Severe Psoriatic Arthritis - Early intervEntion to Control Disease: the SPEED Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Severe Psoriatic Arthritis - Early intervEntion to Control Disease: the SPEED Trial (NCT03739853), a randomized phase 4 study. The primary outcome is Psoriatic Arthritis Disease Activity Score (PASDAS).

2. Clinical Question

The study's clinical question concerns whether earlier, more intensive treatment strategies are associated with different disease-activity scores than the standard step-up approach at the stated assessment time. The primary results do not, by themselves, establish why groups differ or how broadly the findings apply. The registry record does not report a formal primary analysis model, adjustment strategy, or analysis population beyond the description “With complete data.”

3. Trial Design

SPEED is a phase 4, randomized, single-masked, three-arm parallel-group trial nested within a cohort, described in the record as a Trials Within Cohorts (TWiCs) design. The sponsor is the University of Oxford. The record lists an enrollment of 192 and a completed status. The listed study start was 2019-05-14, and primary completion was 2024-10-31. The identity of those masked is not reported in the registry record.

The three groups were standard care, combination csDMARD therapy, and early TNF inhibition. Standard care was cohort step-up therapy. The combination strategy prescribed methotrexate with an additional csDMARD, either sulfasalazine or leflunomide, at baseline. In the early TNF inhibition arm, participants were prescribed methotrexate with adalimumab at baseline. The description says adalimumab was given every two weeks and that TNF inhibitor treatment was continued until week 24, then tapered to week 32. The registry record does not report the complete treatment description beyond the text posted for each arm.

The primary endpoint was PASDAS at 24 weeks, reported as a mean and standard deviation among participants with complete data. The record also posts PASDAS results at 48 weeks. Secondary outcomes named in the record include Minimal Disease Activity (MDA) and Psoriatic Arthritis Impact of Disease (PsAID), as well as PASDAS. Numerical results for those other named outcomes are not reported in the registry record. The original design assumptions, including the planned sample-size justification and detailed statistical analysis plan, 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: With complete data

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: No difference

6. Primary Endpoint

The registry lists the primary outcome as Psoriatic Arthritis Disease Activity Score (PASDAS), measured over 24 weeks.

The primary endpoint is the Psoriatic Arthritis Disease Activity Score (PASDAS) at 24 weeks. It is a composite measure of psoriatic arthritis disease activity, reported as a single total score on a scale from 0 to 10. Higher values indicate more active disease, while the record defines low disease activity as <3.2. The posted endpoint is continuous, so each arm is summarized by a mean and standard deviation rather than by a count of participants meeting a threshold.

A composite score can represent several aspects of disease activity in one measure, giving investigators a way to compare the overall level of measured activity between groups. The registry record does not list the components or scoring algorithm used for PASDAS, so those details cannot be described from this record. It also does not report the proportion in each arm who reached low disease activity at 24 weeks.

The endpoint population is described as participants “With complete data.” The arm summaries include 39 participants in standard care, 47 in combination csDMARD, and 53 in early TNF inhibition. The registry record does not explain how missing endpoint data were handled or whether the complete-data population was specified before analysis. These features matter when interpreting a mean: it describes the participants contributing data, and its meaning for all randomized participants depends in part on the pattern of missingness and the analysis approach.

7. Results Posted in the Registry

• Combination csDMARD: mean 4.00 versus 4.70 on control; difference -0.70 (95% CI -1.34 to -0.06), standardized effect d = -0.47; z-test p = 0.031.
• Early TNF Inhibition: mean 3.60 versus 4.70 on control; difference -1.10 (95% CI -1.80 to -0.40), standardized effect d = -0.63; z-test p = 0.002.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Psoriatic Arthritis Disease Activity Score (PASDAS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Combination csDMARD: mean 3.80 versus 4.10 on control; difference -0.30 (95% CI -0.98 to 0.38), standardized effect d = -0.18; z-test p = 0.390.
  • Early TNF Inhibition: mean 3.50 versus 4.10 on control; difference -0.60 (95% CI -1.31 to 0.11), standardized effect d = -0.34; z-test p = 0.097.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Combination csDMARD and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Early TNF Inhibition and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes named in the registry record are PASDAS, Minimal Disease Activity, and Psoriatic Arthritis Impact of Disease. PASDAS is also posted at 48 weeks, giving a later descriptive view of the same disease-activity measure used as the primary endpoint at 24 weeks. The record does not report numerical results for MDA or PsAID, and it does not list sponsor-reported analyses for the secondary outcomes.

