Transcript Slide 1

Patient questionnaires:
Standardized quantitative
“scientific” data from a
patient history, the primary
source of rheumatology
treatment decisions
Theodore Pincus MD
Clinical Professor of Medicine
New York University
[email protected]
Disclosures
Theodore Pincus, MD
Sources of Funding for Research: Amgen Inc.; BristolMyers Squibb Company
Consulting Agreements: Abbott Laboratories; Amgen
Inc.; Bristol-Myers Squibb Company; UCB
Speakers’ Bureau/Honorarium Agreements: Abbott
Laboratories; Wyeth Pharmaceuticals, Genentech
Financial Interests/Stock Ownership: None
Discussion of Off-Label, Investigational, or
Experimental Drug Use: None
Many, if not most, doctors have
extensive information about their
patients with a few mouse clicks
concerning:
Scheduling
Billing
Laboratory tests
Medications
BUT NOT: Is the patient better, worse,
or the same? With which treatments?
Why not ask the patient in a
structured, “scientific” format, ie,
self-report questionnaire?
Why measurement?
 This wine is expensive –
$60 or $6,000
 The patient has a fever –
101º or 106ºF, 38º or 40ºC
 The blood pressure is high –
150/95 or 250/125
 The patient is “doing well” –
What is the DAS28, CDAI or RAPID3
Complexities in quantitative
assessment of patients with
RA and rheumatic diseases
• Laboratory tests are limited in diagnosis
and treatment decisions
• Treat radiograph before damage
• No single ‘Gold Standard’ measure, eg,
blood pressure, cholesterol, glucose, for
diagnosis and management in all
individual patients
• Therefore, need indices of 3–7 measures
American College of Rheumatology (ACR)
Core Data Set & Disease Activity Score (DAS)
3 Physician/Assessor measures
1. Tender joint count (also in DAS)
2. Swollen joint count (also in DAS)
3. Assessor Global status
3 Patient self-report measures
4. Physical Function - HAQ, HAQ II, MDHAQ
5. Pain
6. Patient Global status (also in DAS)
1 Laboratory Measure
7. Acute phase reactant –ESR, CRP–also in DAS
(8. Radiograph – longer than 1 year)
Felson et al, Arth Rheum 36:729, 1993. van Riel, Br J Rheumatol 31:793, 1994.
Types of measures to
assess patients with RA
•
•
•
•
•
Joint counts
Radiographs
Laboratory tests
Patient questionnaires
Global estimates
Formal Joint Counts in
Management of Patients With RA
 Most specific measure to assess RA
 Most important measure in clinical
trials – 20, 50, 70% required for ACR
improvement criteria
 Widely-accepted by rheumatologists
and FDA as “best” measures
 28-joint count as useful as 68–70
joint count
Changes in ACR Core Data Set Measures Over 12 Months:
Leflunomide (LEF) vs Methotrexate (MTX) vs Placebo (PBO)
Measure:
LEF
PBO MTX
Tender Jts
Swollen Jts
MD Global
ESR
FN- HAQ
FN-MHAQ
Pain
Pt Global
-7.7
-5.7
-2.8
-6.3
-0.45
-0.29
-2.2
-2.1
-3.0
-2.9
-1.0
+2.6
+0.03
+0.07
-0.4
+0.1
-6.6
-5.4
-2.4
-6.5
-0.26
-0.15
-1.7
-1.5
Effect Relative
Size Efficiency
-0.59
1.00
-0.44
0.56
-0.68
1.33
-0.41
0.48
-0.80
1.84
-0.69
1.37
-0.65
1.21
-0.81
1.88
Strand V, et al. Arch Intl Med. 1999; 159:2542-2550;
Tugwell P, et al. Arthritis Rheum. 2000; 43:506-514.
Question for Rheumatologists
For patients with RA under your care (not including
patients in clinical trials), how often do you perform
formal tender and swollen joint counts?
