Transcript lau_final_exam - UW Departments Web Server

Systems and Methods for
Patient-Centered e-Health Services
Chris Lau
Ph.D. Final Examination
August 14, 2003
1
Patient-Centered e-Health



Purpose

Improve productivity and efficiency in health care

Improve quality of care and patient outcomes
Current status

Content delivery—patient education, information about services offered by
clinics and medical centers

Appointment scheduling

E-mail with provider
Emerging applications—delivery of care

Personal health knowledge management systems (PHKMS)

Referrals

Home monitoring
2
Presentation Outline

Home monitoring with asynchronous video: a feasibility study

A platform for building e-health applications

Personal health knowledge management system applied to referral
management

Diabetic foot care monitoring

Conclusion
3
Telemedicine and Home Health Care


Home health care (nursing, rehabilitation, social services)

# of patients increased 54% from 1992-1994, 7.2 million in 1996

Costs increased 400% from 1992-1997
Telemedicine for home health

Goal: improve clinical outcomes and reduce costs (visits)

Home nursing for congestive heart failure, chronic obstructive pulmonary
disease, diabetes, cancer, wound care

Reduced cost in home nursing

Synchronous videoconferencing with custom equipment
(Johnston, 2000 and Dansky, 2001)
4
E-Medicine



Asynchronous web-based telemedicine
Combination of patient-provider messaging, web-accessible
medical records, and patient monitoring at home
Characteristics

Just-in-time access

Geographic neutrality and economy

Medical recording and service documentation
5
Feasibility Study

Total shoulder replacement arthroplasty

Helps restore comfort and function to shoulders damaged by
degenerative joint disease, osteoarthritis, or rheumatoid arthritis

Patients learn to do their own physical therapy and are discharged
about three days after surgery
6
E-Medicine for Post-Operative Monitoring

Patients can use the E-Medicine website to:

Review treatment plans

Send status reports to doctors with video

Send questions to doctors, optionally with video

File health status surveys to track how surgery is affecting everyday
activities
7
Patients’ Video Recording Interface
8
Physicians’ Status Report Review Page
9
Feasibility Study Conclusions


6 patients with a wide range of prior computing experience successfully
used the system
System useful for patients with questions (e.g., “Would it be okay if I…?”
or “Is it normal that…?”)

Interoperability with other systems desirable (e.g., EMRs)

Permanent storage area for patient-generated data needed

Next steps

Refine feasibility study system into a platform that can support e-health
applications

Implement new applications and evaluate their effectiveness
10
E-Medicine Platform Features

Secure messaging between patients and providers

Interface for implementing data collection instruments

Rapid development of new applications

Interoperability

HIPAA security requirements
Patients’ Homes
Patients
Internet
Home Healthcare Network Application
Service Provider
Hospitals, Clinics, and Doctors’ Homes
11
E-Medicine Architecture
SMTP
Mail Server
Data
Collection
Plug-in
HTTPS
Web
Server
(IIS)
HTTPS
EMedicineClientLib
Pipes
Applications
Pipes
ASP.NET
PubCookie
DHTML
User
Interface
ASP.NET
Web Browser
EMedicineLib
TCP/IP
Database
(SQL Server)
Cookies
.NET Common Language Runtime
Application Server
(Microsoft .NET)
Other Clients
SOAP
HTTPS
12
Personal Health Knowledge Management

Personal health knowledge management systems (KMS)



Web-based patient-owned repository for all of an individual’s medical
information
Problems addressed

Patients’ medical records scattered across points of care

No place in traditional EMR to store patient-generated information
Prior work

Implementations focused on use in emergency and travel

Evaluations compared personal health KMS features against EMR features
(Schneider, 2001 and Kim, 2002)
13
PHIMS

Personal Health Information Management System
Patients
Primary Care
Provider
Hospitals, Clinics,
and ER
Specialists and
Other Clinicians
Home & Mobile
Physiological
Sensors
EMRs
PHIMS
Medical records, X-rays,
glucometer readings,
vaccinations, allergies,
medications, genome
Laboratory Systems
Imaging Systems
Multimedia Digital
Library
14
FARMS


Referral—request to see a specialist for a problem beyond the
experience of patient’s current provider
Prior electronic referral systems

Enable providers to request referrals with patient information in EMR
(Sittig,
1999 and Murphy, 1998)

