Study Evaluating Etanercept in the Treatment of Subjects With Psoriasis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study Evaluating Etanercept in the Treatment of Subjects With Psoriasis (NCT00195507), a randomized phase 4 study. The primary outcome is Physician Global Assessment of Psoriasis (PGA) Score - Mean Value Over 54 Weeks.

2. Clinical Question

This is a comparison of two treatment strategies using the same named drug, rather than a comparison of etanercept with no treatment or with a different drug. The primary outcome summarizes physician-assessed plaque severity across the stated timeframe. Secondary outcomes addressed patient global assessment, time to a physician score of clear or almost clear, and the number of patients reporting at least somewhat satisfied. The registry record does not provide a detailed protocol for the treatment strategies or additional eligibility criteria, so the clinical question should be read within the limits of the descriptions recorded there.

3. Trial Design

The registry describes a completed, randomized, open-label, parallel-group Phase 4 study of etanercept in people with psoriasis. The listed enrollment is 720. The study began in 2004-12. The sponsor is listed as Wyeth, now a wholly owned subsidiary of Pfizer.

Participants were assigned to continuous or intermittent therapy. The study summary characterizes the intermittent strategy as withdrawal and retreatment when psoriasis returned, and its stated aim was a direct comparison of continuous and intermittent treatment. Masking is listed as none, so the trial was open label. The registry record does not report whether outcome assessors were separately masked.

The primary outcome was the mean Physician Global Assessment score over 54 weeks. The primary analysis population was modified intention-to-treat, defined as patients who received at least one dose of study drug and had at least one post-baseline evaluation. Missing observations were handled using last observation carried forward. The primary outcome results list 352 participants in the continuous group and 359 in the intermittent group; those are the group counts for this outcome and should not be assumed to describe every analysis population or randomized group.

Three secondary outcomes are listed: percentage improvement in Patient Global Assessment from baseline; time to a Physician Global Assessment score of clear or almost clear; and the number of patients with a survey response of somewhat satisfied or better. The registry lists four outcome measures and three analyses. Details such as the allocation ratio, treatment dose, precise administration schedules, original sample-size rationale, and analysis of interim data 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: Subjects were screened up to 4 weeks.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis population was the modified intention-to-treat (mITT)population, which included all patients who received at least one dose of study drug and had at least 1 post-baseline evaluation. Last observation carried forward (LOCF).

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.

6. Primary Endpoint

The registry lists the primary outcome as Physician Global Assessment of Psoriasis (PGA) Score - Mean Value Over 54 Weeks, measured over 54 weeks.

The primary endpoint is the mean Physician Global Assessment of Psoriasis score over 54 weeks. The PGA is a 7-point scale used to assess the severity of psoriatic plaques, with 1 indicating severe psoriasis and 7 indicating clear. A higher score therefore corresponds to a clearer assessment on this scale. The endpoint is continuous, and the registry reports group means and standard deviations.

Using a mean outcome allows the analysis to compare the average score across the stated timeframe between the continuous and intermittent strategies. Such a summary can convey a group-level difference in disease-severity assessment, but it does not show how each individual participant's score changed or how scores were distributed at every assessment. The registry does not detail the schedule of assessments used to form the mean over 54 weeks.

The analysis population included patients who received at least one dose and had at least one post-baseline evaluation. The registry specifies last observation carried forward for missing data. This method carries a participant's last recorded value forward to later missing assessments; it is an assumption about the unobserved values, not a direct measurement of them. No threshold for a clinically important difference on the PGA scale is reported in the registry record.

7. Results Posted in the Registry

• Intermittent: mean 2.51 versus 1.98 on control; difference 0.53 (95% CI 0.42 to 0.64), standardized effect d = 0.70; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure summarized by a mean and standard deviation in each group. The registry reports the mean PGA score over 54 weeks for the modified intention-to-treat population. A mean gives a single group-level summary of the scores, while a standard deviation describes how spread out participant scores are around that group mean. Neither statistic alone shows the full distribution or the pattern of scores over time.

What is the unadjusted comparison?

The recalculation compares the posted group means without adjustment for additional variables. It reports a mean difference of 0.53, expressed as intermittent minus continuous. Since higher PGA values indicate a clearer assessment on the scale, the positive difference corresponds to a higher average score in the intermittent group. This is an unadjusted summary comparison, not the same procedure as the registry's ANCOVA result. It should not be treated as a replacement for the sponsor's reported analysis.

The recalculation also reports a pooled standard deviation of 0.761 and a standardized difference of 0.696. A standardized difference expresses the observed mean contrast relative to the variability of the scores, which can help describe the contrast on a scale independent of the PGA's units. It does not itself establish clinical importance. The registry record does not give a threshold for interpreting a standardized difference or a clinically important change on the PGA scale.

