Improving the Outcomes of
Oral Anticoagulation:
Home Monitoring of Warfarin Therapy
Jack Ansell, M.D.
Lenox Hill Hospital, NY
September 22, 2009
Disclosures
Consultant: Roche Diagnostics, ITC, HemoSense
The Dilemma of
Anticoagulation Management
• Warfarin has a narrow therapeutic window
of effectiveness and safety.
• Many factors influence a patient’s stability
within that window.
• Frequent monitoring is required to maintain
patients in the therapeutic window.
• Monitoring is labor intensive and complex.
•
•
Consequences
Increased adverse events with poor management
Physicians avoid warfarin use because of its complexity.
ORAL ANTICOAGULATION IS ALL ABOUT MANAGEMENT
The Desired Outcome: Benefits must be greater than Risks
Hospital Transition to Outpatient
Outpatient
Hemorrhage
Risks1,2
INR
Reduction in risk of stroke or
venous thromboembolism (VTE)
Benefits2,3
1. Ansell J et al. Chest. 2004;126:204S-233S. 2. Hirsh J et al. J Am Coll Cardiol. 2003;41:1633-1652.
3. Rothberg MB et al. Ann Intern Med. 2005;143:241-250.
Transition from Hospital to
Ambulatory Care Settings
4
How well does a University Hospital do in
managing warfarin therapy?
“Inpatient Warfarin Medication Utilization Evaluation”
Treatment decisions involving inappropriate assessment of response
• 349 records reviewed and assessed by established criteria
• 647/2030 (31.8%) warfarin treatment decisions were deemed inappropriate
Total = 647 decisions
8%
Initial dose too high (52 decisions)
Initial dose too low (9 decisions)
1%
10%
Different dose from home therapy
(63 decisions)
1%
1%
Continued home dose but should
have been changed (6 decisions)
10%
Continued home dose but should
have been held (4 decisions)
Held dose when therapy should
have been restarted (66 decisions)
69%
PK/PD not taken into account
(447 decisions)
5
Is the correct starting dose used?
“Inpatient Warfarin Medication Utilization Evaluation”
35.3% (123/349) patients were initiated on warfarin in-house
for the first time
New starts (n=123)
51%
Dose ok
Dose too
high
7%
42%
Dose too low
6
What is the impact on outcomes?
“Inpatient Warfarin Medication Utilization Evaluation”
New starts for VTE indication (n=47)
Initial Dose
Inappropriate (n=23)
Bleeding events
Bleeding risk
distribution
Vitamin K use
4 documented bleeds
3 transfusions ≥ 2 u
5 > 2 g/dL decr. Hgb
15 high risk
6 moderate
2 low
4 pts
Initial Dose
Appropriate (n=24)
2 documented bleeds
4 transfusions ≥ 2 u
5 > 2 g/dL decr. Hgb
11 high risk
8 moderate
5 low
1 pt
7
Models of Chronic Anticoagulation Management
 Routine Medical Care (Usual Care)
AC managed by physician or office staff w/o any systematic program for
education, follow-up, communication, and dose management. May use
POC device or laboratory INR
 Anticoagulation Clinic (ACC)
AC managed by dedicated personnel (MD, RN or pharmacist) with
systematic policies in place to manage and dose patients. May use
POC device or laboratory INR
 Patient Self-Testing (PST)
Patient uses POC monitor to measure INR at home. Dose managed by
UC or ACC
 Patient Self-Management (PSM)
Patient uses POC monitor to measure INR at home and manages own
AC dose
Challenges With Conventional
Laboratory Testing
• Patient issues
– Time for traveling to office or laboratory
– Ability to travel
– Need for venous access
• Labor-intensive and higher costs
– Scheduling visits
– Proper handling and delivery of sample
– Documentation at several time points
• Potential for communication delays
– Laboratory to contact provider with results
– Provider to contact patient with dosage adjustments
Jacobson AK. In: Ansell JE, Oertel LB, Wittkowsky AK, eds. Managing Oral Anticoagulation Therapy. 2nd ed. St. Louis, Mo: Facts and
Comparisons; 2003;45:1-6.
Technology Advances:
Offers a new paradigm for monitoring since 1987
• Use of capillary whole blood1,2
– Allows fingerstick sampling2
– Appropriate for self-testing1
• Consistency of INR results1
• Portability1
– Can be done anywhere
• Simplicity1
– Patient can easily perform test
1. Leaning KE, Ansell JE. J Thromb Thrombolysis. 1996;3:377-383. 2. Ansell JE. In: Ansell JE, Oertel LB, Wittkowsky
AK, eds. Managing Oral Anticoagulation Therapy. 2nd ed. St. Louis, Mo: Facts and Comparisons; 2003;44:1-6.
