DMARDs exposure was evaluated for all participants. "Prior DMARDs treatment" included participants, who were treated with DMARDs 8 weeks and according to physician's discretion before being included in the study. "DMARDs treatment at baseline" included participants who were receiving DMARDs when they were included in the study and continued with this concomitant medication in addition to TCZ. Only those participants who received DMARDs prior to start of study and at Baseline up to 6 months were reported.

Objective intolerance was determined by medical observation; subjective intolerance was determined by the participant; lack of efficacy was determined by physician discretion.

Biological RA treatment exposure was evaluated for all participants. "Prior biological RA treatment" included participants who were treated with biological RA treatment 8 weeks before being included in the study. Percentage of participants who did not receive any biological RA treatment and percentage of participants who received one or more biologic RA treatments were reported.

Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment). Same participants may be counted in more than one previous biologic RA treatment category.

Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment).

Lack of efficacy was determined by physician discretion. Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment).

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. Only those participants who had dose modifications were reported.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. The mean dosing interval between different TCZ administrations (Adm) by participants within 6 months observational period was presented.

The safety variable measured the number of participants who discontinued TCZ due to adverse reactions to TCZ, and the efficacy variable measured the participants who discontinued from TCZ due to lack of efficacy according to criteria of the treating physician.

A participant's adherence was calculated based on the adverse event or laboratory abnormality experienced by the participants who required dose modifications as per local TCZ label or protocol.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. Only those participants who had TCZ as monotherapy were reported.

The number of tender joints was recorded on the joint assessment form, with no tenderness = 0, and tenderness = 1, for 68 joints, giving a total possible TJC score of 0 to 68.

The number of swollen joints was recorded on the joint assessment form, with no swelling = 0, and swelling =1, for 66 joints, giving a total possible SJC score of 0 to 66.

DAS28 was calculated from SJC and TJC using an assessment of 28 joints, the erythrocyte sedimentation rate (ESR) (milliliter per hour [mm/hr]), and Patient's Global Assessment (PGH) of disease activity (measured on a 0 to 100 mm Visual Analogue Scale [VAS] where 0=no disease activity and 100=worst disease activity). DAS28 was calculated using the following formula: DAS28 = 0.56*square root (sqrt) (TJC28) + 0.28*sqrt(SJC28) + 0.70*natural logarithm (ln) (ESR) + 0.014*PGH of disease activity. Total score range: 0-10, with a higher score indicated more disease activity. DAS28 <=3.2 implied low disease activity, DAS >3.2 to 5.1 implied moderate disease activity and DAS >5.1 implied high disease activity, and DAS28 <2.6 = clinical remission.

Clinical response assessed as per EULAR categorical DAS28 response criteria was defined as clinically meaningful improvement at a particular time point. EULAR response was based on change from baseline (CFB) in the DAS28 score and also on the actual DAS28 score at the time point so was more reflective of the current status of the participant. The DAS28 score was a measure of the participant's disease activity, based on the TJC (28 joints), SJC (28 joints), PGH (mm), and ESR (mm/hr). DAS28 total scores ranged from 0 to approximately 10. Scores <2.6 = best disease control and scores >5.1 = worse disease control. A negative CFB indicated clinically meaningful improvement. EULAR Good response: DAS28 <=3.2 and a CFB <-1.2. EULAR Moderate response: DAS28 >3.2 to ≤ 5.1 or a CFB < -0.6 to ≥ -1.2. EULAR No response: DAS28 ≤3.2 or CFB >=-0.6, DAS28 >3.2 to <=5.1 or CFB >=-0.6 and DAS28 >5.1 or CFB >=-0.6.

The SDAI was a combined index for measuring disease activity in RA which reflected the numerical sum of five outcome parameters: TJC and SJC based on a 28-joint assessment, PGH and physician's global assessment (PhGH) of disease activity, assessed on 0-100 mm VAS where 0 = no disease activity and 100 = worst disease activity, and C-reactive protein (CRP) (milligrams per deciliter [mg/dL]). SDAI total score = 0-86. A SDAI score of <=3.3 represented clinical remission, a score of <=11.0 represents low disease activity, a score of <=26.0 represented moderate disease activity and a score of >26.0 represented high (or severe) disease.

