations

If use standard drug concentrations, then rely on dosing charts

56

Errors in AdministrationIV Infusions

IV syringe pumps Traditional infusion pumps prone to ADEs Errors due to keypad data entry mistakes Free-flow phenomena

57

Smart Pumps

Library of medications with standard concentrations specific to patient population

Makes calculations

“Safety net”

Hard limits (range cannot be overridden)

Soft limits (can be overridden)

Pump alarms and halts infusion if medication dose is programmed outside preset limits

58

0

0.5

1

1.5

2

2.5

3

3.5

Reported Errors

PreinterventionErrors, per 1000

doses

PostinterventionErrors, per 1000

doses

Larsen et al. Pediatrics 2005

*

*p<0.001

Smart Pump with Standard Concentrations and Redesigned Labels

59

Safe Practices for Communicating Test Results

Critical ambulatory safety issue 75% of physicians did not notify patients

of normal results 33% of physicians did not even notify of

abnormal results (Boohaker et al, Archives 1996)

Approximately 1/3 of women with abnormal mammograms or pap smears do not receive appropriate follow-up care

60

Improving Result Management Systems

Can be integrated with Electronic Medical Records A tool that allows focus on truly abnormal test

results A tool that warns physicians if patients have missed

tests Use of standardized features, such as “ticklers” Paper systems can also be highly successful Standardized procedures rather than every physician

doing it his/her own way

61

Results Manager Application

Strategy: Automate CarefullyStrategy: Automate Carefully

EHREHR CPOECPOE Smart IV systemsSmart IV systems RobotsRobots Automated drug delivery systemsAutomated drug delivery systems Bar codingBar coding

Inexpensive technologyInexpensive technology 63% reduction in serious dispensing errors 63% reduction in serious dispensing errors

at BWHat BWH

63

Conclusions

Make safety top priority Move from a culture of blame to one of safety

Stop blaming people for making errors To err is human

Improve the system Utilize technology as it becomes available Involve patients/parents to the fullest extent Measure and iteratively refine