What are the outcomes with Home
Monitoring?
Thromboembolism with PST or PSM vs Control
Heneghan et al. Lancet 2006;367:404
Thromboembolism with PST or PSM
PSM
Usual
Care
PST
AMS
Heneghan et al. Lancet 2006;367:404
Major Hemorrhage with PST and PSM vs Control
Heneghan et al. Lancet 2006;367:404
Author
Year
Intervention
#
Patients
TTR
(% or time in range)
Major
Hemorrhage
Thromboembolism
PST vs UC
Beyth
2000
PST/ams* vs UC
163 vs 162
56 vs 32 p<0.001
5.6% vs 12% p=0.049
8.6 % vs 13% p = 0.2
23 vs 24
93 vs 75 p=0.003
0
0
0 vs 1 event
0
PST vs AMS
White
1989
Kaatz
2001
Gadisseur
2003
THINRS
2009
PST/ams* vs AMS
PST/ams* vs AMS
PST/ams* vs AMS
63 vs 65 p=NS
52 vs 60
PST /ams vs AMS
63.9 vs 61.3 p=0.14
~68% vs 63% p = NS
PSM vs UC
Horstkotte
1998
Sawicki
1999
Fitzmaurice
2002
Kortke
2001
Sidhu
2001
Sunderji
2004
Voller
2005
PSM vs UC
75 vs 75
92.4 vs 58.8
PSM vs UC
83 vs 82
57 vs 33.8 p=0.006
1 event vs 1 event
1 event vs 2 events
PSM vs UC
23 vs 26
74 vs 77 p=NS
0 vs 1 event
0
PSM vs UC
305 vs 295
78.3 vs 60.5 p=<0.001
1.7 % vs 2.6% p=NS
1.2% vs 2.1% p=NS
PSM vs UC
34 vs 48
76.5 vs 63.8 p<0.0001
1 event vs 0
1 event vs 0
PSM vs UC
69 vs 70
71.8 vs 63.2
PSM vs UC
101 vs 101
67.8 vs 58.5 p=0.0061
2 events vs 0
0 vs 1 event
PSM vs AMS
49 vs 53
84.5 vs 73.8
1 event vs 0
1 event vs 0
PSM vs AMS
47 vs 52
66.3 vs 63.9 p=0.14
1 event vs 1 event
0
PSM vs AMS
40 vs 39
71.1 vs 70.4
PSM vs AMS
368 vs 369
58.6 vs 55.6 p=NS
4 events vs 7 events
4 events vs 20 events
PSM vs AMS
Watzke
2000
Gadisseur
2003
Khan
2004
MenendezJandula 2005
Improving AC Outcomes at the Time of Discharge
128 patients randomized to home POC monitoring (n= 60) or UC (n=68)
after discharge. POC testing on d 2,4,6,8 vs UC on d 8
Discharge
Home
UC
Monitoring Monitoring
Sub-therapeutic
49%
47%
Therapeutic
42%
45%
Supra-therap
9%
8%
Day 8
Home
Monitoring
29%
67%
4%
UC
Monitoring
33%
41%
26%
p
value
<0.01
Adverse events up to day 90
Major Bleeding
Total Bleeding
Embolic Event
Readmit due
to AC
Complication
Death
Home Monitoring
2
15
9
3
7
(n=59)
Usual Care Monitoring
10
36
10
8
8
(n=68)
0.05
0.009
NS
0.32
NS
Jackson et al. J Intern Med 2004:256:137
Who is able to perform Home
Monitoring?
