An Global Comparative Observational Study of RoActemra/Actemra (Tocilizumab) in Patients With Rheumatoid Arthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Global Comparative Observational Study of RoActemra/Actemra (Tocilizumab) in Patients With Rheumatoid Arthritis (NCT01543503), a registered phase not stated study. The primary outcome is Mean Change From Baseline in Calculated Disease Activity Score Based on 28 Joint Count Erythrocyte Sedimentation Rate at Week 24.

2. Clinical Question

The clinical question is whether disease activity changes over time among people with rheumatoid arthritis starting tocilizumab or a TNF inhibitor as their first biologic therapy, and how those changes compare in the observational data. The primary endpoint is change from baseline in the calculated Disease Activity Score based on 28 joints and erythrocyte sedimentation rate (DAS28-ESR) at Week 24. The registry also describes a broader assessment of efficacy and safety over 52 weeks.

This is a comparison between treatment cohorts, not a randomized comparison. The registry does not state how treatment allocation occurred, and the cohort descriptions do not provide further detail about the groups. As a result, any difference in observed outcomes could reflect treatment as well as differences between the people receiving each treatment or other features of care. The posted record does not provide numerical primary endpoint results, so it is not possible to describe the magnitude or precision of the observed mean change or a contrast between cohorts from the registry information alone.

3. Trial Design

This was a prospective, multicenter, observational study of people with rheumatoid arthritis treated with a TNF inhibitor or tocilizumab as their first biologic therapy. The registry describes a global comparative study and an intended data-collection period of 52 weeks. The record lists enrollment of 1225 participants, a start date of 2012-02, primary completion in 2015-02, and a completed status. Hoffmann-La Roche is listed as the sponsor.

The study compares treatment cohorts, but it is not a randomized trial. The allocation method is not stated in the registry record. The arm descriptions do not add information about how clinicians or participants selected a treatment, nor do they describe other treatment details. Therefore, the group comparison should be read as observational: the record does not establish that the groups were comparable at baseline or that treatment choice was independent of participants’ clinical characteristics.

The primary outcome is a continuous measure: mean change from baseline in calculated DAS28-ESR at Week 24. The registry lists secondary outcomes including DAS28-ESR change at Week 52, changes in erythrocyte sedimentation rate, C-reactive protein, swollen and tender joint counts, Clinical Disease Activity Index, Simplified Disease Activity Index, and physician global assessment. Loss of efficacy or development of intolerance to biologic therapy is also listed.

The primary endpoint description defines an effectiveness analysis population drawn from the safety population. It includes participants whose first biologic administration occurred within 60 days after their last rheumatoid arthritis disease activity assessment; data available at the time of assessment were included. This definition means the endpoint population is not simply synonymous with everyone enrolled. Further details about recruitment, treatment selection, follow-up procedures, and the analysis model are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 1250 participants were screened for entry into the study, with 1225 participants enrolled in the study. One participant whose randomization status was unknown withdrew informed consent.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants belonging to the safety population who had first biologic administration within 60 days after the last Rheumatoid Arthritis (RA) disease activity assessment were included in the effectiveness analysis population. Data of participants available at the time of the assessment were included in the analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Baseline characteristics were described for the safety population. All enrolled participants who received at least one dose of a TNF inhibitor or tocilizumab during the study were included in the safety population.

6. Primary Endpoint

The registry lists the primary outcome as Mean Change From Baseline in Calculated Disease Activity Score Based on 28 Joint Count Erythrocyte Sedimentation Rate at Week 24, measured over Baseline and Week 24.

The primary endpoint is mean change from baseline in calculated DAS28-ESR at Week 24, with baseline and Week 24 named as the assessment times. DAS28 is a composite measure of rheumatoid arthritis disease activity. It combines swollen joint count and tender joint count across 28 joints, the patient’s global assessment of disease activity using a visual analogue scale, and an acute-phase inflammatory marker. For this study, erythrocyte sedimentation rate was used whenever possible to calculate the score.

