Treatment effectiveness and treatment patterns among rheumatoid arthritis patients after switching from a tumor necrosis factor inhibitor to another medication

Figures & data

Figure 1 Study sample selection.

Note: ^aIn the 12-month pre-index period or the first 30 days post-index.
Abbreviations: DMARD, disease-modifying antirheumatic drug; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; RA, rheumatoid arthritis; TNFi, tumor necrosis factor inhibitor.

Table 1 Demographic and clinical characteristics of targeted DMARD switchers

	All patients (N=6,945)	TNFi cyclers (n=5,020)	Non-TNFi switchers (n=1,925)
Age (years), mean ± SD	49.9±9.6	49.6±9.7	50.9±9.3
Female, n (%)	5,673 (81.7)	4,058 (80.8)	1,615 (83.9)
DCCI, mean ± SD	1.4±0.8	1.4±0.8	1.5±0.9
Prior targeted DMARD, n (%)
Adalimumab	2,188 (31.5)	1,648 (32.8)	540 (28.1)
Certolizumab pegol	213 (3.1)	121 (2.4)	92 (4.8)
Etanercept	3,200 (46.1)	2,568 (51.2)	632 (32.8)
Golimumab IV	0 (0.0)	0 (0.0)	0 (0.0)
Golimumab SC	358 (5.2)	257 (5.1)	101 (5.2)
Infliximab	986 (14.2)	426 (8.5)	560 (29.1)
No of prior targeted DMARDs received at any time pre-index, n (%)a
One	6,418 (92.4)	4,729 (94.2)	1,689 (87.7)
Two	479 (6.9)	271 (5.4)	208 (10.8)
Three or more	48 (0.7)	20 (0.4)	28 (1.5)
Prior conventional synthetic DMARD use, n (%)
Hydroxychloroquine	1,578 (22.7)	1,173 (23.4)	405 (21.0)
Leflunomide	1,151 (16.6)	816 (16.3)	335 (17.4)
Methotrexate	3,863 (55.6)	2,864 (57.1)	999 (51.9)
Sulfasalazine	535 (7.7)	394 (7.8)	141 (7.3)
Most common comorbid conditions, n (%)
Hypertension	1,373 (19.8)	924 (18.4)	449 (23.3)
Benign hypertension	1,085 (15.6)	751 (15.0)	334 (17.4)
Pain in limb	979 (14.1)	688 (13.7)	291 (15.1)
Malaise and fatigue	904 (13.0)	643 (12.8)	261 (13.6)
Acute sinusitis	907 (13.1)	662 (13.2)	245 (12.7)
RA-related health care utilization, mean ± SD
No. of outpatient office visits	5.3±3.3	5.3±3.1	5.9±3.5
No. of pharmacy prescriptions	10.2±6.8	11.3±6.6	10.6±6.5

Note:

a Number of prior targeted DMARDs included any pre-index claim, without a time restriction.

Abbreviations: DCCI, Deyo–Charlson Comorbidity Index; DMARD, disease-modifying antirheumatic drug; IV, intravenous; RA, rheumatoid arthritis; SC, subcutaneous; SD, standard deviation; TNFi, tumor necrosis factor inhibitor.

Figure 2 Proportion of targeted DMARD switchers who switched therapy again.

Note: ^aP<0.001 compared with TNFi cyclers.
Abbreviations: DMARD, disease-modifying antirheumatic drug; TNFi, tumor necrosis factor inhibitor.

Table 2 Treatment persistence after the index switch

Treatment persistence	All patients (N=6,945)	TNFi cyclers (n=5,020)	Non-TNFi switchers (n=1,925)	P
Mean ± SD duration (days)	283.4±114.8	279.6±116.8	293.3±109.1	<0.001a
Persistent at 12 months, n (%)	4,110 (59.2)	2,920 (58.2)	1,190 (61.8)	<0.001b

Notes:

a Pairwise comparison of TNFi cyclers and non-TNFi switchers by t-test.

b Pairwise comparison of TNFi cyclers and non-TNFi switchers by chi-square test.

Abbreviations: SD, standard deviation; TNFi, tumor necrosis factor inhibitor.

Table 3 Treatment effectiveness for the 12 months after the index switch, using the claims-based algorithm

Algorithm criterion	Patients achieving the criterion, n (%)			Pa
	All patients (N=6,945)	TNFi cyclers (n=5,020)	Non-TNFi switchers (n=1,925)
Satisfied all six algorithm criteria	1,724 (24.8)	1,205 (24.0)	519 (27.0)	0.011
Adherent	2,809 (40.4)	2,006 (40.0)	803 (41.7)	0.182
No increase in the targeted	6,257 (90.1)	4,429 (88.2)	1,828 (95.0)	<0.001
DMARD dose
No new conventional synthetic	5,901 (85.0)	4,253 (84.7)	1,648 (85.6)	0.353
DMARD
No targeted DMARD switch	4,635 (66.7)	3,282 (65.4)	1,353 (70.3)	<0.001
No new/increased glucocorticoid	5,935 (85.5)	4,304 (85.7)	1,631 (84.7)	0.285
Less than two intra-articular injections	6,294 (90.6)	4,549 (90.6)	1,745 (90.6)	0.968

Note:

a Pairwise comparison of TNFi cyclers and non-TNFi switchers by chi-square test.

Abbreviations: DMARD, disease-modifying antirheumatic drug; TNFi, tumor necrosis factor inhibitor.