The CDAI was a combined index for measuring disease activity in RA and used to evaluate disease activity in the absence of laboratory testing of CRP and ESR. It was the numerical sum of 4 outcome parameters: TJC and SJC based on a 28-joint assessment, PGH and PhGH (assessed on 0-100 mm); VAS (0 = no disease activity and 100 = worst disease activity). CDAI total score = 0-76. A CDAI score of <=2.8 represented clinical remission, a score of <=10.0 represented low disease activity, a score of <=22.0 represented moderate disease activity and a score of >22.0 represented high (or severe) disease.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR20 response required at least a 20% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), Acute phase reactant (CRP or ESR). A reduction in the level of and acute phase reactants was considered an improvement.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR50 response required at least a 50% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: (participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), and acute phase reactant (CRP or ESR). A reduction in the level of acute phase reactants was considered an improvement.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR70 response required at least a 70% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: (participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), and acute phase reactant (CRP or ESR). A reduction in the level of acute phase reactants was considered an improvement.

The test for CRP was a laboratory measurement for evaluation of an acute phase reactant of inflammation through the use of an ultrasensitive assay. The serum concentration of CRP was measured in mg/dL. A reduction in the level was considered an improvement.

ESR was a laboratory test that provided a non-specific measure of inflammation. The test assessed the rate at which red blood cells fell in a test tube. ESR was measured in mm/hr. A reduction in the level was considered an improvement.

The PhGH was assessed using a 0 to 100 mm horizontal VAS by the physician. The left-hand extreme of the line equaled 0 mm, and was described as "no disease activity" (symptom-free and no arthritis symptoms) and the right-hand extreme equaled 100 mm, as "maximum disease activity" (maximum arthritis disease activity). A negative change from baseline indicated improvement.

The PGH of disease activity was assessed using a 0 to 100 mm horizontal VAS by the participant. The left-hand extreme of the line equaled 0 mm, and was described as "no disease activity" (symptom-free and no arthritis symptoms) and the right-hand extreme equaled 100 mm, as "maximum disease activity" (maximum arthritis disease activity). A negative change from baseline indicated improvement.

The HAQ-DI was a participant self-reported questionnaire for assessing the extent of a participant's functional ability. It consisted of 20 questions in 8 categories (dressing and grooming, rising, eating, walking, reach, grip, hygiene, and carrying out daily activities). Each question had 4 response options, ranging from "no difficulty" to "unable to do", corresponding to scores from 0 to 3. The HAQ-DI scale was an average of all the scores and ranged from 0 to 3, where higher scores represented higher disease activity. A score of <0.5 represented clinical remission. A participant achieved a clinically meaningful improvement in HAQ-DI if they had a reduction from baseline of >=0.22.

Fatigue was evaluated by a VAS. Participants marked on a 100 mm horizontal VAS the level of fatigue that they have experienced, ranging from 0 (no fatigue) to 100 (extreme fatigue).

Severity of pain was evaluated by a VAS. Participants marked on a 100 mm horizontal VAS the severity of pain that they had experienced because of their RA, ranging from 0 (no pain) to 100 (unbearable pain). A decrease of 10 points was considered clinically meaningful.

The FACIT Measurement System was a collection of health-related quality of life questionnaires targeted to the management of chronic illness and included questions on Physical Well-Being, Social/Family Well-Being, Emotional Well-Being and Functional Well-Being. The FACIT Fatigue Scale was a 13-item tool that measured an individual's level of fatigue during their usual daily activities over the past week. The level of fatigue was measured on a five-point Likert scale (4 = not at all fatigued to 0 = very much fatigued). The total score of FACIT ranged from 0 to 160. An increase of 4 points in the FACIT-Fatigue score was considered clinically meaningful. Change from Administration 1 was reported for individual administration schedules. TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

Morning stiffness was defined by the time elapsed between the time of usual awakening (even if not in the morning) and the time the participant was as limber as he/she would be during a day involving typical activities. Morning stiffness was assessed on a 100 mm VAS, where 0= none and 100= very severe.

Systemic manifestations included fibromyalgia, osteoporosis, sjogren's syndrome, anemia, rheumatoid nodules, pulmonary fibrosis, vasculitis, peripheral neuropathy or mononeuropathy, scleritis, episcleritis, hypertension, hyperlipidemia, low high-density lipoproteins, diabetes type 1 and type 2, metabolic syndrome, carotid artery disease, transient ischemic attacks, embolic stroke, atherothrombotic stroke, atrial fibrillation, heart failure New York Heart Association (NYHA) Class l/ll, coronary heart disease(angina pectoris/myocardial), aortic aneurism, peripheral arterial occlusive disease, percutaneous coronary interventions, coronary artery bypass, carotid endarterectomy and peripheral arterial bypass.

FAS population.

DMARDs exposure was evaluated for all participants. "Prior DMARDs treatment" included participants, who were treated with DMARDs 8 weeks and according to physician's discretion before being included in the study. "DMARDs treatment at baseline" included participants who were receiving DMARDs when they were included in the study and continued with this concomitant medication in addition to TCZ. Only those participants who received DMARDs prior to start of study and at Baseline up to 6 months were reported.