What does the p-value describe?

A p-value evaluates how compatible the observed data are with a specified null hypothesis under a statistical model. The unadjusted calculation and the registry's ANCOVA analysis each report p < 0.001. In general, a p-value is not the probability that the null hypothesis is true, nor does it quantify the size or clinical relevance of the contrast. Its meaning depends on the analysis and assumptions used. The registry reports ANCOVA but does not report the covariates or further model details in the record.

The registry's ANCOVA result is a mean difference (net) of -0.53, with a 95% confidence interval from -0.64 to -0.43. Its negative sign reflects the reverse direction of the group contrast from the recalculation. ANCOVA generally compares outcome means while accounting for one or more covariates; the specific covariates and model specification are not reported in the registry record. As the registry analysis is the trial's own reported analysis, it remains distinct from a simple re-analysis of the two posted group means and standard deviations.

How should the confidence interval be understood?

A confidence interval describes a range of values reasonably compatible with the data and the method used to calculate it. In repeated use of the same method under its assumptions, confidence intervals at the stated level would cover the target parameter at the corresponding long-run rate. A 95% confidence interval is not a statement that there is a 95% probability that this one fixed interval contains the true value. It also does not capture every possible source of uncertainty, such as limitations in trial conduct, missing-data assumptions, or generalizability.

The recalculated interval is based on an unadjusted contrast, while the registry interval accompanies an ANCOVA estimate. The estimates and intervals have similar magnitudes when expressed in the same direction, but their methods are not interchangeable. The registry reports its interval as 95%; the recalculation also gives a confidence interval. The registry record does not describe any further adjustment of the interval for multiple outcomes or comparisons.

What does last observation carried forward assume?

Last observation carried forward, or LOCF, uses a participant's most recent observed value as the value for later assessment points that are missing. This creates a complete set of values for the analysis under that rule, but it does not recover the participant's unknown score. In general, LOCF can be problematic when the outcome may change after the last observed assessment, because carrying the value forward assumes no further change for that participant over the missing period. The registry identifies LOCF but does not report the amount or pattern of missing data, or sensitivity analyses using alternative missing-data assumptions.

The modified intention-to-treat definition excludes people who did not receive any dose or did not have a post-baseline evaluation. This population is therefore defined by treatment receipt and availability of a follow-up measure. The registry does not provide a detailed account of participants excluded from this analysis or their outcomes.

What is the role of planning illustrations?

The sample-size illustrations use the observed contrast as a planning input and consider smaller fractions of that effect. They are not the trial's original design rationale, which is not reported in the registry record. The stated illustration assumptions include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustrations describe numbers of participants per arm for the continuous endpoint. General statistical planning asks how many participants may be needed to detect an effect under specified assumptions; it does not reveal what the trial actually had power to detect after completion.

10. Confidence Intervals and Precision

The recalculated mean difference is 0.53, with a 95% confidence interval from 0.418 to 0.642, for intermittent relative to continuous therapy. The interval gives a range of values compatible with the unadjusted contrast and its calculation method. It is above zero throughout, consistent with the reported p-value of p < 0.001 under that method. The interval's width reflects sampling uncertainty under the assumptions of the recalculation; it does not account for every source of uncertainty in trial conduct or interpretation.

The registry's ANCOVA estimate is -0.53, with a 95% confidence interval from -0.64 to -0.43. The sign is reversed because the group order is opposite to the recalculated contrast. Read in the same direction, the registry estimate also indicates a group contrast with an interval that does not include zero. The two intervals arise from different analyses: one is an unadjusted recalculation from group summaries, the other is the trial's reported ANCOVA result. The registry does not report the covariates used in the ANCOVA.

Precision is not the same as clinical importance. The confidence intervals describe uncertainty around an average group difference, not the range of outcomes individual participants experienced. The standard deviations of 0.86 for continuous therapy and 0.65 for intermittent therapy show variation among the observed scores, but the interval for the mean contrast does not describe that individual variation. No clinically important difference threshold is reported in the registry record, so the confidence intervals cannot by themselves indicate whether the estimated contrast is meaningful to patients or clinicians.

These estimates pertain to the modified intention-to-treat population and the LOCF approach described for the primary outcome. The registry record does not report the extent or pattern of missing data or the results of sensitivity analyses. Confidence intervals quantify statistical uncertainty under the analysis used, but they do not resolve uncertainty about missing-data assumptions, open-label assessment, or the applicability of the trial population and treatment strategies beyond the trial.