The broad pattern of the posted PASDAS means is that both early-treatment groups have lower observed means than standard care at both assessment times. The recalculated 24-week intervals lie below zero, while the 48-week intervals include zero. This offers context for the primary result but is not an independent confirmation of it: the later outcome is the same measure at a different assessment time, and the named secondary outcomes do not have posted numerical results in the registry record.

Unadjusted secondary comparisons generally carry less weight than a clearly specified primary analysis because multiple comparisons create more opportunities for chance findings. General statistical explanation: when many hypotheses are tested, interpreting each result as if it were the only test can overstate the strength of evidence. The registry record does not report a multiplicity strategy, such as a hierarchy of tests or an adjustment procedure. The number of outcome measures listed is 4, but the record does not identify how the investigators planned to control interpretation across them.

Secondary outcomes can still help describe the range of questions the study addressed, but the absence of posted estimates limits what can be said about them. A result for a secondary endpoint should not be inferred from a primary endpoint, even when the measures concern the same disease. Without the missing outcome estimates and analysis details, the secondary endpoint set contributes context about intended measurement rather than a complete body of comparative evidence.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Psoriatic Arthritis Disease Activity Score (PASDAS)
• Minimal Disease Activity (MDA)
• Psoriatic Arthritis Impact of Disease (PsAID)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference represent?

PASDAS is a continuous score, and the registry reports a mean for each arm. A mean difference compares the two group means: the early-treatment group's mean minus the standard-care mean in the recalculated comparisons. A negative estimate therefore indicates a lower observed mean in the early-treatment group. Because higher PASDAS indicates more active disease, that direction corresponds to lower measured activity in the posted summaries. It does not account for differences in participant characteristics or other factors that might affect the comparison.

The arm standard deviations describe variation in individual scores around each arm's mean. The reported standard deviations at 24 weeks were 1.5 for standard care, 1.5 for combination csDMARD, and 1.9 for early TNF inhibition. At 48 weeks they were 1.8, 1.5, and 1.7, respectively. Standard deviation is not the uncertainty in the mean itself: uncertainty in a mean difference also depends on the number of participants contributing to each group and the analysis assumptions.

How should the confidence intervals be read?

A confidence interval gives a range of estimates compatible with the observed data under the method used. It conveys both the direction and precision of an estimate. A relatively narrow interval constrains the plausible values more closely than a wide interval, while an interval spanning zero includes both a negative and positive difference under this comparison's sign convention. The intervals quoted for this trial are recalculated, unadjusted intervals based on the posted summary statistics. The registry record does not report an adjusted confidence interval from the sponsor's own analysis.

The 24-week intervals are entirely negative, whereas both 48-week intervals include zero. This describes the geometry of the recalculated intervals, not a declaration about whether an effect is clinically important. Confidence intervals do not remove uncertainty about missing data, selection, analysis choices, or whether the chosen endpoint captures outcomes important to participants.

What do the p values say?

The p values attached to the recalculated comparisons are unadjusted. In general, a p value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It does not measure the probability that a treatment works, the size or importance of a difference, or the probability that the null hypothesis is true. The record does not report whether these calculations correspond to the trial's prespecified tests. It also does not report adjustment for testing multiple comparisons or multiple outcomes.

The posted p values should therefore be considered alongside the estimate, confidence interval, endpoint definition, and analysis population, rather than interpreted as a stand-alone conclusion. The sponsor's own analysis takes precedence if it differs from these simple recalculations; the posted record does not identify a discrepancy in this case. That does not establish that the recalculated comparisons were the planned primary analysis.

Why do analysis populations and missing data matter?

The primary result is reported for participants with complete data, rather than explicitly for all randomized participants. A complete-data summary can be informative about those observed, but it may not represent all randomized participants if the reasons for missing scores are related to disease activity, treatment, or other characteristics. General statistical practice distinguishes an analysis limited to observed values from methods that make assumptions about unobserved values. The registry record does not report imputation, sensitivity analyses, reasons for missing scores, or an intention-to-treat analysis.

