Table 6

Multiple logistic regression of possible independent predictors of radiographic progression at two years.

Model with x- ray scores^a

Model without x- ray scores^b

OR

95% CI

p

OR

95% CI

p

Change DXR 1 yr

1.001

.0.974

1.029

.956

0.971

0.955

0.988

.001

Change SHS 1 yr

2.667

2.027

3.511

.000

Erosions at baseline

.666

.262

1.691

.392

2.592

1.486

4.519

.001

Anti- CCP

3.475

1.332

9.066

.011

3.132

1.701

5.766

.001

Number of swollenjoints 1 yr

1.086

.931

1.267

.296

1.129

1.021

1.248

.018

DAS28 1 yr

1.220

.645

2.307

.541

1.181

.791

1.764

.417

General health 1 yr

.990

.961

1.021

.527

.997

.979

1.016

.761

ESR baseline

.999

.979

1.018

.886

.991

.979

1.004

.178

ESR 1 yr

1.014

.980

1.050

.415

1.005

.982

1.030

.665

HAQ 1 yr

1.327

.570

3.092

.512

1.079

.608

1.916

.795

CRP 1 yr

.974

.929

1.020

.261

1.006

.980

1.032

.656

Constant

.014

.000

.060

.001

^aR² = 77%. 91% correct classification. ^bR² = 36%. 75% correct classification. Anti-CCP, antibodies against cyclic citrullinated peptides; CI, confidence interval; CRP, C-reactive protein; DAS28, Disease Activity Score calculated on 28 joints; DXR, digital X-ray radiogrammetry; ESR, erythrocyte sedimentation rate; HAQ, Health Assessment Questionnaire; OR, odds ratio; SHS, Sharp van der Heijde Score; yr, year.

Forslind et al. Arthritis Research & Therapy 2012 14:R219   doi:10.1186/ar4058