11. 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 participants per arm for the continuous primary outcome, not total events. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the stated fraction of the observed effect. The registry's original sample-size rationale and design assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm estimated to provide an 80% chance of detecting the specified effect under those planning assumptions, if that effect and the model assumptions were correct. The 90% power column gives the corresponding number for a 90% chance. These are planning probabilities under a hypothetical repeated-sampling framework, not probabilities that a particular result is true.

The listed scenarios use 100%, 75%, and 50% of the observed effect. For 100% of the observed effect, the illustration gives 33 participants per arm for 80% power and 44 for 90% power. For 75% of the observed effect, it gives 58 and 78 participants per arm, respectively. For 50% of the observed effect, it gives 130 and 174 participants per arm, respectively. Smaller effects need more participants to achieve the same target power under the same assumptions. These figures are not observed power: they are prospective planning illustrations based on the observed effect, not a measure of the trial's achieved power.

12. 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 = 0.530; pooled standard deviation σ = 0.761.
2. n = 2 × (1.96 + 0.842)² × 0.761² / 0.530² = 32.4, rounded up to 33 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 <- 0.53                      # observed difference in means
sigma <- 0.761238875934573                     # 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(195507)
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 = 33 (80% power), 44 (90% power)
 75% of observed effect: n per arm = 58 (80% power), 78 (90% power)
 50% of observed effect: n per arm = 130 (80% power), 174 (90% power)
power.t.test: 33.4 
Simulated power at n = 33 per arm: 0.799 (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.

13. Analyses Reported by the Sponsor

The registry reports the trial's own primary comparison as an ANCOVA mean difference (net) of -0.53, with a 95% confidence interval from -0.64 to -0.43 and p < 0.001. The unadjusted recalculation from the posted group means reports a difference of 0.53 for intermittent relative to continuous therapy, with a 95% confidence interval from 0.418 to 0.642 and p < 0.001. The signs are opposite because the contrasts use opposite group ordering. In direction and magnitude, the two estimates are consistent.

The analyses are not identical. The recalculation uses the two group summaries without adjustment; the registry identifies its analysis as ANCOVA. The record does not state which covariates were included in that model. As the two results are not flagged as discrepant, the recalculation serves as a transparent comparison based on the posted summaries, while the registry's ANCOVA remains the trial's reported analysis. Neither analysis alone answers questions about clinical importance, missing-data sensitivity, or applicability beyond the participants and strategies studied.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and three secondary outcomes: percentage improvement in Patient Global Assessment from baseline, time to a Physician Global Assessment score of clear or almost clear, and the number of patients with a survey response of somewhat satisfied or better. It lists four outcome measures and three analyses. Results for those secondary outcomes are not reported in the registry record, so they cannot be interpreted alongside the primary result.

When multiple outcomes or multiple comparisons are examined, the chance of at least one apparently notable result can increase. Statistical methods can adjust for multiplicity, or a prespecified hierarchy can define how outcomes are tested. The registry record does not report whether any multiplicity adjustment, testing hierarchy, or prespecified strategy for interpreting multiple outcomes was used. It also does not report whether interim analyses were conducted or how they would have been handled.

The primary outcome analysis population was modified intention-to-treat, consisting of patients who received at least one dose and had at least one post-baseline evaluation. The record does not report the number of participants excluded from that population, the reasons for exclusion, or whether other analysis populations were used for the primary outcome. The outcome group counts are 352 for continuous therapy and 359 for intermittent therapy.

Missing primary-outcome observations were handled using last observation carried forward. In general, this approach substitutes a participant's last recorded score for later missing scores. The registry does not report how many assessments were missing, when they were missing, or whether missingness differed between groups. Results from sensitivity analyses using other missing-data assumptions are not reported in the registry record. Consequently, the extent to which alternative approaches might affect the estimated group contrast cannot be assessed from the recorded information.

The open-label design is also relevant because the primary outcome is a physician assessment. The registry lists no masking, but does not state whether physicians assessing the outcome were separate from those aware of treatment assignment. This is a consideration when interpreting the measurement process; it does not establish that the ratings were biased. The registry does not report additional safeguards or assessment procedures in the record.

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

The posted primary data show different average PGA scores for the two treatment strategies over 54 weeks in the modified intention-to-treat population. The continuous group mean was 1.98 and the intermittent group mean was 2.51. Because higher scores on this scale indicate a clearer assessment, the group means are directionally consistent with a higher average physician score in the intermittent group. The unadjusted mean difference was 0.53, while the registry's ANCOVA estimate was -0.53 using the opposite group order.

The reported confidence intervals do not include zero, and both analyses report p < 0.001. These statistics describe evidence for a difference in group averages under the respective analyses. They do not show that every participant in one group had a clearer score than every participant in the other group. The group standard deviations indicate that scores varied within both groups.

