Table 6.

Type of protocol deviation

Type	Number of deviations		Total	P -value #
	Criteria-based	DAS28-based
Continued rather than escalating
Toxicity	33	26	59
Patient-related	92	63	155
Physician-related	83	60	143
Comorbidity	79	42	121
Other*	42	17	59
Total, n (%)	329 (58.1)	208 (63.0)	537 (59.9)	0.005
Tapered rather than continuing/escalating
Toxicity	31	22	53
Patient-related	25	13	38
Physician-related	11	12	23
Comorbidity	4	3	7
Other*	1	0	2
Total, n (%)	72 (12.7)	50(15.2)	122 (13.6)	0.077
Discontinued rather than continuing/escalating
Toxicity	68	28	96
Patient-related	20	5	25
Physician-related	0	0	0
Comorbidity	37	14	51
Other	0	0	0	0.076
Total, n (%)	125 (22.1)	47 (14.2)	172 (19.2)
Dose escalated rather than continuing
Toxicity	0	0	0
Patient-related	2	0	2
Physician-related	29	20	49
Comorbidity	0	0	0
Other*	2	1	3
Total, n (%)	33 (5.8)	21 (6.4)	54 (6.1)	0.402
Other**
Toxicity	1	0	1
Patient-related	0	1	1
Physician-related	3	3	6
Comorbidity	0	0	0
Other*	3	0	3
Total, n (%)	7 (1.3)	4 (1.2)	11 (1.2)	0.843
Total deviation	566	330	896	0.016

*Examples of other type of protocol violation include awaiting laboratory results, awaiting approval of biological disease modifying anti-rheumatic drugs, prophylaxis for latent tuberculosis before initiating biologics, and other logistic reasons or unknown reason. ^#Statistical tests based on the chi-square test compared the frequency of responses (clinic visits with deviation versus no deviation) between patient groups due to specific types of deviation. **Unknown, addition of very high unusual dose or intensification of two or more drugs at a time, intensifying instead of tapering/discontinuing.