Pediatric ePrescribing:

1
Pediatric e-Prescribing

Is the PDA a Prescription for Safety?
Kevin B. Johnson, MD, MS Department of Biomedical
Informatics Department of Pediatrics Vanderbilt
University
2
How Common?

• From 5.7 (pediatric inpatient) -11 (critical
  care)
• Dosage mis-calculations the most common
  contributor to medication prescribing errors in
  pediatric patients

JAMA. 2001 Apr 25285(16)2114-20. Pediatric
Critical Care Medicine. 5(2)124-132, March 2004.
3
How Big Is the Issue?

• 40 of medication errors due to prescribing
  (Hariharan, S, 2001)
• 16 illegible prescriptions
• Children are twice as likely to incur a
  prescription error (IOM, 2001)
• 70 of pediatric care is outpatient-based

(National Ambulatory Care Survey, 1998)
4
How Big Is the Issue?

• 70 of pediatric cases are seen in ambulatory
  settings

Represents 230 million ambulatory cases seen
(National Ambulatory Care Survey, 1998)
5
Theres A Little Dory In ALL Of Us
6
Error Prevention Strategies

• Use printed (typed) characters
• Add indication
• Use metric weights
• Do NOT use abbreviations
• Use a leading zero (0.5)
• Do NOT use a trailing zero (5.0)

American College of Physicians, June, 2002
7
Error Prevention Strategies

• We should comply with safe prescribing
  recommendations
• Show mg/kg dosing strategy
• Show childs weight
• Include childs age
• Show indication for the prescription
• Sign and date all prescriptions

Pediatrics 112(2)431
8
Paper-based Prescription Compliance

• Number of prescriptions with missing information
  3,096 / 7951 (39)

Johnson, KB et al, 2004
9
For children 40
30
40
Residents N 47
30
30
20
20
20
10
10
10
Mean 4.7
Mean 4.6
Mean 4.1
0
0
0
5.0
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
5.0
4.0
3.0
2.0
1.0
Weight
Dosing Strategy
Indication
12
7
10
Faculty n 12
10
6
8
5
8
6
4
6
3
4
4
2
2
2
1
Mean 4.83
Mean 4.67
Mean 4.3
0
0
0
5.00
4.00
3.00
2.00
1.00
5.00
4.00
3.00
2.00
1.00
5.0
4.0
3.0
2.0
1.0
measured on Likert scale 1 strongly disagree
5 strongly agree
10
For patients 16 years, it is important to
include
Residents N 47
Indication
Dosing Strategy
Weight
Faculty n 12
measured on Likert scale 1 strongly disagree
5 strongly agree
11
Why Not Always Prescribe Safely?

• Residents and faculty may not value all safe
  prescribing practices for all age groups.
• Education about safe prescribing practice will
  need to specifically address rationale for
  documenting weights, indications, and dosing
  guidelines.
• Faculty are not exempt from this education!

12
Why Consider PDAs?

• 17-20 of providers use PDA in practice (Harris
  Poll, 2002)
• 64 of medical facilities (represented by
  membership in AMDIS) use PDAs at an enterprise
  level (HIMSS, 2002)
• Prescription reference material the most common
  resource on physicians PDAs

13
PDAs Current Uses

• Daily schedules
• Reference tool
• Charge capture
• E-prescribing

14
Handheld Prescription Writing (HPW) Advantages

• Low startup infrastructure costs when using
  physician-owned devices
• Minimal training
• Integrated prescribing and decision support
• Easily integrated into workflow (geographically)
• May be integrated into workflow functionally

15
Potential HPW Challenges

• Robustness
• Integration into EMR
• Integration into workflow (including pharmacy!)
• Security
• Wireless capabilities
• Battery life
• Cost of properly configured HPW device\
• State regulations may impede use (EDI, scanned
  signatures)

16
The True Cost Of Enterprise PDAs

• Hot spares (20 failure rate in high-use
  environments
• Wireless (wi-fi)
• Extra battery power
• Spare batteries
• Chargers
• 1200 or more per device