Facilitated Accurate Referral Management System (FARMS) linked with
PHIMS

Contains information not just from a single provider

Allows patients to request referrals electronically

Patients can maintain a permanent record of their consultations
15
Referral Procedure with FARMS
Patient establishes
PHIMS account
Patient provides
medical history and
problem information
FARMS checks for
records required to
make referral request
FARMS directs
referral to
appropriate provider
Provider evaluates
referral request
Schedule
consultation?
No
Yes
Appointment
Suggest alternative
treatment and/or
redirect to another provider
16
System Architecture
SMTP
Mail Server
HTTPS
Client
HTTPS
Client
Web
Server
(IIS)
Pipes
PHIMS/FARMS User Interface
FARMSLib
Pipes
ASP.NET
PubCookie
HTTPS
ASP.NET
Client
EMedicineLib
TCP/IP
Database
(SQL Server)
Cookies
.NET Common Language Runtime
…
Application Server
(Microsoft .NET)
17
Study Overview


Deployed into an orthopaedics surgical clinic to evaluate usability and
clinical utility of PHIMS and FARMS for managing self-referrals
Procedure

Patients recruited through UW Dept. of Orthopaedics web site

Patients fill out essential parts of PHIMS for referral request and share with
clinic staff (PCC—patient care coordinator)

Patients asked to complete
rest of PHIMS before the
appointment

Specialist reviews patients’
information in PHIMS before
and/or during appointment

Patients asked to fill out survey
2 weeks after initial contact
18
Dept. of Orthopaedics Arthritis Source
19
PHIMS Health Problem Entry
20
Referral Review
21
Referral Management
22
PHIMS Message
23
PHIMS Allergies
24
Subject Demographics

Study period: December 2002-July 2003

61 patients, 5 specialists

10 times more visitors to Age (Mean (SD))
study recruiting web page
Location (No. (%))
West
Other
Clinic
Study
51.79 (16.79) 45.70 (12.93)
2691 (97.54)
68 (2.46)
54 (88.52)
7 (11.48)
p-value
0.00461
0.0000365
41
2
1
2
1
1
2
1
1
1
1
3
1
3 patients outside of the US
25
Usage
Referrals
PCC
messages
Patient
messages
17-Dec
6-Jan
10-Mar
30-Mar
19-Apr
9-May
29-May
18-Jun
8-Jul
28-Jul
Date
26
Survey Response
Strongly Agree = 1
Agree = 2
Avg. Response: 1.44
2.56
Somew hat Agree = 3
2.44
1.81
Somew hat Disagree = 4
2.09
3.74
Disagree = 5
2.19
Strongly Disagree = 6
2.00
2.06
20
18
Number of Responses
16
14
12
10
8
6
4
2
he
al
th
in
ng
4.
F
il li
ng
il li
3.
F
ea
re
sy
fe
rra
lr
5.
eq
V
ue
ie
w
st
of
ea
he
sy
al
th
re
co
6.
rd
Le
he
ar
lp
ne
fu
d
l
m
or
e
ab
7.
ou
C
om
th
ea
fo
rta
lth
bl
e
co
m
8.
m
un
R
es
ic
at
po
in
ns
g
e
t
9.
im
e
S
at
ac
is
ce
fie
pt
d
ab
wi
le
th
re
fe
rr
al
pr
oc
es
s
re
co
rd
ea
sy
in
cc
ou
nt
2.
A
1.
Lo
gg
in
g
in
w
se
tu
p
as
ea
sy
0
Question
32 respondents, 85% satisfied with usability, 94% with online referral experience
27
Patient Comments
Topic
Enhanced quality of service
Online forms preferred vs. paper forms
Convenience of messaging
Technical suggestions
System-related service complaint
Other service complaint
# of Comments
10
5
3
6
3
3
28
Provider Comments



Physician acceptance

Physicians satisfied with completeness of PHIMS and found it to improve the
usage of time during visits

Usefulness of patient-entered information

PHIMS data gives physicians a unique baseline assessment of patients in
their own words

Nowhere else is this type of information captured for later comparison
on effectiveness of treatment
PCC acceptance

PCCs felt messaging system resolved the problem of telephone tag

Workflow-oriented design of interface allows them to prioritize patients and
efficiently respond to each request
Physician request: image and video uploads
29
Image and Video Uploads
30
Conclusion


