The Norfolk Arthritis Register Alan Silman arc Epidemiology Unit University of Manchester UK.

The Norfolk Arthritis RegisterThe Norfolk Arthritis Register

Alan SilmanAlan Silman

arc Epidemiology Unit arc Epidemiology Unit

University of University of

Manchester UKManchester UK

Manchester

The Norfolk Arthritis The Norfolk Arthritis RegisterRegister

A primary care based A primary care based

inception cohort study of inception cohort study of

patients with inflammatory patients with inflammatory polyarthritispolyarthritis

Norfolk

Why Norfolk?Why Norfolk?

Geographically ‘isolated’Geographically ‘isolated’ Stable populationStable population Single central major hospitalSingle central major hospital Excellent links primary to Excellent links primary to

secondary caresecondary care Local enthusiasmLocal enthusiasm

TopicsTopicsThe NOAR methodologyThe NOAR methodology

Key resultsKey results– Classification of RAClassification of RA

– Environmental risk factorsEnvironmental risk factors

– OutcomeOutcome

– Predictors of outcomePredictors of outcome

– Treatment effectsTreatment effects

Manchester

Norwich

NOAR : Recruitment

Entry criteria- age > 16 years- registered with local GP- swelling of > 2 joints- duration > 4 weeks- onset since 1/1/90

Metrology assessment

Apply ACR criteria

Metrologist AssessmentMetrologist Assessment

Data CollectedData Collected

(())------(())(())--X-raysX-rays

------(())(())Blood testsBlood tests

----------Pain scorePain score

------------SF36SF36

HAQHAQ

- - - deformity- deformity

----- - - activity- activity

ManikinManikin

DrugsDrugs

Co-morbidityCo-morbidity

ReproductiveReproductive

OccupationOccupation

101077554433221100

AnniversaryAnniversary

The Norfolk Arthritis Register (NOAR)

To establish the incidence of IP and subset with RA

To identify risk factors for the development of IP and RA

To study the natural history of treated IP and RA

To identify predictors of outcome in IP and RA

Initial aims

The Norfolk Arthritis Register (NOAR)

To investigate the epidemiology of cardiovascular disease in patients with IP (risk factors, incidence and outcome)

To identify predictors of treatment response and non-response

Current Major Aims

Key resultsKey results

Incidence of IP Incidence of IP and RAand RA

15-24 25-34 35-44 45-54 55-64 65-74 75+0

20

40

60

80

100

120

Incidence /100,000 pyrs

MaleFemale

[Symmons et al - Br J Rheum, 1994; 33:735-9]

Age group

IP RA

15-24 25-34 35-44 45-54 55-64 65-74 75+0

20

40

60

80

100

120

Incidence /100,000 pyrs

MaleFemale

Incidence of IP and RA in 1990

Estimates of the incidence of RA:Estimates of the incidence of RA:Application of ACR criteriaApplication of ACR criteria

0

10

20

30

40

50

60

70

80

90

100

15 25 35 45 55 65 75+

0

10

20

30

40

50

60

70

80

90

100

15 25 35 45 55 65 75+

Inci

denc

e ra

te p

er

100,

00

Females Males

ACR criteria applied at baseline ACR criteria applied over 5 years

Age Age

Issues of ClassificationIssues of Classification

IP vs RAIP vs RA

ConceptConcept

Early IPEarly IP

Recovery

Another disease

Established Established RARA

ConceptConcept

Early IPEarly IP

Recovery

Another disease

? Treatment

Established Established RARA

Does early RA Does early RA exist?exist?

Are there differences between IP Are there differences between IP destined to differentiate into RA and destined to differentiate into RA and other ‘causes’ of IP?other ‘causes’ of IP?

Immunisation X

Parvovirus X

Psoriasis X

Can we distinguish early RAfrom other forms of early

arthritis?