17
PDA for Prescribing

• Settings primary care pediatric clinic and
  pediatric emergency department
• Timeframe 3/04 11/04
• 2 months using PDA, 2 months using paper
  alternating randomly from April through November
• Residents asked to complete survey in October

18
Survey Details

• Modified Questionnaire for User Interface
  Satisfaction (QUIS Chin, J. P., Diehl, V. A,
  Norman, K. (Sept. 1987))
• Questions added/modified specifically to address
  PDA workflow issues, use of stylus, specific
  domain functionality

19
Survey Process

• Distributed through mail, in clinic, in emergency
  department (77)
• Response rate 51/77 66

20
HPW Use

• When given a choice of PDA versus desktops,
• 17/51 used PDA routinely
• 34/51 did not use PDA

21
Overall Reaction to HPW
22
Overall Reaction to HPW

• Wonderful (38)
• Easy (36)
• Satisfying (23)
• Stimulating (23)
• Flexible (14)
• Fast (27)

• Selecting patients is easy (50)
• Logon is short (54)
• Faxing is easy (53)
• Faxing is safe (58)
• Good terminology (58)

percent percent with score between 6-9 on 0-9
point scale
23
Screen Design and Layout
24
HPW Screen Layout

• Easy to read (72)
• Stylus easy (88)
• Message position (72)
• Messages clear (83)
• Organization on screen clear (72)

• Sequence of steps clear (72)
• Computer keeps me informed of state (44)
• Error messages helpful (22)

N 28/77 percent percent with score between
6-9 on 0-9 point scale
25
Learning the PDA
26
Learning the PDA

• Easy (66)
• Exploring features is easy (60)
• Remembering commands (64)
• Straightforward process (60)

• Speed is fast enough (47)
• Good reliability (27)
• Correcting mistakes is easy (20)
• Experience makes process easier(36)

percent percent with score between 6-9 on 0-9
point scale
27
PDA Issues
28
Other Issues
29
Conclusions

• HPW not well accepted in the setting of widely
  available desktop workstations
• HPW user interface presents barriers to adoption
  of eRx
• Desktop workstation-based eRx may be more readily
  adopted, and offers more benefits for a similar
  cost

30
Implications

• In environments where desktop workstations are
  easily installed, these devices may be easier to
  adopt than existing PDA technology.

31
Thank you!

• AHRQ HS11868
• Coda Davison
• Lynette Forrest
• Mary Landrum
• PedsSTEP team!

32
The PedSTEP Project
33
Our Major Aims

• Assess the medication error rate and types of
  errors in ambulatory pediatric settings
• Identify the facilitators and barriers to
  implementing and successfully adopting handheld
  prescription writing (HPW) tools and e-Rx in
  general
• Evaluate the effect of e-Rx on the medication
  error rate and types of errors in ambulatory
  pediatric settings

34
Step 1 Baseline Data Collection
35
Prescription Compliance Summary

• Total number of prescriptions written 7,951
• Participants writing prescriptions
• Attendings 6
• First Years 18
• Second Years 13
• Third Years - 12
• Fellows, 1st year 2
• Fellows, 2nd year 1
• Fellows, undeclared years 2
• No position defined - 1

36
Hypothesis

• Resident behavior (not completing required
  prescription information) is due to
  misinformation about causes of error and uses of
  prescription data

37
Design

• Setting Primary care clinic in an urban academic
  medical center
• Subjects Pediatrics residents (74) and
  preceptors (14)
• Methods self-administered anonymous survey
• Data analysis primarily descriptive Kendalls
  Tau B and one-way ANOVA to assess effect of
  co-variates (gender, year of training, exposure
  to medication errors)

38
Results

• Surveys returned 48/74 (65)
• PGY1 18/25 (72)
• PGY2 14/25 (56)
• PGY3 14/24 (58)
• Preceptors 12/14 (86)
• Gender Female 42 (72)