Patient health knowledge management system built on top of the EMedicine platform

Applied to referral management in a surgical clinic

Improved efficiency for patients, PCCs, and physicians

System found to be user friendly, convenient, and able to facilitate patientprovider communications

Extensible to include surveys, photos, videos, and other data types
Limitation

Work on standards and universal identifiers for all parties needed to enable
more automatic consolidation of patients’ medical records
31
Diabetes Foot Care Monitoring


Background: peripheral neuropathy

Loss of pressure and pain sensations

Affects >30% of diabetics

Most common complication: foot ulcers in 5-10% of diabetics, up to 3% will
require lower limb amputation

Diagnosis requires physical exam, but general foot care can be monitored
using digital photographs
Monitoring with E-Medicine

Digital camera photo upload to E-Medicine using web browser plug-in

Image quality evaluated at a diabetes clinic

Home test conducted with patient from diabetes self-management project
32
Integration with Diabetes Self Management


Patients issued a UW NetID and SecurID hardware token
Pubcookie single sign-on authentication allows users to access any
application inside the box
Pubcookie
MyUW
MIND
WebPine
Diabetes
Monitoring
E-Medicine
Patient
Education
33
Diabetes Foot Care Application
SMTP
Mail Server
Digital Photo
Plug-in
HTTPS
Web
Server
(IIS)
HTTPS
EMedicineClientLib
Pipes
Diabetic Foot Care
Pipes
ASP.NET
PubCookie
DHTML
User
Interface
ASP.NET
Web Browser
EMedicineLib
TCP/IP
Database
(SQL Server)
Cookies
.NET Common Language Runtime
Application Server
(Microsoft .NET)
HTTPS
MyUW
HTTPS
34
Foot Examination User Interface (1/3)
35
Foot Examination User Interface (2/3)
36
Foot Examination User Interface (3/3)
37
Results

Photographs examined by endocrinologist specializing in
diabetes care

Calluses, fissures, corns, warts, ulcers, and fungus
38
Fungus
39
Summary
Patient requests
referral
Clinic and
patient
correspond
Appointment(s)
Intervention follow-up
Orthopaedic
surgery
E-Medicine
Chronic
disease
management
Monitoring and intervention follow-up
Time
40
Unified E-Medicine Architecture
SMTP
Mail Server
Data
Collection
Plug-in
HTTPS
Web
Server
(IIS)
HTTPS
EMedicineClientLib
Pipes
PHIMS
&
FARMS
Diabetes
Shoulder
Pipes
ASP.NET
PubCookie
DHTML
User
Interface
ASP.NET
Web Browser
EMedicineLib
TCP/IP
Database
(SQL Server)
Cookies
.NET Common Language Runtime
Application Server
(Microsoft .NET)
Other Clients
SOAP
HTTPS
41
Conclusion


Low cost asynchronous telemedicine system implemented using
commercial off-the-shelf components

Utility shown in post-operative monitoring

Both patients and surgeons appreciated the ability to communicate between
clinic visits in a mode richer and more convenient than telephone
Post-operative monitoring application transformed into a modular
platform for building e-health applications


Personal health knowledge management system

Applied to referral management

Patients and providers reported enhanced quality of communications

Unique storage area where patients can record health status in their
own words
Home monitoring of diabetic foot care
42
Future Directions


Clinical

Outcome measurements for diabetic foot care

Long-term effects of PHIMS on quality of care
Technical

Use of PHIMS in a public key infrastructure

Patient-targeted decision support tools built on PHIMS
43
Acknowledgements


Committee: Yongmin Kim, Rick Matsen, Bill Lober, Ceci Giachelli,
Buddy Ratner, Scott Eberhardt
E-Medicine: Hugh Chang, Maisie Wang, Justine Liu, Janice Kim,
Mark Wilson, Judy Gattinella

Clinical:
Sean Churchill, Irl Hirsch, Harold Goldberg, Seth Leopold,
Paulette Gayton, Suzi Hughes

PETTT:
Scott Macklin, Aaron Louie, Kristen Shuyler

ICSL:
All current and former members and in particular
Sayan Pathak, Jim Cabral, Niko Pagoulatos, Ravi Managuli,
Chris Chung, Eung-Hun Kim
44