Article Text

OP0289 Comorbidity with Depression and Anxiety is Associated with Discontinuation of Tumour Necrosis Factor Inhibitor (TNFI) Therapy in Ankylosing Spondylitis (AS)
  1. E. Lie1,2,
  2. U. Lindström1,
  3. L.E. Kristensen3,
  4. H. Forsblad-d'Elia1,
  5. J. Askling4,
  6. L.T. Jacobsson1
  7. on behalf of the ARTIS Study Group
  1. 1Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
  2. 2Diakonhjemmet Hospital, Oslo, Norway
  3. 3The Parker Institute, Copenhagen, Denmark
  4. 4Karolinska Institutet, Stockholm, Sweden


Background Studies using validated questionnaires have shown a high prevalence of depression and/or anxiety in patients (pts) with AS, and such comorbidity has been found to be associated with AS outcome measures such as Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and Functional Index (BASFI). A study of pts with rheumatoid arthritis showed that depression and anxiety were associated with poorer TNFi drug survival1. To our knowledge, no such data has been published for AS.

Objectives To assess whether comorbidity with depression and/or anxiety has an impact on drug survival in pts with AS treated with their first TNFi, and secondarily, to explore whether such comorbidity had a particular influence on discontinuations due to primary inefficacy.

Methods The Swedish biologics register ARTIS provided data on previously biologics naïve pts with AS starting TNFi treatment from 1 July 2006 to 31 Dec 2010. Data on anxiety/depression were provided through linkage to the National Patient Register (NPR) and the Prescribed Drug Register (PDR): Depression/anxiety comorbidity was defined as ≥1 ICD-10 code F30-34 or F38-41 registered in the NPR and/or a dispensed prescription of drug classes N06A (Antidepressants) or N05AN (Lithium) registered in the PDR during the year prior to TNFi start. We assessed 3-year drug survival by uni- and multivariable Cox regression analysis (covariates: age, sex, TNFi type, TNFi start year, DMARD co-medication, income, education, civil status, number of hospital days prior 2 yrs, diabetes, uveitis, IBD, psoriasis). Variables with p<0.25 in the univariable analyses were included in multivariable analysis with stepwise backward elimination.

Results 965 pts with AS were included. 123 pts (12.7%) met the above criteria for depression/anxiety; these pts were more often female and had lower baseline levels of ESR and CRP, but higher levels of patient global and pain. The univariable HR (95% CI) for TNFi discontinuation over 3 years was 1.78 (1.35-2.35) for pts with vs without depression/anxiety (p<0.001). The corresponding HR for discontinuation due to primary inefficacy was 2.71 (1.63-4.50) (p<0.001). In the multivariable analysis the risk of discontinuing the first TNFi over 3 years was greater in pts with depression/anxiety with a HR of 1.53 (1.15-2.04) (p=0.003; Figure). Sex, DMARD co-medication, type of TNFi and uveitis were the other statistically significant predictors. In the secondary analysis of primary inefficacy the effect of depression/anxiety was more pronounced, with a HR of 2.39 (1.41-4.05) (p<0.001). Higher age, lower level of education, having no history of uveitis and not being married were also independently associated with primary inefficacy discontinuation.

Conclusions In this large register-based study, we found that comorbidity with depression and/or anxiety was associated with poorer TNFi drug survival in pts with AS, and the association with primary inefficacy discontinuations was particularly strong. Depression/anxiety may have influenced the pts' symptoms before and during TNFi treatment, including aggravation of back symptoms that are not caused by inflammation and thus not likely to improve upon TNFi therapy.


  1. Mattey DL, et al. J Rheumatol. 2010;37(10):2021-4.

Disclosure of Interest E. Lie Consultant for: AbbVie, Bristol-Myers Squibb, Hospira, Pfizer, UCB, Speakers bureau: AbbVie, Bristol-Myers Squibb, U. Lindström: None declared, L. E. Kristensen: None declared, H. Forsblad-d'Elia: None declared, J. Askling: None declared, L. Jacobsson Consultant for: Pfizer, UCB, AbbVie, Speakers bureau: Pfizer, UCB, AbbVie

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.