Background In Denmark approximately 0.7% (35,000) of the population is diagnosed with rheumatoid arthritis (RA). RA is a risk factor in the development of comorbidity, and comorbidities are not well managed in RA patients. AS well as being a first-line treatment of early RA, methotrexate (MTX) gives a 70% reduction in cardiovascular disease-caused mortalities, and if treatment exceeds 1 year, the general mortality risks are lowered by 60%. Discontinuation of MTX is therefore a bad outcome for RA. It remains unclear whether side-effects and treatment response to MTX is associated with comorbidity in early RA.

Purpose To evaluate the association between comorbidity and persistence to MTX treatment and side-effects for RA patients.

Material and methods Patient files from three centres were evaluated retrospectively. Inclusion criteria were: diagnosis obtained according to ACR/EULAR 2010 criteria for RA in the period 1 January 2010 to the present, and MTX as a first line of treatment. Medical records were reviewed for side-effects, dose changes of MTX, formulation changes and persistence. Comorbidities and comedication was evaluated by usage of the Danish National Patient Registry (DNPR) and the Odense Pharmacoepidemiological Database (OPED). Comorbidities were scored according to the Charlson Comorbidity Index (CCI), and analysed by the Cox proportional hazards model for discontinuation of MTX treatment and dose reduction.

Results Five hundred and one patients were screened, 177 were eligible and analysed at baseline for disease characteristics, medication besides MTX and comorbidities in a 5 year window before RA diagnosis baseline. The highest risk of MTX discontinuation was a CCI of 3–4, they had crude 4.18 (95% CI: 1.67 to 10.45) increased risk compared to the reference group (RA with no comorbidities). Risk of dosage reduction was highest at CCI 1–2: 1.38 (95% CI: 0.72 to 2.62). A CCI of 5 or higher gave a −4.83 mg (95% CI: −10.24 to −0.59) adjusted difference in maximum weekly tolerable MTX dosage. Side-effects occurred for 23.7%. Most likely dosage causing side-effects was 20 mg (IQR 15–20 mg). Nausea occurred in 29% and hepatic events in 21%.

Conclusion Patients with CCI in the range of 3–4 had an increased risk for discontinuing MTX treatment.

References and/or Acknowledgements Acknowledgements are given to all my co-authors for great encouragement and supervision.

No conflict of interest