“…While our design did not account for all possible unmeasured confounders, when the set of measured demographics were compared for the two groups, no significant differences were present. In addition, our cohort was similar to the validation cohort (Hill et al, 1992) and other inflammatory arthritis cohorts (Bykerk et al, 2012;Pappas et al, 2015) that reported similar results. Taken together, these findings add credence to our interpretation that, despite the potential existence of uncontrolled confounders, there is a positive relationship between collaborative care models and patient satisfaction.…”