Leiden model: Leiden model: prediction of outcomeprediction of outcomeGoal:Goal:To discriminate at first visit To discriminate at first visit

between patients who will go between patients who will go on to have:on to have:

self-limiting arthritisself-limiting arthritis persistent non-erosive persistent non-erosive

arthritisarthritis persistent erosive arthritispersistent erosive arthritis

Leiden model: 7 variablesLeiden model: 7 variables

Symptom duration at presentationSymptom duration at presentation

Morning stiffness > 1 hourMorning stiffness > 1 hour

Arthritis of > 3 jointsArthritis of > 3 joints

Bilateral compression pain of MTPsBilateral compression pain of MTPs

Rheumatoid factorRheumatoid factor

Anti-cyclic citrullinated peptide Anti-cyclic citrullinated peptide antibodyantibody

Erosions in hands or feetErosions in hands or feet

Validation of Leiden Validation of Leiden modelmodelerosive vs non-erosive erosive vs non-erosive arthritisarthritis In presence of persistenceIn presence of persistence

Radiological criterion omittedRadiological criterion omitted

LeidenLeiden NOARNOAR(n= 526)(n= 526) (n=486)(n=486)

Prediction model ROCPrediction model ROC 0.830.83 0.760.76

ACR criteria ROCACR criteria ROC 0.770.77 0.660.66


Risk factors for the Risk factors for the development of IP development of IP and RAand RA

Sources of DataSources of Data

Descriptive AnalysisDescriptive Analysis

Local Clustering of RALocal Clustering of RA

Silman et al., 1999

Jan1990

June Jan1991

June Jan1992

June

25

20

15

10

5

0

30

Month of onset

Number of

new cases

All casesUIPRA

Onset of Disease by Onset of Disease by Month 1990-92Month 1990-92

Silman et al., 1997

Observed & Expected Observed & Expected Events in Relation to Events in Relation to Time & DistanceTime & Distance

Silman et al., 1997

0

-5e7

-1e8

400300

200100

D

Time 5001000

1500

Distance

Socioeconomic Deprivation vs RA Socioeconomic Deprivation vs RA Incidence by census wardIncidence by census ward

Bankhead et al., J Rheum 1996

IndicatorIndicator rrss

Households in rented Households in rented accommodationaccommodation

-0.09-0.09

Overcrowded Overcrowded accommodationaccommodation

-0.14-0.14

Householders with no CHHouseholders with no CH -0.26-0.26

Households with no access Households with no access to a carto a car

-0.06-0.06

Male unemployment (age Male unemployment (age 26-60)26-60)

-0.03-0.03

Socioeconomic Deprivation & Socioeconomic Deprivation & RARA

Bankhead et al., J Rheum 1996

0

10

20

30

40

50

I & II IIIN IIIM IV V

Men

Women

Social Class

Incidence/100,000

* IV & V combined for men

*

Sources of dataSources of data

Case control studiesCase control studies

1. Internal NOAR1. Internal NOARCases 1992 (n=165) : Cases 1992 (n=165) :

– aged 18-70aged 18-70– symptom duration < 12 monthssymptom duration < 12 months

Controls: 2 per case from referring Controls: 2 per case from referring primary careprimary care

Lifestyle Factors

SmokingObesity

20

10

5

4

3

2

1

Association of Smoking with Association of Smoking with Severe RA: Rheumatoid NodulesSevere RA: Rheumatoid Nodules

CurrentExNever

20

10

4

2

1

0.4

Harrison Harrison et al.et al., Arth Rheum 2003, Arth Rheum 2003

Odds Ratio(95% CI)

Hormonal Risk Factors

TerminationOral ContraceptiveMiscarriage

8

6

4

2

1.8

.6

.4

.2

.1

Symmons et al., 1997

Cases Controls0

5

10

15

20

%

Association between Association between Prior Blood Transfusion Prior Blood Transfusion and RAand RA

