Transcript Inappropriate clopidogrel adherence explains stent related adverse

Inappropriate clopidogrel
adherence explains stent
related adverse outcomes
Leonardo Tamariz, MD, MPH
University of Miami
Antiplatelet use post-stent
• Angioscopic studies show lack of neointimal coverage and thrombi post-stent.
• Drug eluting stents have been associated
to late MI.
• ACC/AHA/SCAI recommends in patients
with low bleeding risk maintaining
clopidogrel for 12 months after a stent.
• Clopidogrel is now a chronic medication.
Consequences of lack of
medication adherence in CV
disease
• Inadequate disease control.
• Low adherence to beta-blockers or statins
in post MI patients increases the death
rate.
• Represents a significant burden to
healthcare utilization – the estimated
yearly cost is $396 to $792 million.
• One – two thirds of all medication-related
hospital admissions are attributed to
nonadherence.
Specific aims
• To evaluate clopidogrel medication
adherence in a cohort of patients
with stents.
• To evaluate if clopidogrel adherence
affects stent related outcomes.
Methods: Data source
Humana EDW
Medical File
Member File
ICD 9 codes
Demographics
Costs
Provider information
Pharmacy File
GPI codes
Dosage
Refill patterns
Methods: Study design
Clopidogrel adherence
Stents
(n=7,091)
Clopidogrel users
(n=5,838)
MI
Death
3.2 ± 0.7 years
Non- clopidogrel users
(n=1,253)
Inclusion criteria
•18 years or older
• Procedure claim for bare metal stent (36.06) or drug eluting stent (36.07)
between January 1, 2003-June 1, 2005.
Methods: Clopidogrel adherence
• Clopidogrel use defined by Generic
Product Identifiers (GPI code:85158020)
• Adherence defined by medication
possesion ratios (MPR).
– Appropriate adherence: 80% or more MPR
– Innapropriate adherence: <80% MPR.
∑ days supply of medication
MPR =
∑ number of days between first and last refill + days supply
of last refill
Methods: Outcomes
• Myocardial infarction
– ICD 9 code 410.x with a hospitalization
• Death (all cause mortality)
– Social security death index match
• Combined
– MI or death
Methods: CV risk factors
(ICD-9 definitions)
• Diabetes (250.xx)
• Hypertension (401.xx, 402.xx,
403.xx,404.xx,405.xx)
• Abnormal lipid panel (272.xx),
• Obesity (278, 278.0, 278.00,
278.01, 278.1).
Methods: Statistical analysis
• Baseline characteristics with chi-square and t-test
• Predictors of innapropriate use with logistic
regression
• Person-time and hazard ratios of events using
Cox proportional adjusted for demographics,
claims for CV risk factors, claims for heart failure,
type of stent and MI at presentation.
Results: Baseline characteristics
Entire
cohort
(n=5,838)
Clopidogrel
users with
appropriate
adherence
(n=4,548)
Clopidogrel users
with
inappropriate
adherence
(n=1,290)
63.311.6
63.511.5
62.711.9
0.04
Female gender, %
33
24
21
<0.01
Previous history of
myocardial
infarction, %
18
18
18
0.92
Previous history of
hypertension, %
55
54
57
0.18
Previous history of
diabetes, %
28
27
32
<0.01
Previous history of
abnormal lipids, %
48
48
48
0.95
Previous history of
heart failure, %
10
9
11
0.01
Medicare, %
58
55
59
0.08
Drug eluting stents,
%
84
84
85
0.49
Characteristic
Age, years
p-value
20
10
0
Percent
30
40
Results: Distribution of MPRs
0
.2
.4
.6
Medication possesion ratio
.8
1
Results: Predictors of inappropriate
clopidogrel use
Predictors
OR (95% C.I)
Age
0.97 (0.97-0.98)
Female gender
0.79 (0.69-0.91)
Diabetes
1.2 (1.0-1.4)
Medicare insurance
1.5 (1.26-1.81)
0.10
Results: Incidence of MI by
adherence to clopidogrel
0.02
0.04
0.06
0.08
HR 1.35(1.08-1.70)p=0.009
Appropriate clopidogrel adherence
Innappropriate clopidogrel adherence
0
1
2
Years
3
Results: Incidence of death by
adherence to clopidogrel
0.40
Appropriate clopidogrel adherence
Innappropriate clopidogrel adherence
0.00
0.10
0.20
0.30
HR 1.32(1.12-1.55)p<0.01
0
1
2
Years
3
0.15
Results: Incidence of combined
outcome by adherence to
clopidogrel
Appropriate clopidogrel adherence
Innappropriate clopidogrel adherence
0.00
0.05
0.10
HR 1.31(1.11-1.54)p<0.01
0
1
2
Years
3
Conclusions
• Twenty eight percent of clopidogrel
users post-stent deployment are not
adherent.
• Lack of adherence to clopidogrel
post-stent increases the risk of
myocardial infarction and death.
Limitations
• Lack of validation of exposure and
outcomes.
• Insured patient population, less
burden of disease compared to other
studies.
• Unable to document ASA use,
coronary anatomy where stent was
placed due to use of administrative
claims.
Clinical implications and
unresolved issues
• We need to stress the importance of
clopidogrel adherence after a stent.
• We need to inquire patients about
clopidogrel adherence post-stent.
• We need to identify individual reasons for
lack of clopidogrel adherence.
• Need a randomized trial to determine the
efficacy of different interventions to
improve clopidogrel adherence.