Never
13%
1–24% of visits
25–49% of visits
50–74% of visits
75–99% of visits
Always
32%
11%
14%
16%
14%
Time to Score RA Measures - Seconds
150
114
100
50
106
94
42
9.6
4.6
0
28 Joint HAQ-DI DAS28
Count
CDAI
RAPID3 RAPID3
(0-10) (0-30)
Pincus et al 2009; Arthritis Care Res. in press
Some Limitations of Formal Joint Counts
 Relative efficiencies similar or lower
than global and patient measures in
clinical trials
 May improve over 5 years while joint
damage and functional disability may
progress
 Poorly reproducible
 Not performed at most visits in usual
care
The most specific measure
for diagnosis is not
necessarily the most
significant measure for
prognosis and management.
Radiographs in Diagnosis and
Management of Patients With RA
 Excellent quantitative scoring
systems - Sharp, van der Heijde,
Larsen, Genant
 Erosions are closest to
pathognomonic sign in RA
 Reflect cumulative damage of
disease
9- to 10-Year Survival According to Quantitative
Markers in Three Chronic Diseases
A
Rheumatoid Arthritis – Activities of Daily Living
B
100
>90%
81%–90%
80
% Active “With Ease”
60
40
71%–80%
70%
20
Survival (%)
Survival (%)
100
Rheumatoid Arthritis – Formal Education Level
>12 Years
80
9–12 Years
60
8 Years
40
20
(Data from Pincus et al, 1987)
(Data from Pincus et al, 1987)
Months
0
40
60
80
100
Hodgkin Disease – Anatomic Stage
100
0
D
Stage I
80
60
Stage II
All Stages,
All Causes
Stage III
Stage IV
40
20
(Data from Kaplan, 1972)
0
2
4
6
Years
8
10
Survival (%)
Survival (%)
C
20
Months
20
40
60
80
100
Coronary Artery Disease – No. of Involved Vessels
100
80
1 Artery
60
2 Arteries
40
3 Arteries
LCA
20
(Data from Proudfit et al, 1978)
0
2
4
6
8
Years
10
Change from baseline (Mean +/- SE)
TEMPO Trial: Year 2 Radiograph:
Change in Total Sharp Score from
8
Baseline to Year 2
7
6
5
MTX = 206
E = 203
MTX+E = 213
3.34
(CI 1.18, 5.50)
4
3
2
1.10*
(CI 0.13, 2.07)
1
0
-1
* p < 0.05, E vs MTX
† p < 0.05, Combination vs MTX
‡ p < 0.05, Combination vs E
-0.56†‡
(CI –1.05, -0.06)
450
400
350
300
250
200
150
100
50
1
1.59
-0.54
ERA ETA
ERA MTX
TEMPO
Combi
0.52
2.8
0.4
3.7
1.3
3
5.7
IFX MTX
PREMIER
Combi
PREMIER
ADA
PREMIER
MTX
0
TEMPO ETA TEMPO MTX IFX Combi
Yazıcı Y, Yazıcı H, Arthritis Rheum 2006;54(supl)
2 Year Change in Total Sharp/van der Heijde
X-ray score (0–448): TEMPO probability plot
TEMPO=Trial of Etanercept and MTX with radiographic Patient Outcomes.
van der Heijde, et al. Arthritis Rheum 2006;54:1063–74.
19
Strongly and Weakly Related
Measures to Assess RA
Radiographs
ESR, CRP
Shared epitope
Rheumatoid factor
Joint deformity
Duration of disease
Functional disability
Pain
Patient global estimate
Socioeconomic status
Joint tenderness
Age
Pincus T, Sokka T:
Best Pract Res Clin Rheumatol 17:753-781, 2003.
Predicting Mortality in RA: Most Baseline
Measures Are Worse in Patients Who Will
Die Over a 5-Year Period
Mean Baseline Values
P Value
Alive
Dead
Age (years)
ARA functional class
Number of comorbidities
Walking time
ESR
mHAQ score
Learned helplessness
Global self-report
Number of extra-articular features
Duration of disease
Years of education
Joint count
55.1
2.2
1.1
10.8
33.8
1.98
2.41
2.6
0.2
9.1
10.8
12.8
65.5
2.6
2.1
16.8
48.3
2.32
2.55
3.0
0.5
12.7
9.4
15.9
< 0.001
< 0.001
< 0.001
< 0.001
0.004
0.005
0.007
0.01
0.02
0.03
0.03
0.04
Radiograph score
RF titer
Pain
1.2
2.7
5.40
1.4
2.9
5.19
0.20
0.28
0.68
Callahan LF, et al. Arthritis Care Res. 1997;10:381–394.
