Interactive Patient Computer Dialogue

1
Interactive Patient Computer Dialogue
2
Hypothesis

• The best way to obtain comprehensive historical
  data from a patient in an efficient cost
  effective manner is to have a patient enter
  information directly into a computer

3
Types of Health Care Outcomes
PRIORITY

• Physical
• e.g., patients perceptions of benefit, change in
  functional status, complication rate
• Service
• e.g., consumer satisfaction, access, waiting
  times, bedside manner
• Cost
• e.g., direct cost, out-of-pocket cost, indirect
  cost

Health Professional
Patient
Payor
1
1
1
So WHICH outcome(s) is/are the most important?
4
The price ofprofessional autonomy is
Accountability for all three outcomes
Service Outcomes
Cost Outcomes
Physical Outcomes
5
The IRON Triangle
Access
Quality
Cost
Optimizing among the three pillars of clinical
care is a fundamental and career-long process.
6
Patient Computer Dialogue will improve all three
7
(No Transcript)
8
Traditional Methods of History Taking
9
How do we do?

• Physicians miss 54 of patients problems and 45
  of their concerns
• In 50 of visits patients and doctors do not
  agree on the presented problem
• 50 of psychological problems are missed
• Only 23 seconds before patient is interrupted

10
Study with standardized patient

• 134 physicians
• Nuts and bolts exams 17 cases and 20 minutes(21
  of time used full)
• 14-16 patients on a Saturday
• Realism 2.3 (1 best 10 least)
• 59 Essential Questions were covered
• 15/17 cases 100 by one doctor

11
You are at the office

• A partners patient who has hypertension comes to
  see you because in the last ten days she has
  noted that her blood pressure is elevated from
  its baseline.
• Meds Lisinopril 20 mg daily
• BP 152/93

12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
What did we learn

• We can not judge quality of care by reviewing a
  chart!
• Inputs are important
• Computerized history provided more information
  that was critical to this case, and was valuable
  to the clinician

16
We should all have patients answer questions on
computers to enhance their quality

• Improve quality of history
• Improved access
• Cost savings

17
Whoa there partner

• Confronted with a new idea what is our reaction
• It is negative-Limbic System
• Read your minds
• My patients could not do this
• It is so impersonal
• I do a much better history then any computer
• That is just too much information to deal with
• These Yanks-why did they ever leave the mother
  country

18
Evolved-Cerebral Cortex

• Concerns you have
• Possibilities
• Overcome the potential obstacles

19
How does it work?
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
How many of your patients can do this?
24
Your patients can not do this

• Yes-ATM machines
• Your mind 25-75 of your patients
• Literature says Expect 83-95
• Interpreters?
• People who come to the office with elderly
• Elderly more careful

25
Another Concern How would they do it?

• Office waiting room-kiosk
• At home on the internet

26
Unstructured E-mail
Dear Dr. Sample, My back hurts. I am taking my
arthritis medication like you told me every night
but the pills do not seem to help my back pain.
The pain is so bad that I toss and turn all
night. The pain in my back started after I
picked up a heavy box at work day before
yesterday. It is bad when I go down stairs, but
not when I go up stairs. The pain has started to
go down my leg. Do you think I strained my back?
27
CHIEF COMPLAINT Test Case is a 54 year old male.
His reason for visit is "back pain". HISTORY OF
PRESENT ILLNESS 1. "BACK PAIN" He reported
Initial back pain evaluation. LOCATION He
reported Pain in the lower back. Lumbosacral
pain distribution only. He denied Upper back
pain. Mid back pain. Sacral back pain.
Paravertebral back pain. SEVERITY He denied
Crescendo pattern of back pain. DURATION He
reported Back pain 1 to 2 days. TIMING He
reported Back pain continuous. Acute onset back
pain. No prior episodes of similar back pain. He
denied Back pain pattern different from initial
onset. CONTEXT He reported Back pain preceded by
lifting, carrying, or bending. MODIFYING
FACTORS He reported Back pain changed by
bending, lifting, carrying and ascending or
descending stairs. Back pain changes with
movement. He denied Back pain changed by
sitting or valsalva, walking or standing, and
lying down or stretching. Back pain worsened by
valsalva (cough or sneeze and relief from
conservative measures.
28
Patients will not like it

• Very efficient
• Patients 90 satisfaction
• 10 dislike
• 10 like more then their doctor

29
It is impersonal

• Of course it is It is a machine
• There are some advantages to being impersonal

30
Better at Socially Sensitive Information
31
Better at Socially Sensitive Information
32
This will replace me!

• What is more important to agriculture a tractor
  or a farmer?
• This is an aid
• Your histories are more accurate
• You can zero in on things better
• 40 important information
• Pilot Check List

33
How dangerous is health care?
Note both dimensions are logarithmic scales
REGULATED
DANGEROUS
ULTRA-SAFE
(gt1/1000)
(lt1/100K)
Driving
Scheduled
Deaths per year
Airlines
Chartered
Flights
European
Mountain
Railroads
Climbing
Nuclear
Manufacturing
Bungee
Power
Jumping
10
Number of encounters per death
34
Highly Adaptable

• SARS

35
(No Transcript)
36
(No Transcript)
37
Others

• Patient controls interview-length of time
• Doctor only deals with positives
• Research-Structured Data
• Multimedia
• High patient acceptance
• Scales

38
(No Transcript)
39
Scales

• Zung Depression
• Rahe Stress Scale
• Urology Scales
• Wast
• Conners
• Anxiety
• Pain impairment
• Patient Education Needs

40
Jane Doe Chief Complaint Sore throat VS
stable Sore throat 3 days duration
41
(No Transcript)
42
(No Transcript)
43
Ok How would I get started

• Interviewing Software
• At least 50,000 questions (False Positives will
  be less)
• Literature supporting the software
• Integrates with an electronic medical record
• Inexpensive
• Specialty specific
• Ability to generate patient friendly information

44
IMH

• Primetime Practice
• www.medicalhistory.com
• Over 80,000 questions
• Only vendor that has EMR integration 42 in US and
  e-mail
• Inexpensive

45
Resources

• On the Internet Mayo Clinic Proceedings
• Check out www.medicalhistory.com
• Process-15 inventive early adapter personality
• Follow me around
• Look for lectures.

46
Summary

• The best way to obtain comprehensive historical
  data from a patient in an efficient cost
  effective manner is to have a patient enter
  information directly into a computer.
• It will improve your quality of care for patients
  and will help you to be a better clinician
• Bachman.john_at_mayo.edu