Transcript Control - Northeast Iowa Family Practice Center

Team-based Care to Improve Blood
Pressure Control: The CAPTION Trial
Barry L. Carter, Pharm.D., FCCP, FAHA, FASH, FAPHA
The Patrick E. Keefe Professor in Pharmacy
Department of Pharmacy Practice and Science
College of Pharmacy and
Professor
Department of Family Medicine
Roy J. and Lucille A. Carver College of Medicine
University of Iowa
Objectives
See Agenda
Cluster, Randomized Efficacy Trial
OriginalPaper
A Cluster Randomized Trial to Evaluate
Physician/Pharmacist Collaboration to
Improve Blood Pressure Control
Barry L. Carter, PharmD; George R. Bergus, MD; Jeffrey D. Dawson, ScD;
Karen B. Farris, PhD; William R. Doucette, PhD; Elizabeth A.
Chrischilles, PhD; Arthur J. Hartz, MD, PhD
Funded by NHLBI:
RO1 HL69801
Journal of Clinical Hypertension 2008;10:260-271
Collaborative Management of
Hypertension Study: Efficacy Trial
•
•
•
•
•
Only faculty / private physicians involved in the study.
Patients 21-85 years with diagnosis of hypertension.
Baseline BP: 145-179 SBP or 95-109 DBP for uncomplicated.
•
135-179 SBP or 85-109 DBP for diabetes.
Clinic BP at 0, 2, 4, 6, 8, 9 months
24-hour BP at baseline and 9 months
Journal of Clinical Hypertension 2008;10:260-271
Physician/Pharmacist
Collaborative Management
Intervention
• Pharmacist conducted interview and assessed
patient for strategies to improve BP control.
• Pharmacist made recommendations to MD and
patient to improve BP control.
• Pharmacists and physicians worked to
overcome/prevent sub-optimal treatment, clinical
inertia, poor adherence, adverse reactions, drug
interactions
• Pharmacists saw patients at least every 2 months x
9 months.
NHLBI: RO1 HL69801
Data Analysis
• Continuous variables – likelihood-based mixed
models with random patient effects fit to SAS
Proc Mixed in an intention-to-treat analysis.
• Models adjusted for baseline BP, age, gender,
race, education, insurance status, household
income, marital status, smoking status, alcohol
intake, BMI, number of co-existing conditions,
baseline medication adherence and total
number of visits during the study.
Baseline Demographics
Control (n=78)
Age
BP meds
Baseline med
adherence
# co-existing DX
Diabetes
BMI (kg/m2)
* - p < 0.001
61.0 + 11.3
Intervention
(n=101)
59.6 + 13.7*
1.4 + 1.0
1.5 + 1.0
88.6%
71.1%*
0.46 + 0.78
0.47 + 0.81
24.4%
24.8%
31.8 (+14.7)
32.3 (+7.7)
Results
Intervention - CMH (n=101)
Control - CMH (n=78)
Systolic BP vs. Time
Intervention - post-study (n=64)
160
Control - post-study (n=39)
155
Sustainability
150
SBP mm Hg)
145
140
135
130
*
125
*
120
115
***
*
** **
**
***
***
***
110
0
2
4
6
8
10
12
14
16
18
20
Time in Months
* p<0.05, ** p<0.01, *** p<0.001,
Carter BL, Bergus GR, Dawson et al. Journal of Clinical Hypertension 2008;10:260-271.
Carter BL, Doucette WR. Franciscus CL, et al. Pharmacotherapy 2010;30:228-235.
22
24
26
*
Results: BP Control Rates
Main Finding: The major reason for the high control
was due to intensification of medications.
Control
Interven- Adjusted
tion
OR
CI;
p value
All
patients
52.9%
89.1%
8.9
3.8-20.7
P<0.001
Diabetes
23.5%
81.8%
40.1
4.1-394.7
P=0.002
- Carter BL, Bergus GR, Dawson et al. Journal of Clinical
Hypertension 2008;10:260-271.
- Von Muenster SJ, et al. Pharmacy World & Science
2008:30:128-135.
Physicians accepted 95.8% of 267 pharmacist recommendations
Recommendation
Frequency by Visit
0 Mo
Opt
Added Thiazide n=45
40
2
3
0
0
0
NA
Added Other Drug n=79
30
13
18
9
6
3
NA
Increased Dose n=89
28
21
14
9
9
8
NA
Changed Dose Frequency n=7
2
0
1
3
1
0
NA
Switch Within Class n=15
6
3
1
3
2
0
NA
Decreased Dose n=14
3
3
3
2
3
0
NA
Drug Discontinued n=18
2
4
8
3
1
0
NA
111
46
48
29
22
11
NA
0
-
Total n=267
BP Control Rate n=101
2 Mo 4 Mo 6 Mo 8 Mo 9 Mo
52% 67% 73% 84%
89%
• Von Muenster SJ, Carter BL, Weber CA et al. Description of pharmacist
interventions during physician-pharmacist co-management of hypertension.