RA Cohort #2- Cox Proportional Hazards Model
Analyses Including Demographic, Functional, SelfReport, Joint Count, X-ray, Laboratory and Disease
Variables in 206 patients
Age
Univariate
RR
P
(95% CL) Value
1.07
<0.001
Stepwise Model
RR
P
(95% CL) Value
1.06
<0.001
Comorbidity
MHAQ ADL Score
Disease duration
Education
ESR
Joint count
1.63
2.00
1.04
0.89
1.01
1.02
<0.001
0.003
0.02
0.007
0.005
0.10
1.40
1.76
-----
Walking time
X-ray
1.03
1.40
0.04
0.17
---
Arthritis Care Res 10:381,1997
0.02
0.02
-------
MRI can better identify early bone
erosions than X-ray
Some Problems With
Radiographs in RA
1. Quantitative score tedious to perform
2. Treatment initiated prior to erosions –
MRI, ultrasound more sensitive
3. Radiographic damage has poor
prognostic value for work disability,
death and even joint replacement
4. Treatment prior to erosions
Laboratory Tests in Diagnosis and
Management of Patients With RA
1. Most important measure in most
clinical situations, e.g.,
cholesterol, hemoglobin,
creatinine, glucose, etc.
2. Many tests may be of value –
CBC, ESR, CRP, RF, anti-CCP
3. No work for the rheumatologist
Textbook statements concerning
ESR in RA
"the erythrocyte sedimentation rate is increased
in nearly all patients with active RA”
Lipsky PE. Rheumatoid arthritis. In: Fauci AS, Langford CA,
eds. Harrison's Medicine. New York: McGraw-Hill,2006:85.
“at least 5% of patients with clinically active
disease may have a normal ESR”
Chatham WW, Blackburn WD, Jr. Laboratory findings in
rheumatoid arthritis. In: Koopman WJ, Moreland LW, editors.
Arthritis and allied conditions: a textbook of rheumatology.
Philadelphia, PA: Lippincott, Williams &
Wilkins,
2005:1207
Traditional approaches to clinical expertise:
EMINENCE BASED MEDICINE - making the same
mistakes with increasing confidence over an
impressive number of years
ELOQUENCE BASED MEDICINE - a year-round
suntan and brilliant oratory may overcome absence
of any supporting data
ELEGANCE BASED MEDICINE - where the sartorial
splendor of a silk-suited sycophant substitutes for
substance
The modern alternative?
EVIDENCE BASED MEDICINE - the best approach
to clinical data - requires information from clinical
observational data in addition to clinical trials
Pincus and Tugwell J Rheumatol 2006
ESR Values in Patients With RA
Wolfe F, Michaud K, J Rheumatol.