FAS population.

Objective intolerance was determined by medical observation; subjective intolerance was determined by the participant; lack of efficacy was determined by physician discretion.

FAS population.

Biological RA treatment exposure was evaluated for all participants. "Prior biological RA treatment" included participants who were treated with biological RA treatment 8 weeks before being included in the study. Percentage of participants who did not receive any biological RA treatment and percentage of participants who received one or more biologic RA treatments were reported.

FAS population.

Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment). Same participants may be counted in more than one previous biologic RA treatment category.

FAS population.

Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment).

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

Lack of efficacy was determined by physician discretion. Previous biologic RA treatment included adalimumab, infliximab, golimumab, etanercept, certolizumab, and other (any other previous biologic RA treatment).

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. Only those participants who had dose modifications were reported.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. The mean dosing interval between different TCZ administrations (Adm) by participants within 6 months observational period was presented.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

The safety variable measured the number of participants who discontinued TCZ due to adverse reactions to TCZ, and the efficacy variable measured the participants who discontinued from TCZ due to lack of efficacy according to criteria of the treating physician.

FAS population.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

A participant's adherence was calculated based on the adverse event or laboratory abnormality experienced by the participants who required dose modifications as per local TCZ label or protocol.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations. Only those participants who had TCZ as monotherapy were reported.

FAS population. Here, "n"= participants who were evaluable for each category.

The number of tender joints was recorded on the joint assessment form, with no tenderness = 0, and tenderness = 1, for 68 joints, giving a total possible TJC score of 0 to 68.

FAS population. Here, number of participants analyzed = participants with valid data to calculate TJC at the end of study (6 months).

The number of swollen joints was recorded on the joint assessment form, with no swelling = 0, and swelling =1, for 66 joints, giving a total possible SJC score of 0 to 66.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

DAS28 was calculated from SJC and TJC using an assessment of 28 joints, the erythrocyte sedimentation rate (ESR) (milliliter per hour [mm/hr]), and Patient's Global Assessment (PGH) of disease activity (measured on a 0 to 100 mm Visual Analogue Scale [VAS] where 0=no disease activity and 100=worst disease activity). DAS28 was calculated using the following formula: DAS28 = 0.56*square root (sqrt) (TJC28) + 0.28*sqrt(SJC28) + 0.70*natural logarithm (ln) (ESR) + 0.014*PGH of disease activity. Total score range: 0-10, with a higher score indicated more disease activity. DAS28 <=3.2 implied low disease activity, DAS >3.2 to 5.1 implied moderate disease activity and DAS >5.1 implied high disease activity, and DAS28 <2.6 = clinical remission.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

Clinical response assessed as per EULAR categorical DAS28 response criteria was defined as clinically meaningful improvement at a particular time point. EULAR response was based on change from baseline (CFB) in the DAS28 score and also on the actual DAS28 score at the time point so was more reflective of the current status of the participant. The DAS28 score was a measure of the participant's disease activity, based on the TJC (28 joints), SJC (28 joints), PGH (mm), and ESR (mm/hr). DAS28 total scores ranged from 0 to approximately 10. Scores <2.6 = best disease control and scores >5.1 = worse disease control. A negative CFB indicated clinically meaningful improvement. EULAR Good response: DAS28 <=3.2 and a CFB <-1.2. EULAR Moderate response: DAS28 >3.2 to ≤ 5.1 or a CFB < -0.6 to ≥ -1.2. EULAR No response: DAS28 ≤3.2 or CFB >=-0.6, DAS28 >3.2 to <=5.1 or CFB >=-0.6 and DAS28 >5.1 or CFB >=-0.6.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The SDAI was a combined index for measuring disease activity in RA which reflected the numerical sum of five outcome parameters: TJC and SJC based on a 28-joint assessment, PGH and physician's global assessment (PhGH) of disease activity, assessed on 0-100 mm VAS where 0 = no disease activity and 100 = worst disease activity, and C-reactive protein (CRP) (milligrams per deciliter [mg/dL]). SDAI total score = 0-86. A SDAI score of <=3.3 represented clinical remission, a score of <=11.0 represents low disease activity, a score of <=26.0 represented moderate disease activity and a score of >26.0 represented high (or severe) disease.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The CDAI was a combined index for measuring disease activity in RA and used to evaluate disease activity in the absence of laboratory testing of CRP and ESR. It was the numerical sum of 4 outcome parameters: TJC and SJC based on a 28-joint assessment, PGH and PhGH (assessed on 0-100 mm); VAS (0 = no disease activity and 100 = worst disease activity). CDAI total score = 0-76. A CDAI score of <=2.8 represented clinical remission, a score of <=10.0 represented low disease activity, a score of <=22.0 represented moderate disease activity and a score of >22.0 represented high (or severe) disease.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR20 response required at least a 20% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), Acute phase reactant (CRP or ESR). A reduction in the level of and acute phase reactants was considered an improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR50 response required at least a 50% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: (participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), and acute phase reactant (CRP or ESR). A reduction in the level of acute phase reactants was considered an improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