Pharmacy World & Science 2008:30:128-135.
“Mixed” Efficacy-Effectiveness trial
ORIGINAL INVESTIGATION
HEALTH CARE REFORM
Physician and Pharmacist Collaboration
to Improve Blood Pressure Control
Barry L. Carter, PharmD; Gail Ardery, PhD; Jeffrey D. Dawson, ScD; Paul A.
James, MD; George R. Bergus, MD; William R. Doucette, PhD; Elizabeth A.
Chrischilles, PhD; Carrie L. Franciscus, MA; Yinghui Xu, MS
Trial Registration: clinicaltrials.gov Identifier:
NCT00201019
Arch Intern Med. 2009;169(21):1996-2002
Adherence Study: Combination of
Efficacy and Effectiveness
• Prospective, cluster-randomized controlled trial in
6 community-based family medicine residency
clinics all with clinical pharmacist faculty in the
medical office.
• Research nurse in each clinic measured BP at
baseline, 3 and 6 months and 24-hour BP at
baseline and 6 months.
Intervention
• Pharmacist conducted interview and assessed
patient for strategies to improve BP control.
• Pharmacist made recommendations to MD and
patient to improve BP control.
• Pharmacists and physicians worked to
overcome/prevent sub-optimal treatment, clinical
inertia, poor adherence
• Pharmacists only encouraged to see patients at
baseline and 1 month with a telephone call at 3
months with a goal to achieve BP control by 6
months (but they could see patients more often).
Research BP Measurement
• Automated Omron Device
• Measure 1 BP, record but do not use for
research value
• Measure 2 BP values and average them if
less than 4 mm Hg apart.
• If more than 4 mm different, measure a
4th BP and average the 2 closest BP
values (from the 2nd to 4th BP
measurements).
*
Systolic Blood Pressure
160
Systolic Blood Pressure (mmHg)
150
*
140
**
130
120
Prospective study ended
Intervention stopped
Retrospective evaluation of sustainability…
110
100
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Time
(months)
15
16
•- p<0.001; **- p=0.0015; *** - p=0.0023
Arch Intern Med. 2009;169(21):1996-2002
Journal of Clinical Hypertension 2011;13:431-437.
17
18
19
20
21
22
23
24
Meta-analysis of Potency of individual components of teambased care
Odds that BP was controlled
(95% confidence Interval)
Studies involving nurses
1.69 (1.48-1.93)
[69% increased chance]
Studies involving pharmacists within
physician offices or clinics
2.48 (2.05-2.99)
[148% increased chance]
Studies done in community
pharmacies
2.89 (1.83-4.55)
[189% increased chance]
Conclusion: All were effective but interventions by
pharmacists appear to be more potent than by
nurses.
Carter BL, et al. Archives of Internal Medicine 2009; 169:1748-1755.
17
Collaboration Among Pharmacists and
Physicians To Improve Outcomes Now
(CAPTION)
Barry L. Carter, Pharm.D.
Principal Investigator, CCC
Christopher Coffey, Ph.D.
Principal Investigator, DCC
Department of Pharmacy Practice and Science,
College of Pharmacy and
Professor
Department of Family Medicine
Roy J. and Lucille A. Carver College of
Medicine
Professor and Director, Clinical Trials Data
Management Center
* The study is being funded
by NHLBI/NIH, R01
HL091841-01A1.
College of Public Health
Important Concepts that Determine
Various Study Methodologies
Implementation: is the process of putting to use
or integrating evidence-based interventions within
a specific setting.
Sustainability: describes to what extent an
evidence-based intervention can deliver its
intended benefits over an extended period of time
after external support from the donor agency is
terminated.
Rabin BA et al. A glossary for dissemination and
implementation research in health. J Public Health
Management Practice 2008;14:117-123.
CAPTION Study Outcomes
Primary outcome = BP control @ 9 months
(determined via research measurement)
BP control defined as:
• < 140/90 for patients with uncomplicated
hypertension
• < 130/80 for patients with diabetes or
chronic kidney disease**
Outcomes
(cont.)
Secondary endpoints include:
• Mean BP @ 12, 18, 24 months
CAPTION
32 offices stratified by degree of pharmacy
services and racial minorities:
Then randomized to:
1. Usual care group
2. 9-month pharmacist intervention
3. 24-month pharmacist intervention.
Subjects followed for 24 months to determine:
1. What happens when the intervention is
stopped?