1994;21:1227–1237. Wichita KS, USA
ESR ≥ 28
mm/h
ESR < 28
mm/h
Females
63%
37%
Males
55%
45%
Similar results have seen reported from:
Nashville, TN USA
Jyvaskyla, Finland
Oslo, Norway
Nancy, France
Gronigen, the Netherlands Belfast, Ireland
Mean ESR (mm/Hr) 4 Locations – 1996:
Location
Oslo,Norway
Nancy, France
Gronigen, Netherlands
Belfast, N Ireland
n
237
135
283
51
ESR
6
9
8
8
26
29
28
28
Smedstad LM, Moum T, Guillemin F,Kvien TK, Finch MB, Suurmeijer
TPBM, Van Den Heuvel WJA
Br J Rheumatol 1996; 35:746-51
ESR and CRP at 1st visit in US
and Finland – 1980-2005
CRP
ESR
≥28 mm/hr
<28 mm/hr
Jyvaskyla, Finland n=1744
Total
55%
45%
<10 mg/L
11%
33%
>10 mg/L
44%
12%
Nashville, Tennessee, USA n=170
Total
45%
55%
<10 mg/L
17%
42%
>10 mg/L
28%
13%
Total
100%
44%
56%
100%
59%
41%
Sokka and Pincus – J Rheumatol 2009
Mean/median baseline ESR in RA patients in
23 studies, by first year of recruitment
First year of
recruitment
Period of
recruitment
Median ESR
(mm/h)
Mean ESR
(mm/h)
1954-1980
(7 studies)
1954-1995
47
50
1981-1984
(8 studies)
1981-1999
38
41
1985-1996
(8 studies)
1985-2000
36
35
Abelson B, Sokka T, Pincus T. J Rheumatol 2009
Meta-analysis: Anti-cyclic citrullinated peptide
(CCP) antibody and rheumatoid factor (RF)
Anti-CCP
Number of studies
37
Positive likelihood ratio
12.5
Odds ratio for RA 16.1 – 39.0
RF
50
4.9
1.2 – 8.7
Nishimura K et al. Annals of Internal Medicine 146:797-808, 2007
Meta-analysis: Anti-cyclic citrullinated peptide
(CCP) antibody and rheumatoid factor (RF)
Number of studies
Positive likelihood ratio
Odds ratio for RA
Sensitivity
Specificity
% of Patients with
negative test result
Anti-CCP
37
12.5
16.1 – 39.0
67%
95%
RF
50
4.9
1.2 – 8.7
69%
85%
33%
31%
Nishimura K et al. Annals of Internal Medicine 146:797-808, 2007
RA Cohort #2- Cox Proportional Hazards Model
Analyses Including Demographic, Functional, SelfReport, Joint Count, X-ray, Laboratory and Disease
Variables in 206 patients 1985-1990
Age
Univariate
RR
P
(95% CL) Value
1.07
<0.001
Stepwise Model
RR
P
(95% CL) Value
1.06
<0.001
Comorbidity
MHAQ ADL Score
Disease duration
Education
ESR
Joint count
1.63
2.00
1.04
0.89
1.01
1.02
<0.001
0.003
0.02
0.007
0.005
0.10
1.40
1.76
-----
Walking time
X-ray
1.03
1.40
0.04
0.17
---
Arthritis Care Res 10:381,1997
0.02
0.02
-------
5-Year Survival in 206 Patients
With RA: Cohort #2 – 1985-1990
Rheumatoid Factor
MHAQ Score
80
80
Survival (%)
100
Survival (%)
100
60
60
40
Absent (29)
Present
20
(175)
0.00 (12)
0.01–0.99 (91)
1.00–1.99 (86)
>2.00 (21)
40
20
0
0
0
12
24
36
48
Months After Baseline
Arthritis Care Res 10:381,1997
60
0
12
24
36
48
Months After Baseline
60
IgM rheumatoid factor binding IgG
MultiDimensional
Health
Assessment
Questionnaire
(MDHAQ)
Page 1
% of RA patients with abnormal
measures at presentation: evidence,
not eminence-based
• RF positive
-
69% (1)
• Anti-CCP positive
-
67% (1)
• ESR >28 mm/Hr
-
57% (2,3)
• CRP >10
-
58% (2)
1- Nishimura et al, Ann Int Med 146:797-808, 2007
2 - Wolfe and Michaud, J Rheumatol 21:1227–1237, 1994
3 - Sokka and Pincus, J Rheumatol 36:1387--1390, 2009
Some Problems With Laboratory Tests in
Diagnosis and Management of RA
1. ESR & CRP - normal in 40% at
presentation
2. Anti-CCP & RF - negative in 20–50% of
patients
3. Treatment decisions are based primarily
on clinical criteria
4. Lab tests have good prognostic value for
radiographic damage but poor prognostic
value for work disability or death
CRP = C-reactive protein; CCP = cyclic citrullinated protein
Patient self-report questionnaires