Considerations for Patient Selection
Willing to:
Learn and perform testing procedure
Keep accurate written records
Communicate results in timely fashion
Able to:
Participate in a training program to acquire
skills/competencies to perform self-testing
Generate an INR
Understand implications of test result
Maintain records
Reliable to:
Perform procedure with acceptable technique to obtain
accurate results
The THINRS Trial: Design
•
Purpose:
Compare HQACM with PST to HQACM alone on major
health outcomes
•
Patient population:
Atrial fibrillation or mechanical heart valve
•
Participating Centers
28 VA Med Ctrs with ACC of > 100 patients
• Two parts:
Part 1: Training and home testing for 2-4 weeks
Part 2: Competency assessment and, if capable,
randomization to HQACM every 4 weeks or PST
every week
Matchar. Amer J Med 2002;113:42-51
The THINRS Trial: Design
A key attribute
“everyone” was
those who were
randomized to
management
of this trial is that
trained for PST and
deemed capable, then
either PST or ACC
Matchar. Amer J Med 2002;113:42-51
The THINRS Trial: Intervention & Outcomes
Interventions:
•
HQACM (monthly INR)
Designated, trained staff person
Local standard management algorithm
•
PST (Weekly INR)
Interactive value response reporting system with
web-based local monitoring
Outcomes:
•
Primary
time to first major event (stroke, major bleed, death)
•
Secondary
time in range, satisfaction, quality of life
Matchar. Amer J Med 2002;113:42-51
The THINRS Trial: Participants
3,644 Trained
78 did not pass training
3,566 home with
meter for 2-4 weeks
508 dropped out
3,058 competency
assessment
136 did not pass
assessment or dropout
2,922 randomized
2,922 / 3,644 = 80% Passed Competency
Matchar. Amer J Med 2002;113:42-51
Summary from THINRS: Outcomes
• 80% of screened subjects demonstrated
PST competency and were randomized
– approx. 4 out of 5 pass
• Patients were less likely to pass PST, if
– Older, h/o CVA, poor cognition, low literacy,
poor manual dexterity
Matchar. Amer J Med 2002;113:42-51
Summary from THINRS:
Outcomes: Stroke, Bleed, Death
PST
HQDM
4,235
pt yrs
4,495
pt yrs
Event
Type
HQACM
Rate
per
pt-yr
PST
Rate
per
pt-yr
Total
Rate
per
pt-yr
Stroke
32
0.76%
31
0.69%
63
0.72%
Major
Bleed
189
4.46%
173
3.85%
362
4.15%
Death
157
3.71%
152
3.38%
309
3.54%
Total
378
8.93%
356
7.92%
734
8.41%
Summary from THINRS
• 80% of screened subjects demonstrated
PST competency and were randomized
– approx. 4 out of 5 pass
• Patients were less likely to pass PST, if
– Older, h/o CVA, poor cognition, low literacy,
poor manual dexterity
• Outcomes (TTR & AEs) were improved to
a small degree with PST
How Does Home Monitoring Achieve
Good Outcomes ?
• Access to testing
Frequency (convenience), timeliness
Greater Time-in-Range
• Consistency of testing
Instrument & thromboplastin
Consistent Results
• Awareness of test results
Knowledge, empowerment, compliance
Greater Time-in-Range
Managing Home Monitoring?
Trusting the INR Result
FS LT INR = .258 + .89 * FS RT INR; R^2 = .956
5.0
INR by left hand fingerstick
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
1.0
1.5
2.0
2.5
3.0
3.5
4.0
INR by right hand fingerstick
4.5
5.0
Trusting the INR Result
Thromboplastin — Reagent Combinations and
observed variation in INR
Ortho 1.00 BFA
DADE 1.03 BFA
Behring 1.08 BFA
Pacific Hem 1.20 BFA
IL Test 1.43 BFA
5.5
DADE 1.96 BFA
5
Ortho 1.00 ACL
4.5
4
DADE 1.03 ACL
Behring 1.08 ACL
Pacific Hem 1.20 ACL
IL Test 1.43 ACL
3.5
DADE 1.96 ACL
3
Ortho 1.00 MLA
2.5
2
DADE 1.03 MLA
Behring 1.08 MLA
Pacific Hem 1.20 MLA
IL Test 1.43 MLA
1.5
DADE 1.96 MLA
Courtesy A. Jacobson
Optimal Frequency of INR Monitoring*
Test Interval vs % In Range
100
90
% in Range
80
70
60
50
40
30
More Frequent
testing increases
% in range
20
10
0
0
7
14
21
28
35
42
49
Days Between Tests
Summary 18 published studies: PST Coalition Report, July 2000
Barriers to PST/PSM
• Lack of physician awareness or
acceptance1,2
• Fear it will lead to unintended
self-management3
• Implementation of PST/PSM3
• Reimbursement3
1. Jacobson AK. In: Ansell JE, Oertel LB, Wittkowsky AK, eds. Managing Oral Anticoagulation Therapy. 2nd ed. St. Louis, Mo: Facts and
Comparisons; 2003;45:1-6. 2. Roche Diagnostics. CoaguChek System: Why Use? Available at: http://www.coaguchek-usa.com/
information_for_professionals/why_use/content.html. Accessed May 12, 2006. 3. Wittkowsky AK et al. Pharmacotherapy. 2005;25:265-269.