2. 2. NOAR EPIC LinkNOAR EPIC Link

Co-occurrence of NOAR & EPICCo-occurrence of NOAR & EPIC in same population in same population

Area for new cases of IP referred to NOAR

EPIC

practices

European Prospective study of the Incidence of Cancer (EPIC-Norfolk)

Baseline assessmentsBaseline assessments

Random sample (n= 25,000)Random sample (n= 25,000)

45 – 75 years45 – 75 years

Recruited 1993 – 1997Recruited 1993 – 1997

Health and lifestyle questionnaireHealth and lifestyle questionnaire

Height and weightHeight and weight

Prospective ‘nested’ case Prospective ‘nested’ case control studycontrol study

Free of IP at Free of IP at baselinebaseline

Subsequent Subsequent registration with registration with NOARNOAR

2 per case2 per case

Matched:Matched:- age (- age (± 3 years)± 3 years)

- - gendergender- - within 3 months ofwithin 3 months of

baseline assessmentbaseline assessment

73 Cases Controls

EPIC Diet SurveyEPIC Diet Survey

7 day detailed food diary with 7 day detailed food diary with portion sizesportion sizes

Fruit Intake (g) and Fruit Intake (g) and Development of IPDevelopment of IP

Highest (ref)MiddleLowest

86

4

2

1.8.6

.4

.2

*Adjusted for energy intake, smoking, red meat intake

Odds Ratio (95% CI)*

Pattison et al., ARD 2004

Tertile

Tertiles of Vitamin C Intake (mg)Tertiles of Vitamin C Intake (mg)

Highest (ref)MiddleLowest

86

4

2

1.8.6

.4

.2

*Adjusted for energy intake, smoking, protein intake

Pattison et al., ARD 2004

Odds Ratio

(95% CI)*

Tertile

Tertiles of Tertiles of -cryptoxanthin Intake -cryptoxanthin Intake (µg)(µg)

HighestMiddleLowest (ref)

3

2

1

.5

.3

.2

.1

Odds Ratio

(95% CI)*

*Adjusted for energy intake, smoking, protein intake

Pattison et al. 2005

Tertile

HighestMiddleLowest (ref)

3

2

1

.7

.5

.3

.2

.1

RA and Dietary Zeaxanthin RA and Dietary Zeaxanthin IntakeIntakeOdds

Ratio (95% CI)*

*Adjusted for energy intake, protein, smoking

Pattison et al 2005

Tertile

Red Meat & Meat Products and Red Meat & Meat Products and Development of IPDevelopment of IP

*Adjusted for energy intake, smoking, fruit intake

MiddleLowest (ref) Highest

8

6

4

2

1.8

.6

Pattison et al., A & R 2004

Odds Ratio

(95% CI)*

Are the Diet Effects Independent?Are the Diet Effects Independent?

Vitamin C mg/day

HighMiddleLow

86

4

2

1.8.6

.4

.2

.1

Red Meat g/day

HighMiddleLow

86

4

2

1.8.6

.4

.2

.1

Odds Ratio

(95% CI)*


The natural The natural history of history of

treated IP and RAtreated IP and RA

Outcomes investigated

Persistence Radiological damage

Physical function (HAQ) Economic costs

Health status (SF-36) Co-morbidity

Work disability Mortality

Work disabilityWork disability

Year of onset

0

5

10

15

20

25

30

35

Oneyear

Twoyears

Per

cen

tage

1989-19921994-1997

All cause mortalityAll cause mortality

Seropositive patients

Men WomenNorfolk0

1

SMRInflammatory polyarthritis

0

1

2

SMR

Men WomenNorfolk

SMR = 1.13 SMR = 1.01

2

SMR =1.51 SMR = 1.41

Cardiovascular mortality:Cardiovascular mortality:Influence of RF StatusInfluence of RF Status

Males0.5

1

2

3

Females

RF-

SMR(95% CI)

