Table 2

Effect of bDMARDs on the likelihood of progression: comparative studies with continuous exposure to treatment

Study	Study design	Follow-up (years)	Treatment	N	Handling of pretreatment confounders	Definition of progression	OR (95% CI)
Baraliakos et al, 200534	OLE versus cohort (GESPIC)	2	INF	41	Adjust. (not clear which variables)	≥1 mSASSS	1.5 (0.4 to 6.5)*
Baraliakos et al, 200534	OLE versus cohort (GESPIC)	2	No TNFi	41	Adjust. (not clear which variables)	≥1 mSASSS	1.5 (0.4 to 6.5)*
van der Heijde et al, 200831	OLE versus cohort (OASIS)	2	INF	165	Match. (inclusion criteria), adjust. (mSASSS) and strat. (gender, age, CRP, BASDAI, BASFI, BASMI, PGA and HLA-B27)	≥1 mSASSS	0.9 (0.6 to 1.5)†
van der Heijde et al, 200831	OLE versus cohort (OASIS)	2	No TNFi	61		≥1 mSASSS	0.9 (0.6 to 1.5)†
Haroon et al, 201339	Cohort	1.5–9	TNFi	201	PS matching‡	≥1 mSASSS/year	0.3 (0.1 to 0.8)‡
Haroon et al, 201339	Cohort	1.5–9	No TNFi	133	PS matching‡	≥1 mSASSS/year	0.3 (0.1 to 0.8)‡
Kim et al, 201640	Cohort	2–5	TNFi	269	PS matching§	≥2 mSASSS	0.7 (0.3 to 1.6)§
Kim et al, 201640	Cohort	2–5	No TNFi	341	PS matching§	≥2 mSASSS	0.7 (0.3 to 1.6)§

*Calculated from observed data (p value of adjusted analysis not significant and no adjusted effect size reported).
†Calculated from observed data in the matched cohorts (adjusted and stratified analyses also not significant).
‡Adjusted OR after PS matching (n=142); baseline variables used to estimate the PS: gender, HLA-B27, BASDAI, ESR, mSASSS, NSAID index, disease duration, smoking pack-years and age of onset of symptoms.
§adjusted OR after PS matching (n=166 in each group); baseline variables used to estimate the PS: age, gender and CRP.
adjust, adjustment; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; BASFI, Bath Ankylosing Spondylitis Functional Index; BASMI, Bath Ankylosing Spondylitis Metrology Index; bDMARDs, biological disease-modifying antirheumatic drugs; CRP, C reactive protein; ESR, erythrocyte sedimentation rate; GESPIC, German Spondyloarthritis Inception Cohort; HLA, human leukocyte antigen; INF, infliximab; Match, matching; mSASSS, modified Stoke Ankylosing Spondylitis Spinal Score; NSAIDs, non-steroidal anti-inflammatory drugs; OASIS, Outcome in AS International Study; OLE, open-label extension; PGA, patient global assessment; PS, propensity score; strat, stratification; TNFi, tumour necrosis factor alpha inhibitor.