A composite measure can summarize several dimensions of disease activity in one outcome. It may be useful for tracking how the overall score changes between assessments, while its components represent distinct clinical measurements. The registry states that higher DAS28 scores indicate greater disease activity. It identifies a score of <= 3.2 as low disease activity and a score >3.2 to 5.1 as moderate to high disease activity. Those are descriptions of score levels; they should not be mistaken for thresholds defining a meaningful change from baseline.

The endpoint is reported as a least-squares mean with a 95% confidence interval in units on a scale. The numerical mean change and confidence interval are not reported in the registry record. Nor does the record specify a threshold for interpreting the size of a change as clinically important. Thus, the endpoint definition clarifies what was intended to be measured, but the record does not permit assessment of the observed magnitude or its precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

For a continuous outcome, a mean describes the central value of measurements on a numeric scale. A change-from-baseline outcome compares a participant’s later measurement with the baseline measurement, and the group mean summarizes those participant-level changes. The primary endpoint uses this approach for DAS28-ESR at Week 24. In general, change scores can help express how measurements evolve from an initial assessment, but interpretation depends on the underlying scale, the population measured, and the analysis method.

The registry labels the endpoint estimate a least-squares mean. In general, a least-squares mean is a model-based estimate of a group mean rather than necessarily the raw arithmetic average. Such estimates can account for terms included in a statistical model. The registry record does not report the model specification, the covariates included, or whether and how baseline values were incorporated. It is therefore not possible to determine from the record which adjustments contributed to the reported estimate—or to evaluate such an estimate, since its numerical value is not reported.

A 95% confidence interval describes uncertainty around an estimated quantity under the assumptions of the method used. In general, a narrower interval indicates greater precision than a wider interval, while the interval’s location gives context for the plausible values compatible with the analysis. It is not a guarantee that the true value lies within the particular interval. Here, the endpoint is described as having a 95% confidence interval, but the interval limits are not reported in the registry record. Its precision and compatibility with different possible changes cannot be assessed.

The primary outcome is calculated from a composite score. The registry describes the score as incorporating swollen and tender joint counts, a patient global assessment, and an acute-phase marker, using erythrocyte sedimentation rate whenever possible. In general, composite scores can provide a single summary while combining measurements that capture different aspects of a condition. That convenience also means a composite result does not reveal whether each component moved in the same direction. Component results are not reported for the primary endpoint in the registry record.

Over-dispersion is a general concern when observed variability is greater than a statistical model assumes. If a model understates variability, its uncertainty estimates may be too small and inference may be overly confident. The registry record does not state whether over-dispersion was assessed or whether any special variance method was used. This issue cannot be evaluated from the outcome description alone.

Comparisons in observational studies require particular care. In general, random assignment can help balance measured and unmeasured factors between groups, whereas treatment choice in an observational study may be related to participants’ characteristics or clinical circumstances. The registry states that allocation was not stated and cautions that the arms are not a like-for-like control comparison. It does not report a causal adjustment strategy, matching procedure, weighting method, or covariate balance assessment. Those methods should not be assumed to have been used.

The record lists sample-size planning illustrations as based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These are planning assumptions, not observed power, and they do not establish the study’s original design rationale. The numerical planning results and the observed effect used for those illustrations are not reported in the registry record. For time-to-event planning, the record describes total events using the Schoenfeld approach; for other planning illustrations, it describes participants per arm. Such calculations are hypothetical planning aids and do not replace observed results or resolve the limitations of an observational comparison.

Statistical significance and practical importance are different ideas in general. A p-value, when reported, summarizes how compatible the data are with a specified null model under the analysis assumptions; it does not measure the probability that a treatment works. An estimate and its uncertainty interval are needed to understand magnitude and precision. The registry record does not report a primary endpoint estimate, confidence limits, or a p-value, and it does not report an interpretation threshold for meaningful change.

What can be concluded from the method description?

The method description identifies the outcome, its scale, the time point, a least-squares mean, and a confidence-interval format. It does not identify the full statistical model or provide the numerical result. Thus, the analysis approach cannot be reproduced or fully appraised from the registry record.

Why does the design matter?