The posted group counts also differ across assessment times: the 24-week complete-data groups contain 39, 47, and 53 participants, while the 48-week groups contain 45, 45, and 49. The registry record does not explain these differences. They should not be interpreted as treatment retention or follow-up patterns without additional participant-flow information.

Why can variation affect inference?

General statistical explanation: greater variation in scores makes a group mean less precisely estimated when the number of observations is held constant. Differences in group sizes also affect the uncertainty of a comparison. This is why a mean difference must be interpreted with its interval and not just by comparing the arm means. The posted summary-based calculations do not adjust for the trial design or possible covariates, and the registry record does not report the detailed model used in the sponsor's analysis.

General statistical explanation: randomization is intended to support fair comparisons by balancing measured and unmeasured factors on average across groups, but it does not ensure identical groups in a particular realized trial. Nor does randomization resolve problems from missing outcome data or departures from the planned analysis. The nested cohort design and masking designation describe elements of the trial structure, but the registry record does not report the allocation procedure, the identity of masked parties, or the specific methods used to account for the cohort structure.

What can be concluded from the posted calculations?

The calculations offer an accessible comparison of means and uncertainty from the posted summaries. They are not reported as a complete analysis of the randomized trial. No sponsor-reported inferential analyses are listed for the secondary outcomes, and the registry record does not provide the full statistical analysis plan. The estimates can therefore be used to understand the direction and precision of the available descriptive comparisons, while avoiding stronger claims about causal effects, clinical importance, or the intended confirmatory analysis.

11. Confidence Intervals and Precision

At 24 weeks, the recalculated mean difference for combination csDMARD versus standard care is -0.7, with a 95% confidence interval from -1.337 to -0.063. For early TNF inhibition versus standard care, it is -1.1, with a 95% confidence interval from -1.795 to -0.405. The intervals are wholly below zero, but their widths leave a range of possible mean differences rather than one exact value. The posted estimates are based on complete-data groups of 39, 47, and 53 participants, so they describe those with endpoint data.

At 48 weeks, the combination csDMARD comparison is -0.3, with a 95% confidence interval from -0.985 to 0.385. The early TNF inhibition comparison is -0.6, with a 95% confidence interval from -1.309 to 0.109. Both intervals include zero, leaving uncertainty about whether the corresponding underlying mean differences are negative or positive under the recalculated comparison.

Precision is not the same as importance. The registry record does not report a threshold for a clinically important PASDAS difference, so the intervals cannot be judged against such a threshold from this record. These confidence intervals are recalculated from unadjusted summary statistics and do not reflect possible adjustments, missing-data methods, or other features of the sponsor's analysis. They quantify sampling uncertainty under the calculation, not all sources of uncertainty about the study.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effects, not the trial's own original design calculations. For these continuous-outcome comparisons, the figures are participants per arm; the general convention for the other non-time-to-event outcomes is also participants per arm. Time-to-event planning figures, if present, would instead represent total events, but these PASDAS comparisons are not time-to-event outcomes.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses a stated fraction of the observed effect: 100%, 75%, or 50%. For example, the rows labeled planning for 75% of the observed effect use a smaller target effect than the full observed effect. Under the stated assumptions, detecting a smaller effect requires more participants per arm.

The 80% power column gives the planned number of participants per arm for a design that would have an 80% chance of detecting the specified effect under the assumptions, if that effect were the true effect. The 90% column gives the corresponding planning number for a 90% chance. These are design probabilities for hypothetical studies, not probabilities that the observed result is true. The calculations are not observed power.