2. Can the intervention be sustained for 2 years?
3. Does the intervention benefit patients from
minority groups?
Participating Locations
Demographics
Variable
9 Month
(N=194)
N (%)
119 (61.3)
24 Month
(N=207)
N (%)
125 (60.4)
Control
(N=224)
N (%)
133 (59.4)
Total
(N=625)
N (%)
377 (60.3)
p-value
Age (SD) 60.6 (12.4)
56.7 (11.8)
60.5 (13.8)
59.3 (12.8)
0.055
BMI (SD)
33.8 (8.5)
35.2 (9.0)
32.9 (7.7)
33.9 (8.5)
0.090
DM or
CKD *
SBP
(SD)
102
(52.6%)
109
(52.7%)
103
(46.0%)
314
(50.2%)
0.5995
147.6
(13.7)
149.8
(15.6)
149.6
(15.3)
149.1
(15.0)
0.458
Female
0.938
* - Required a lower treatment goal (<130/80) making it
more difficult to achieve control
Pharmacist Interventions
Group
9 –Month
Group
24 – Month
Group
First 9 Months
(rate/month)
9-24 Months
(rate/month)
0.576
0.074
0.503
0.261
Primary Outcome 9-Month BP Control
Variable
BP
Control
All
subjects
BP
Control
Minorities
Intervention
Groups
(N = 401)
(N=226
minorities)
Control
Group
(N = 224)
(N=111
minorities)
Model-Adjusted
Difference –
Intervention vs.
Control
(95% CI)
p-value
43%
34%
1.57
( 0.99 , 2.50 )
0.052
37%
28%
1.54
( 0.84 , 2.81 )
0.16
* - Defined as <140/90 for uncomplicated BP and <130/80 for
diabetes or CKD
Provider Level Attitudes
Variable
Pharmacy Structure
Score ***
Low
High
Pharmacist TPB
Mean (SD)
Physician TPB
Mean (SD
9 Month
(N=11)
24 Month
(N=9)
Control
(N=12)
p-value
3 (27%)
8 (73%)
3 (33%)
6 (67%)
4 (33%)
8 (67%)
1.00
21.0 (2.18)
21.8 (1.42)
21.2 (3.89)
0.76
19.8 (2.39)
20.2 (1.56)
17.4 (2.85)
0.02
TBP = Theory of Planned Behavior
*** - There was a significant relationship between offices with
higher PSC and better BP control, adjusted odds ratio and CI =
1.75 ( 1.23 , 2.49 ), p = 0.002
*** - Subjects from sites with higher PSC had a mean lower SBP
of 4 mm Hg (p=0.007) and 2 mm DBP (P=0.009)
Primary Outcome 9-Month BP Control
When Adjusted for Baseline Pharmacy
Structure Score
Variable
BP
Control
All
subjects
BP
Minority
Subjects
Intervention
Groups
(N = 401)
(N=226
minorities)
Control
Group
(N = 224)
(N=111
minorities)
Model-Adjusted
Difference –
Intervention vs.
Control
(95% CI)
p-value
43%
34%
1.52
( 1.02 , 2.29 )
0.042
37%
28%
1.54
( 0.84 , 2.81 )
0.016
* - Defined as <140/90 for uncomplicated BP and <130/80 for
diabetes or CKD
9 – month BP- All subjects
Intervention Control
Variable
Groups
Group
(N = 401)
(N = 224)
ModelAdjusted
Difference –
Intervention
vs. Control
(95% CI)
p-value
SBP
Mean
(SD)
131.6
(15.8)
138.2
(19.7)
-6.07
( -9.64, -2.50 )
0.001
DBP
Mean
(SD)
76.3
(11.1)
78.0
(14.5)
-2.89
( -4.80, -0.99 )
0.003
Results – Minority subjects
Intervention Control
Variable
Groups
Group
(N = 226)
(N = 111)
Model
Adjusted
Difference –
Intervention
vs. Control
(95% CI)
pvalue
SBP
Mean
(SD)
133.0
(16.3)
140.3
(21.4)
-6.42
0.006
( -10.97, -1.87 )
DBP
Mean
(SD)
77.9
(10.7)
78.8
(15.9)
-2.98
( -5.76, -0.20 )
0.036
Systolic BP Results
BP Control Rates Results
Summary: Comparison Of Other
Studies With CAPTION
• Mean difference (control group minus the
intervention group) in SBP was -8 to -10
mm Hg in other studies and -6.1 mm Hg
in CAPTION
• Mean difference in DBP was -3 to -6 mm
Hg in other studies and -2.9 mm Hg in
CAPTION
• While the effect was slightly less in
CAPTION, the results were expected and
very similar to other studies
Comments and Questions