Because participants were observed in treatment cohorts rather than assigned at random, a difference in group outcomes would not by itself show that treatment caused the difference. Interpretation would require details about the groups, confounding control, outcome availability, and the analysis model. Those details are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures and 25 analyses. Multiple outcomes and analyses create a general multiplicity concern: the more comparisons examined, the greater the chance that some findings may appear noteworthy by chance if results are interpreted without accounting for the number of analyses. A prespecified primary endpoint can help focus interpretation, but the record does not report a multiplicity adjustment, a testing hierarchy, or other plan for controlling the interpretation of multiple results. Whether any adjustment was used is not reported in the registry record.

The effectiveness analysis population is defined as participants in the safety population whose first biologic administration occurred within 60 days after the last rheumatoid arthritis disease activity assessment. The record says that data available at the time of assessment were included. It does not report the number of participants contributing to the primary analysis, the amount of missing outcome data, or the reasons data were unavailable.

How missing data were handled is not reported in the registry record. In general, missing-data methods can affect estimates, particularly when the chance of a measurement being unavailable is related to health status or treatment experience. The registry does not describe imputation, a complete-case approach, sensitivity analyses, or assumptions about missingness. Those methods should not be inferred from the statement that data available at assessment were included.

The record also does not report whether the analysis population definition was finalized before results were examined, or whether any participants were excluded for other reasons. These omissions limit assessment of how the enrolled population relates to the people whose outcomes contributed to the primary estimate. The treatment of protocol deviations, follow-up discontinuations, and repeated assessments is not reported in the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: lower respiratory tract infection (0.8%), pneumonia (0.8%), pregnancy (0.5%), chest pain (0.4%), sepsis (0.4%).

The registry reports serious adverse events by treatment cohort. In the tocilizumab group, 22 of 423 participants were listed as affected; in the TNF inhibitor group, 64 of 793 were listed as affected. These counts describe the posted record and should not be read as proof that one treatment caused more or fewer serious events. This was an observational comparison, and the record does not provide enough information about group comparability or event ascertainment to support a causal interpretation.

The record also lists other adverse-event counts: 132 of 423 participants in the tocilizumab group and 231 of 793 in the TNF inhibitor group. The registry does not provide further detail about how those events were defined or their severity, so they should be kept distinct from the serious-event counts rather than treated as interchangeable measures.

Serious event terms listed with percentages include lower respiratory tract infection (0.8%), pneumonia (0.8%), pregnancy (0.5%), chest pain (0.4%), sepsis (0.4%), and osteoarthritis (0.4%). The registry does not report arm attribution for these terms in the listed information, so they cannot be assigned to a particular treatment cohort from these figures. Deaths are not reported in the registry record.

These figures are descriptive safety information, not a controlled estimate of comparative risk. The record does not give details about event follow-up, exposure duration by cohort, or statistical comparisons for safety outcomes.

12. What the Result Does, and Does Not, Mean

The registry record establishes that an observational study compared people with rheumatoid arthritis receiving tocilizumab or a TNF inhibitor as first biologic therapy. It identifies mean change in calculated DAS28-ESR at Week 24 as the primary endpoint and describes a 52-week data-collection period. It also provides selected safety counts and names several secondary outcomes. These details clarify what the study set out to assess, but they do not supply the numerical primary endpoint result.

Without the primary mean change and its 95% confidence interval, the direction and size of the Week 24 change cannot be evaluated from the registry record. There is also no reported numerical treatment-group contrast for that endpoint. It is therefore not possible to assess the precision of the estimate, how the cohorts differed on the outcome, or whether the observed result was compatible with a range of effect sizes. A stated endpoint is not itself evidence about its result.

Even if numerical differences were available, the nonrandomized design would matter. Treatment allocation is not stated, and the registry warns that the cohorts are not a like-for-like control comparison. In general, observational group differences can reflect who received each treatment and how care was delivered, as well as treatment effects. The registry does not describe enough about confounding control to establish that a group difference would be attributable to treatment.

The safety counts show that serious and other adverse events were recorded in both cohorts. They do not, on their own, establish a comparative safety effect. The record does not report deaths or provide sufficient detail about event ascertainment and exposure to support broader safety conclusions.

