Article Text

FRI0416 Remission definitions affect cost savings estimates in anti-TNF treated rheumatoid arthritis patients
  1. C. Barnabe1,
  2. N.X. Thanh2,
  3. A. Ohinmaa2,
  4. S.G. Barr1,
  5. J. Homik3,
  6. L. Martin1,
  7. W.P. Maksymowych3
  1. 1Medicine, University of Calgary, Calgary
  2. 2Institute for Health Economics
  3. 3Medicine, University of Alberta, Edmonton, Canada


Background Multiple remission definitions have been proposed for rheumatoid arthritis (RA), including the 2011 ACR/EULAR Boolean Definition, SDAI ≤3.3, CDAI ≤2.8, and DAS28 ≤2.6. The stringency of the definition affects the proportion of patients classified as being in remission. We hypothesized that the remission definition used will also affect the estimates of cost-savings observed in anti-TNF treated patients.

Objectives To compare total healthcare costs, RA-attributable costs, and non-RA attributable costs across different definitions of remission.

Methods We determined healthcare utilization (physician visits, outpatient department visits, and hospitalizations) from a provincial administrative database (years 2004-2009) for a prospective cohort of anti-TNF treated RA patients (n=1,086, mean age 54 years). We compared the mean annual costs (standardized to 2008 Canadian dollars) for individuals achieving a minimum 1 year period of remission for each definition above. A propensity score matching technique was used to account for confounding by individual variables affecting healthcare utilization (including therapy received, baseline function, smoking, age, sex, disease duration and medical comorbidities).

Results The proportion of patients attaining a 1 year remission period by the DAS28 definition was 16%, compared to 9% for the ACR/EULAR Boolean definition, 6% for the CDAI definition, and 4% for the SDAI definition. The mean reduction in HAQ score was similar across remission definitions, (0.85 for DAS28 remission, 0.80 for ACR/EULAR remission, 0.60 for SDAI remission, and 0.84 for CDAI remission). The estimated cost savings observed varied by the remission definition that was used. The mean annual total cost difference between patients attaining SDAI remission relative to those that did not was estimated at $1904 (95%CI 112,2964), compared to $1630 (95%CI 1016,2508) for DAS28 remission, and $1266 (95%CI 517,2138) for the ACR/EULAR Boolean definition. The difference in cost for for those patients achieving CDAI remission relative to those that did not was not significant at $430 (95%CI -3319, 2214). These savings occurred predominantly in non-RA attributable costs for all definitions except the CDAI (Table 1).

Table 1. Mean Annual Healthcare Savings (95%Confidence Interval) for Patients Achieiving Remission, by Remission Definition, standardized to 2008 Canadian Dollars

Conclusions The magnitude of the cost savings observed with anti-TNF treatment varies according to the remission definition used in classifying patient disease status. The best cost-savings was observed in patients attaining SDAI remission, and the least was with the CDAI. Improvements in physical function likely contribute to these savings. Cost-effectiveness evaluations should consider various remission definitions in their analyses.

Disclosure of Interest None Declared

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.