RF+


Predictors of Predictors of outcomeoutcomegeneticgenetic

environmentalenvironmental

treatmenttreatment

X-ray strategy NOAR

Time fromregistration

Patients X-rayed

0

1

2

5

None

3 ACR criteria at baseline 2 ACR criteria at year one

2 ACR criteria at year two and no erosions on anyprevious X-rays

All patients

Timing of first erosions

Risk set Time of 1st“erosion free”

X-rayMedian (IQR)

Timing of 2ndX-ray

Median (IQR)

% erosiveat 2nd X-ray

Incidence rateof 1st erosions(per 1000 pm)

(95% CI)

1

2

3

4

18 (16-20)

29 (26-31)

41 (37-45)

18 (16-22)

66 (64-69)

69 (66-73)

75 (70-84)

36

23

28

47

24 (21-29)

5 (4-8)

7 (5-10)

13 (9-19)

NOAR: Predicting NOAR: Predicting radiological erosionsradiological erosions

Risk group RF > 40Initial duration > 3 months

Probability of erosions

1

2

3

4

X

X

X

X

0.79

0.52

0.33

0.10

Overall performance: PPV 61% NPV 74%

Role of genetic factorsRole of genetic factors

HLA.DRBIHLA.DRBI CytokineCytokine

– TNFTNF– IL1 etc etcIL1 etc etc

MMPMMP MBLMBL MIFMIF

Weak association with shared Weak association with shared epitope, less strong than in epitope, less strong than in clinic based studiesclinic based studies

Few candidates tested were Few candidates tested were predictors of presence/severity predictors of presence/severity erosionserosions

Genetic Factors

? Confounding effect of ? Confounding effect of therapytherapy

Propensity models

Bias in treatment assignments“Confounding by indication”

Variable duration of exposure to treatment

Problems

Solution

Assessing the effect of treatment

In observational studies :

It is not random who will get DMARD therapy

Treated patients have more severe disease

Therefore ‘bias in allocation’ occurs

Adjustment for this effect is needed

Propensity modelling

Logistic model used to predict treatment decision

Using disease characteristics that inform treatment decision

Each individual given probability of being treated = propensity score

Distribution of HAQ scores at year 5

Never on DMARDs

<6 months 6-12 months >12 months

Delay from symptom onset to start of first DMARD

0.00

0.50

1.00

1.50

2.00

2.50

3.00H

AQ

sco

re a

t ye

ar 5

Odds of moderate disability(HAQ1.0) at 5 years

Delay from onset to start of treatment

Odd

s ra

tio (9

5% C

I)

DMARDs/steroidsMonthsNever on < 6 6-12 > 12

0.5

1

2

5

10

20

Odds of moderate disability (HAQ 1.0) at 5 years(Models include propensity scores & hospital referral)


Odd

s ra

tio (9

5% C

I)

Never on < 6 6-12 > 120.2

0.3

0.5

1

2

3

5

10

DMARDs/steroidsMonths

Odds of moderate disability(HAQ 1.0) at 5 years

0.2

0.5

1

2

5

10

20

Odd

s ra

tio

(95%

CI)

Without propensity score

With propensity score


Never onDMARDs/steroids < 6 6-12 > 12 Months

Larsen score at year 5 adjusted for propensity score

Delay to start of first DMARD

< 6 months 6-12 months > 12 monthsNo Treatment

0.5

1

2

3

0.5

1

2

3

Patients treated with DMARDS had worse disease at presentation and worse outcome

The greatest benefit of treatment was seen in those treated within six months

Jt Principal Investigator : Deborah Symmons

Research Fellows : Marwan Bukhari

Beverley Harrison

Nicola Goodson

Research Assistants : Clare Bankhead

Nicola Wiles

Dorothy Pattison

Statisticians : Paul Brennan

Mark LuntResearch nurses, consultant rheumatologists

Acknowledgements

The Norfolk Arthritis Register Alan Silman arc Epidemiology Unit University of Manchester UK.

Documents