Table 5 Multivariate Cox regression analysis of bDMARDs discontinuation in patients of ankylosing spondylitis.

From: Effectiveness and drug retention of biologic disease modifying antirheumatic drugs in Korean patients with late onset ankylosing spondylitis

Method

Variate

Hazard ratio

95% CI

p value

Propensity score-based

LOAS (compared to YOAS)

1.229

0.931–1.622

0.146

Covariate adjustment

LOAS (compared to YOAS)

1.184

0.934–1.501

0.162

 

Female (compared to male)

1.291

1.030–1.617

0.027

 

BMI (kg/m2)

 

< 18.5

1.263

0.857–1.861

0.239

 

18.5–22.9

1.000 (Reference)

  
 

23.0–24.9

0.997

0.803–1.236

0.975

 

≥ 25.0

1.104

0.908–1.344

0.320

 

Smoking status

 

Non smoker

1.000 (Reference)

  
 

Ex-smoker

1.116

0.882–1.413

0.361

 

Current smoker

1.381

1.122–1.700

0.002

 

HLA B27 negative

1.473

1.155–1.878

0.002

 

ASDAS

0.945

0.864–1.034

0.216

 

BASFI score

1.000

0.965–1.036

0.996

 

Peripheral arthritis, yes vs. no

1.178

0.979–1.417

0.082

 

Biologics naïve (compared to previous TNFi exposure group)

0.959

0.784–1.172

0.679

 

Biologics type

 

Etanercept

1.000 (Reference)

  
 

Infliximab

0.995

0.775–1.277

0.969

 

Adalimumab

0.719

0.567–0.911

0.006

 

Golimumab

0.570

0.433–0.751

< 0.001

 

Secukinumab

0.595

0.185–1.917

0.385

  1. bDMARDs, biologic disease modifying antirheumatic drugs; YOAS, Young Onset Ankylosing spondylitis (< 50-year-old); LOAS, Late Onset Ankylosing Spondylitis (≥ 50-year-old); TNFi, tumour necrosis factor alpha inhibitor; BMI, body mass index; ASDAS, Ankylosing Spondylitis Disease Activity Score; BASFI, Bath Ankylosing Spondylitis Functional Index.
Back to article page