The figures should not be read as evidence that the original trial was designed around the effects seen in its results. The original sample-size justification and design assumptions are not reported in the registry record. Because the observed effects themselves are uncertain, planning directly from them is an illustration rather than a definitive design recommendation. The registry record also gives no assurance that the assumptions of equal allocation or the normal approximation match the trial's original planning method.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.700; pooled standard deviation σ = 1.500.
2. n = 2 × (1.96 + 0.842)² × 1.500² / -0.700² = 72.1, rounded up to 73 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.7                      # observed difference in means
sigma <- 1.5                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3739853)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 73 (80% power), 97 (90% power)
 75% of observed effect: n per arm = 129 (80% power), 172 (90% power)
 50% of observed effect: n per arm = 289 (80% power), 386 (90% power)
power.t.test: 73.1 
Simulated power at n = 73 per arm: 0.801 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry record does not report whether the statistical plan specified a hierarchy among the primary and secondary outcomes, how multiple comparisons were handled, or whether any significance thresholds were adjusted. The recalculated p values are unadjusted. When several arms, assessment times, and outcomes are considered, repeated testing can increase the chance of apparently notable results arising by chance. General statistical explanation: multiplicity procedures are used to manage that issue, but the appropriate approach depends on the planned questions and analysis strategy. No particular procedure can be attributed to this trial from the registry record.

Missing-data handling is also not reported. The primary PASDAS population is described as “With complete data,” but the record does not explain why some participants lack endpoint data, whether missingness differed by arm, or whether any imputation or sensitivity analysis was performed. The 48-week arm counts differ from those at 24 weeks, but the registry record does not give participant-flow details that would explain the pattern. A complete-case estimate can be affected if participants with and without observed data differ in ways related to the outcome.

The record does not report whether the primary comparison used an intention-to-treat population, a per-protocol population, or another prespecified analysis set. It does not provide the detailed statistical model, covariates, handling of the cohort structure, or adjustments for baseline score. Nor does it report interim analyses, stopping rules, or the original sample-size justification. These absences do not show that such methods were or were not used; those details are simply not reported in the registry record.

These reporting gaps limit the ability to connect the descriptive summaries to a complete confirmatory analysis. The recalculated estimates remain useful descriptions of the posted data, but they cannot establish whether the analysis followed a prespecified plan or whether alternative assumptions would materially change the findings. The registry record does not report enough information to resolve those questions.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cerebrovascualr haemorrhage (1.6%), perineal abscess (1.6%), gastroenteritis (1.5%), lower respiratory tract infection (1.5%).

The registry reports serious adverse-event figures for groups of 64 participants at risk in standard care, 63 in combination csDMARD, and 65 in early TNF inhibition. The corresponding numbers affected by serious adverse events are 2, 0, and 2. Deaths are reported as 0 in each group. These are counts, not comparative risk estimates, and the registry record does not report a statistical comparison of safety outcomes.

For other adverse events, the record lists 19 affected among 64 at risk in standard care, 20 among 63 in combination csDMARD, and 18 among 65 in early TNF inhibition. These figures should be read as descriptive counts; the registry record does not provide adjusted comparisons or establish whether events were related to treatment.

The listed serious-event terms and percentages are Cerebrovascualr haemorrhage at 1.6%, perineal abscess at 1.6%, Gastroenteritis at 1.5%, and Lower respiratory tract infection at 1.5%. The record does not report arm-specific attribution for this term list. Rare-event counts can be unstable and do not, on their own, establish comparative safety. The figures describe what is posted, not a complete assessment of risk.

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

The posted results show lower mean PASDAS scores in both early-treatment groups than in standard care at the 24-week primary assessment. The unadjusted recalculated confidence intervals for these comparisons lie below zero. At 48 weeks, the point estimates remain negative, but both corresponding intervals include zero. These are statistical descriptions of the registry results, not a determination that any treatment works or does not work.

The clinical meaning of a mean difference depends on more than its sign or p value. PASDAS ranges from 0 to 10, with higher values indicating more active disease, and the record defines low disease activity as <3.2. However, the registry record does not report a minimum clinically important difference or the proportion of participants achieving low disease activity for the primary endpoint. A difference between group means cannot be translated into the number of individuals who improved or reached a disease-activity state.

The comparisons are recalculated from group summaries and are unadjusted. They do not show how baseline characteristics, missing outcomes, or the cohort design were handled in a sponsor analysis. The registry record does not report the full statistical plan, multiplicity approach, or detailed missing-data methods. These omissions make it important to treat the results as a limited account of the available posted summaries.

Nor do the data by themselves establish how well these findings apply beyond the trial population. The study focused on newly diagnosed moderate to severe psoriatic arthritis within a cohort, and the record's primary analysis includes only participants with complete data. The results should be understood in that context. They do not answer every question about long-term outcomes, other populations, or the balance of benefits and harms.

