Background/Purpose: Real world studies comparing the
healthcare utilization of anti-TNFs vs. non-TNFs as the next alternative after
the failure of the first anti-TNF are scarce. This study aimed to compare healthcare
costs, resource utilization, and treatment patterns in RA patients who discontinued
an anti-TNF and subsequently switched to an alternative anti-TNF versus those
who switched to a non-TNF.

Methods: Adult patients who had ≥2
RA diagnoses and used at least one anti-TNF (adalimumab, etanercept,
infliximab, golimumab, or certolizumab) on or after initial RA diagnosis were
identified from a large commercial claims database. Patients who switched to an
alternative anti-TNF or a non-TNF (abatacept, rituximab, or tocilizumab) following
the initial anti-TNF during 2006-2012 were selected into anti-TNF and non-TNF
cohorts. The switching date was defined as the index date. Patient and disease characteristics
during the 6-month pre-index (baseline) period were compared. All-cause and
RA-related healthcare costs and resource utilization over a 12-month post-index
(study) period were compared between the two cohorts using multivariable two-part
gamma models and negative binomial models, respectively, adjusted for age, gender,
region, insurance plan, index year, comorbidities, co-medications, initial
anti-TNF and treatment duration. Treatment patterns (switching and
discontinuation) were compared using Cox proportional hazards models.

Results: A total of 2,904 and 934
patients were included in the anti-TNF and the non-TNF cohort, respectively. At
baseline, patients in the anti-TNF cohort (vs. non-TNF cohort) were younger (50
vs. 51 years, p<0.01), had less comorbidity burden (Charlson Comorbidity
Index 1.3 vs. 1.4, p<0.01), and had a shorter initial anti-TNF duration (1.0
vs. 1.3 years, p<0.01). During the study period, all-cause costs were
significantly lower for the anti-TNF cohort compared to the non-TNF cohort with
medical ($11,474 vs. $15,846; adjusted diff = $4,590; p<0.01) and pharmacy ($24,433
vs. $29,294; adjusted diff =$5,524; p<0.01) costs contributing to the total cost
differences ($35,907 vs. $45,139, adjusted diff = $9,654; p<0.01) (Table 1).
RA-related costs were also lower in the anti-TNF cohort compared to the non-TNF
cohort (Table 1). Patients in the anti-TNF cohort were likely to have fewer outpatient
visits (all-cause: 23.4 vs. 30.2 visits/patient/year, adjusted incident rate
ratio = 0.77; p<0.01). Inpatient admissions and emergency department utilization
as well as discontinuation and switching rates were not significantly different
between the two cohorts.

Conclusion: For patients discontinuing
anti-TNF therapy, switching to an alternative anti-TNF was associated with lower
medical and pharmacy costs, fewer outpatient visits, and similar treatment
patterns, as compared to switching to a non-TNF biologic.