Accordingly, the posted information is best understood as a partial registry account of study design, outcomes, and selected safety data. It does not support a claim that either treatment is superior or inferior, nor does it establish a causal effect. The missing numerical efficacy results and analysis details leave key questions about magnitude, precision, and comparability unanswered.

13. Why This Trial Matters Statistically

This study is statistically informative as an example of the distinction between a comparative study and a randomized comparison. The title and objectives describe a comparison of two first-biologic treatment cohorts, but the registry classifies the study as observational and does not state allocation. That distinction affects the interpretation of any group contrast: comparisons can describe what was observed while remaining vulnerable to differences between the people receiving each treatment.

The record also highlights why endpoint definitions and numerical results are both needed. It defines a clinically interpretable composite disease-activity measure, identifies the Week 24 assessment, and states the form of the estimate and uncertainty interval. But the numerical mean change and interval are not reported in the registry record. Without those results, readers cannot assess effect magnitude or precision, even though the outcome is clearly specified.

The population definition is another important statistical feature. The primary effectiveness population is drawn from the safety population and uses a timing criterion relative to the last disease activity assessment. That makes it important to distinguish total enrollment from the number contributing to an endpoint analysis; the latter is not reported. The record also states that available assessment data were included, while leaving missing-data methods unspecified.

Finally, the registry lists numerous outcomes and analyses but does not report how multiplicity was handled. In general, many analyses can complicate interpretation if attention is placed selectively on findings. Taken together, the record illustrates that clear outcome labels, design descriptions, analysis populations, numerical estimates, and uncertainty measures all contribute to a statistical assessment. Some of those elements are present, while the result and several analytic details are not reported.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The principal limitation for interpreting the primary outcome is that its numerical result is not reported in the registry record. Although the endpoint is defined as mean change from baseline in calculated DAS28-ESR at Week 24 and a 95% confidence interval is specified, neither the estimate nor the interval limits appear in the record. No numerical between-cohort contrast is reported there either.

The observational design is an additional limitation. Allocation is not stated, and the registry explicitly cautions that the arms are not a like-for-like control comparison. Information about baseline characteristics, treatment selection, confounding adjustment, and balance between cohorts is not reported in the registry record. A causal interpretation of differences between groups is therefore not established.

The registry reports enrollment of 1225, but not the number included in the primary endpoint analysis. The effectiveness population is defined by membership in the safety population and the timing of first biologic administration after the last disease activity assessment. The extent and pattern of missing outcome data, reasons for exclusions, and methods for handling missing values are not reported.

The statistical model underlying the least-squares mean is not described. The registry record does not report included covariates, variance methods, sensitivity analyses, or diagnostic checks. It also lists 20 outcome measures and 25 analyses without reporting a multiplicity strategy. These omissions limit assessment of the robustness and reproducibility of any numerical findings.

For safety, the record gives affected-participant counts and selected event terms, but does not report deaths, arm attribution for the listed terms, or sufficient detail about ascertainment and exposure to interpret comparative risks. These limitations are specific to the information recorded and should not be taken as evidence that unreported procedures or results did not exist; they are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: An Global Comparative Observational Study of RoActemra/Actemra (Tocilizumab) in Patients With Rheumatoid Arthritis — NCT01543503.
• Registry record last updated 2016-02-10; status last verified 2016-01.
• Results first posted 2016-02-10.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry record describes a completed, prospective, multicenter observational study of first biologic therapy for rheumatoid arthritis. It lists 1225 participants, Hoffmann-La Roche as sponsor, a start date of 2012-02, and primary completion in 2015-02. Data collection was described over 52 weeks.

The primary endpoint was mean change from baseline in calculated DAS28-ESR at Week 24, reported as a least-squares mean with a 95% confidence interval. The numerical result is not reported in the registry record. Secondary outcomes include disease-activity measures, inflammatory markers, joint counts, physician global assessment, and loss of efficacy or intolerance.

The study was not randomized; allocation was not stated, and the registry cautions that the cohorts are not a like-for-like control comparison. Serious adverse events were recorded for 22 of 423 participants receiving tocilizumab and 64 of 793 receiving a TNF inhibitor. The registry does not report deaths. Important details about the primary analysis model, missing data, and multiplicity are not reported in the registry record.