The safety figures are also descriptive. The counts of serious adverse events and other affected participants do not provide a formal comparison or show causal attribution. No deaths are listed, but that observation alone does not characterize all possible risks. The registry results support careful reading of the reported outcome measures, while leaving clinical decisions and broader conclusions outside the scope of these summaries.

17. Why This Trial Matters Statistically

Statistically, SPEED is notable for comparing three treatment strategies within a cohort-based randomized design rather than describing a single treatment group. That design context matters: the standard-care group is the cohort's step-up approach, while the other arms represent early treatment strategies. The registry identifies randomization and single masking, but does not report enough detail to assess the allocation process or who was masked.

The primary measure is continuous, and the record gives both group summaries and recalculated comparisons. This allows readers to inspect the size and precision of mean differences rather than relying on a single test result. At 24 weeks, the intervals are wholly below zero; at 48 weeks, they include zero. The contrast underscores why estimates and intervals are informative alongside p values, particularly when the later assessment is less precise about the direction of the difference.

The record also illustrates a common challenge in registry-based interpretation: posted outcome summaries do not necessarily supply the complete analysis plan. Here, the primary population is complete-data participants, while details about missingness, model specification, multiplicity, and the original sample-size justification are not reported in the registry record. Such omissions constrain inference even when numerical summaries are available.

Finally, the sample-size illustrations are explicitly based on observed effects and stated assumptions rather than the original trial design. They show how planning changes when the target effect is smaller, but they cannot establish whether the completed trial had adequate power for its intended primary question. Keeping descriptive results, unadjusted recalculations, and hypothetical planning calculations separate is essential to understanding what the registry record can support.

18. 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.

Several limitations specific to the available record affect interpretation. The primary PASDAS results use complete-data participants, and the record does not explain missing scores, participant flow, or missing-data methods. The primary endpoint counts are 39, 47, and 53 across the three arms, while the 48-week counts are 45, 45, and 49. The reasons for these differences are not reported in the registry record.

The numerical comparisons are unadjusted recalculations from arm-level means, standard deviations, and counts. The sponsor's inferential model, any baseline adjustment, and any method accounting for the nested cohort design are not reported. The record also does not describe the randomization procedure or identify who was masked. These limits make it impossible to assess the full analysis in detail from the posted summary alone.

Multiplicity is another unresolved issue. There are multiple arms, assessment times, and named outcomes, but the record does not report a prespecified testing hierarchy or adjustment strategy. Numerical results for MDA and PsAID are not reported, and no sponsor-reported secondary analyses are listed. As a result, the set of secondary questions cannot be evaluated comprehensively.

The registry record does not report a clinically important difference for PASDAS, so the practical importance of the mean differences cannot be judged against a stated threshold. It also does not report the original sample-size justification or detailed protocol assumptions. The planning illustrations use observed effects and should not be mistaken for the trial's design calculations. Finally, the posted adverse-event figures are descriptive and lack reported comparative analysis or arm-specific attribution for the listed serious-event terms.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Severe Psoriatic Arthritis - Early intervEntion to Control Disease: the SPEED Trial — NCT03739853.
• Registry record last updated 2025-11-26; status last verified 2024-12.
• Results first posted 2025-11-26.
• 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.

22. Record Summary

SPEED (NCT03739853) is a completed, phase 4, randomized, single-masked, three-arm parallel-group trial sponsored by the University of Oxford and nested within a cohort. It enrolled 192 participants and compared standard step-up care with early combination csDMARD therapy and early TNF inhibition for newly diagnosed moderate to severe psoriatic arthritis.

The primary endpoint was mean PASDAS at 24 weeks among participants with complete data. The posted means were 4.7 for standard care, 4.0 for combination csDMARD, and 3.6 for early TNF inhibition. Recalculated unadjusted comparisons had negative mean differences in both early-treatment groups. PASDAS was also posted at 48 weeks; the recalculated confidence intervals for both comparisons at that assessment include zero.

The registry names MDA and PsAID among secondary outcomes but does not report their numerical results. It also does not report the detailed analysis plan, missing-data methods, multiplicity strategy, or original sample-size justification. Safety figures are descriptive counts, with 0 deaths reported in each group.