1. HAQ and RAPID3 score as informative as
ACR20/50/70 or DAS in clinical trials
2. Significant correlation with joint count,
ESR, X-ray – individual measures and
indices
3. Predict work disability, costs, TJR, and
premature death more significantly than
traditional measures
4. Quantitative measures to save time for
patient and MD to focus on major patient
matters
9-10 Year Survival According to Quantitative Markers
in Three Chronic Diseases
Rheumatoid Arthritis Activities of Daily Living
A
Rheumatoid Arthritis -
B
>90%
81–90%
80
% Active “With Ease”
60
40
71–80%
Survival (%)
Survival (%)
100
>12 Years
80
9–12 Years
60
8 Years
40
20
70%
20
Formal Education Level
100
(Data from Pincus et al, 1987)
(Data from Pincus et al, 1987)
20
60
80
100
100
Stage I
80
Stage II
Stage III All Stages,
All Causes
Stage IV
60
40
0
Months
Hodgkin’s Disease Anatomic Stage
C
Survival (%)
40
20
D
0
2
4
6
8
10
Years
40
60
80
100
Months
Coronary Artery Disease # of Involved Vessels
80
60
40
20
(Data from Kaplan, 1972)
20
100
Survival (%)
0
1 Artery
2 Arteries
3 Arteries
LCA
(Data from Proudfit et al, 1978)
0
2
4
6
8
10 Years
5-Year Survival in 206 Patients
With RA: Cohort #2 – 1985-1990
Rheumatoid Factor
MHAQ Score
80
80
Survival (%)
100
Survival (%)
100
60
60
40
Absent (29)
Present
20
(175)
0.00 (12)
0.01–0.99 (91)
1.00–1.99 (86)
>2.00 (21)
40
20
0
0
0
12
24
36
48
Months After Baseline
Arthritis Care Res 10:381,1997
60
0
12
24
36
48
Months After Baseline
60
Significance of 8 variables as predictors of mortality in 53 RA cohorts
Significant in multivariate analyses
100%
6%
4%
22%
30%
Significant in univariate analyses
34%
17%
32%
23%
21%
39%
32%
46%
Not Significant
50%
39%
75%
50%
50%
28%
25%
72%
0%
65%
Physical
Cofunction morbidities
(N=18)
(N=23)
45%
44%
37%
31%
Rheumatoid
factor
(N=29)
Extraarticular
disease
(N=18)
ESR
(N=19)
Socioeconomic
status
(N=13)
22%
11%
Joint
count
(N=18)
Hand
radiograph
(N=18)
Sokka T, Abelson B, Pincus T. Clin Exp Rheumatol 26(suppl):S35-61, 2008
Prediction of premature
mortality according to blood
pressure and cholesterol
converted hypertension and
hypercholesterolemia from
optional treatments to major
public health campaigns.
Imagine doctors saying that they
do not measure blood pressure
or cholesterol because
“it takes too much time” or
“the staff will not cooperate,”
as suggested for why they do not
measure physical function.
The MDHAQ in Clinical Rheumatology
• In rheumatoid arthritis, the MDHAQ
distinguishes MTX or LEF from placebo in a
clinical trial as effectively as a joint count
or the ACR 20
• In osteoarthritis, the MDHAQ distinguishes
NSAID from acetaminophen as effectively
as the WOMAC
• In fibromyalgia, the MDHAQ distinguishes
patients from those with rheumatoid
arthritis as effectively as an ESR
Physical function/activities of daily living (ADL)
in prognosis of non-Rheumatic Diseases
• In congestive heart failure, ADL predict 36month mortality as ejection fraction
Konstam, Am J Cardiology 78:890, 1996
• In AIDS, ADL predict 36-month mortality as
CD4/CD8 ratios, clinical AIDS prognostic
staging (CAPS), severity classification for
AIDS hospitalizations (SCAH)
Justice, J Clin Epidemiology 49:193, 1996
• In hospitalized elder patients, ADL predict
1-year mortality beyond physiologic data
and comorbidities
Covinsky, J Gen Intern Med 12:203, 1997
Some limitations of patient self-report
questionnaires
 Need for translation
 Cultural and linguistic issues
 Possibility of ‘gaming’ by
patient, health professional to
provide desired responses
 Not specific to any disease