Barriers to INR Patient Self-Testing (PST):
National Survey of Anticoagulation Practitioners1
Cost of Device Main Barrier
90
80
78.7
70
60.4
60
Survey
50
Respondents
40
(%)
30
35.7
20
10
0
Cost of device
Cost of
reagent
cartridges
Fear of PST
leading to PSM
Provider Barriers to PST
1. Wittkowsky AK et al. Pharmacotherapy. 2005;25:265-269.
Willingness to Pay for PST is low
• Few patients are willing to pay for self-testing,
despite the benefits of weekly testing.
• Those willing to pay for PST stated an average of
$18 per month as the acceptable out-of-pocket
expense for home testing with a POC device.
Amount Willing to Pay Out-ofPocket Per Month
Percent Respondents
n=71
$0
35%
$5 - $30
51%
$35 - $100
14%
Proprietary information
CMS did the right thing by approving
reimbursement, but they did it the wrong way
As of March 19, 2008 CMS expanded coverage
to patients with VTE and chronic AF
Medicare National Coverage Policy for
Home PT/INR Testing (as of July 2008)
Medicare will cover the use of home INR monitoring for chronic, oral
anticoagulation management for patients with mechanical heart valves
(non-porcine), chronic atrial fibrillation, or venous thromboembolism .
The monitor and the home testing must be prescribed by a treating
physician and all of the following must be met:
• Patient anticoagulated for at least 3 months
• Patient must undergo face-to-face educations program and
demonstrate correct use of device
• Patient continues to correctly use device
• Self-testing no more frequently than once per week
More information at:
http://www.cms.hhs.gov/MLNMattersArticles/downloads/MM6313.pdf
Oral Anticoagulation Patient Self-Testing: Consensus Guidelines for Practical
Implementation. Managed Care 2008;17(#10, Suppl 9):1-9
What is an IDTF ?
CMS defined a new entity independent of a hospital or
physician’s office in which diagnostic tests are
performed by licensed or certified non-physician
personnel under appropriate physician supervision.
This entity is called an Independent Diagnostic Testing
Facility (IDTF). The IDTF may be a fixed location, a
mobile entity, or an individual non-physician practitioner
and in all cases must comply with the applicable laws
of any state in which it operates.
IDTF’s… How They Work
Manages
Doctor
INR
Prof fee
$9/mon
Train fee
$191
Rx
INR
Patient
INR
CMS
IDTF
Inst
order
Inst
sent
Device
Manufacturer
Tech fee
$140/mon
Serv + Inst
Communication with patient doing home monitoring
Oral Anticoagulation Patient Self-Testing: Consensus Guidelines for Practical
Implementation. Managed Care 2008;17(#10, Suppl 9):1-9
PST dosed by internet expert system vs AMS
Criteria
AMS
Supervised PST
P - value
10.7 (+ 5.2)
41.7 (+ 6.6)
<0.001
19.6 days
4.6 days
<0.001
Time in range
58.6%
74%
<0.001
Extreme INRs
6%
1.7 %
<0.001
INRs < 1.5
45.9 %
33.3 %
<0.001
INRs > 5.0 (%)
54.1 %
66.7 %
0.006
# INRs /pt (mean)
Freq of testing (mean)
RCT (cross-over) of 162 patients, followed for 6 months; mean age
59 yr (16-91), 80% male with diverse indications.
Daily time to manage 80 patients 10-45 min (mean 23.2 min)
Ryan et al. J Thromb Haemost 2009;7:1284
Conclusions . . .
•
•
•
•
•
•
•
•
Anticoagulants (oral and parenteral) top the list for adverse events.
Management of warfarin therapy is often poor, even in the best of
circumstances.
The transition from inpatient to outpatient anticoagulation is a critical transition
that requires labor intensive systems and processes for successful
implementation.
POC INR technology can play an important role in facilitating such care.
Anticoagulation management models include Routine or Usual Care,
Anticoagulation Clinics, and PST/PSM (home monitoring)
Point-of-care (POC) provides an alternative to laboratory testing that is easy,
portable, and accurate and allows for testing either by physician or patient
POC home monitoring can be done either with physician management or
patient self-management
Home monitoring requires systems in place to implement and manage results.
IDTFs can perform much of the implementation and follow up tracking of results