ACR response was calculated based on total joint count evaluation (28 or 66/68 joint count) and other clinical and laboratory assessments. A positive ACR70 response required at least a 70% improvement (reduction) compared to baseline in swollen joint count (66 joints) and tender joint count (68 joints) and at least 3 of the following 5 assessments: (participant's global assessment of pain, PGH, PhGH (all 3 assessed at 0 [good] to 100 mm [worst] VAS scale), participant assessment of disability measured by the Health Assessment Questionnaire-Disability Index (HAQ-DI) (assessed on a 0 to 3 scale, where higher scores represented higher disease activity), and acute phase reactant (CRP or ESR). A reduction in the level of acute phase reactants was considered an improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The test for CRP was a laboratory measurement for evaluation of an acute phase reactant of inflammation through the use of an ultrasensitive assay. The serum concentration of CRP was measured in mg/dL. A reduction in the level was considered an improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

ESR was a laboratory test that provided a non-specific measure of inflammation. The test assessed the rate at which red blood cells fell in a test tube. ESR was measured in mm/hr. A reduction in the level was considered an improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The PhGH was assessed using a 0 to 100 mm horizontal VAS by the physician. The left-hand extreme of the line equaled 0 mm, and was described as "no disease activity" (symptom-free and no arthritis symptoms) and the right-hand extreme equaled 100 mm, as "maximum disease activity" (maximum arthritis disease activity). A negative change from baseline indicated improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The PGH of disease activity was assessed using a 0 to 100 mm horizontal VAS by the participant. The left-hand extreme of the line equaled 0 mm, and was described as "no disease activity" (symptom-free and no arthritis symptoms) and the right-hand extreme equaled 100 mm, as "maximum disease activity" (maximum arthritis disease activity). A negative change from baseline indicated improvement.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The HAQ-DI was a participant self-reported questionnaire for assessing the extent of a participant's functional ability. It consisted of 20 questions in 8 categories (dressing and grooming, rising, eating, walking, reach, grip, hygiene, and carrying out daily activities). Each question had 4 response options, ranging from "no difficulty" to "unable to do", corresponding to scores from 0 to 3. The HAQ-DI scale was an average of all the scores and ranged from 0 to 3, where higher scores represented higher disease activity. A score of <0.5 represented clinical remission. A participant achieved a clinically meaningful improvement in HAQ-DI if they had a reduction from baseline of >=0.22.

ITT population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

Fatigue was evaluated by a VAS. Participants marked on a 100 mm horizontal VAS the level of fatigue that they have experienced, ranging from 0 (no fatigue) to 100 (extreme fatigue).

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

Severity of pain was evaluated by a VAS. Participants marked on a 100 mm horizontal VAS the severity of pain that they had experienced because of their RA, ranging from 0 (no pain) to 100 (unbearable pain). A decrease of 10 points was considered clinically meaningful.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

The FACIT Measurement System was a collection of health-related quality of life questionnaires targeted to the management of chronic illness and included questions on Physical Well-Being, Social/Family Well-Being, Emotional Well-Being and Functional Well-Being. The FACIT Fatigue Scale was a 13-item tool that measured an individual's level of fatigue during their usual daily activities over the past week. The level of fatigue was measured on a five-point Likert scale (4 = not at all fatigued to 0 = very much fatigued). The total score of FACIT ranged from 0 to 160. An increase of 4 points in the FACIT-Fatigue score was considered clinically meaningful. Change from Administration 1 was reported for individual administration schedules. TCZ was administered every 4 weeks according to the label. Due to the observational nature of the study, the suggested schedule was subject to changes according to physician and participant considerations.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure and "n"= participants who were evaluable for each category.

Morning stiffness was defined by the time elapsed between the time of usual awakening (even if not in the morning) and the time the participant was as limber as he/she would be during a day involving typical activities. Morning stiffness was assessed on a 100 mm VAS, where 0= none and 100= very severe.

FAS population. Here, number of participants analyzed = participants who were evaluable for this outcome measure.

FAS population