The result does not establish the clinical importance of the difference. The registry does not report a threshold for a meaningful change in PGA, and the mean contrast does not reveal how participants valued the difference. It also does not establish that one strategy is appropriate for a particular person. The analysis concerns the trial's two etanercept strategies, its participants, its outcome definition, and its missing-data method; it is not a comparison against another medication or no treatment.

The result does not by itself establish the comparative safety of the strategies. The registry reports counts of participants affected by serious and other adverse events but does not give the corresponding at-risk denominators in the record. Those counts therefore cannot be interpreted as rates from the details available. Nor does the primary PGA result answer the listed secondary questions about patient satisfaction, patient global assessment, or time to clear or almost clear. Their results are not reported in the registry record.

Finally, the analysis used LOCF for missing outcome data and an open-label design with no masking listed. These features affect how the estimate should be interpreted, but the record does not show that they changed the result in a particular direction. The finding is a trial-specific statistical comparison, not a complete account of all benefits, harms, or treatment choices.

16. Why This Trial Matters Statistically

This trial offers a direct randomized comparison of continuous and intermittent strategies using the same named drug, etanercept. That design focuses the primary statistical contrast on treatment schedule rather than on differences between drug products. The registry states that intermittent treatment involved withdrawal and retreatment when psoriasis returned, although it does not give detailed operational rules for that strategy.

The primary endpoint is a physician-assessed continuous score, and the registry supplies group means, standard deviations, an ANCOVA estimate, a confidence interval, and a p-value. This permits readers to consider both the direction and estimated size of the average contrast, as well as its statistical precision. The unadjusted recalculation provides a separate check based on the posted group summaries, while the registry's own ANCOVA result represents the reported trial analysis.

The trial also illustrates why statistical significance and clinical interpretation are distinct. The reported p-values are small, but the registry does not report a threshold for a clinically important PGA difference. Mean contrasts describe group averages and do not capture all individual outcomes. Open-label assessment, the use of LOCF, and the absence of reported multiplicity details are relevant context for understanding the evidence.

Finally, planning illustrations based on the observed effect show how required sample sizes depend on the size of the effect chosen for planning. They are not the original rationale for the trial and do not measure its achieved power. These distinctions help separate what the statistical results estimate from what remains uncertain or unreported in the registry record.

17. 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 study was open label, with masking listed as none. The primary outcome was a physician assessment, and the registry record does not state whether outcome assessors were separately masked. Awareness of treatment assignment can be a potential source of measurement concern for an assessed outcome, though the record does not establish that any bias occurred.

The primary analysis used a modified intention-to-treat population requiring at least one dose and at least one post-baseline evaluation. This does not include every enrolled participant by definition. The registry lists 720 enrolled and outcome group counts of 352 and 359, but does not report the complete flow from enrollment and assignment through inclusion in the primary analysis. The reasons for any exclusions are not reported in the registry record.

Missing data were handled by last observation carried forward. The number and timing of missing assessments and the results of sensitivity analyses are not reported in the registry record. The original sample-size justification, allocation ratio, treatment dose, detailed continuous and intermittent schedules, and covariates in the ANCOVA are also not reported. No clinically important PGA difference threshold is given.

The registry lists serious adverse events affecting 23 participants in the continuous group and 31 in the intermittent group, and other adverse events affecting 281 and 273, respectively. At-risk denominators are not reported in the registry record, so these counts alone do not permit a rate comparison. Death counts are not reported in the registry record. The record names serious-event terms including psoriasis, skin carcinoma, accidental injury, myocardial infarct, hypertension, and dyspnea, with percentages listed for those terms, but the available details do not provide a complete safety comparison.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Study Evaluating Etanercept in the Treatment of Subjects With Psoriasis — NCT00195507.
• Registry record last updated 2013-02-25; status last verified 2010-04.
• Results first posted 2010-05-04.
• 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.

21. Record Summary

This completed randomized, open-label, parallel-group Phase 4 trial compared continuous with intermittent etanercept therapy in people with psoriasis. Enrollment was 720. The primary outcome was mean Physician Global Assessment score over 54 weeks, on a 7-point scale where 1 indicates severe psoriasis and 7 indicates clear.

In the modified intention-to-treat analysis using last observation carried forward, mean scores were 1.98 for continuous therapy and 2.51 for intermittent therapy. The registry's ANCOVA estimate was -0.53, with a 95% confidence interval from -0.64 to -0.43 and p < 0.001. The unadjusted recalculation expresses the contrast in the opposite direction and reports 0.53, with a 95% confidence interval from 0.418 to 0.642 and p < 0.001.

The result is specific to the primary outcome and analysis described. Details including the original sample-size rationale, multiplicity procedures, extent of missing data, and results for the listed secondary outcomes are not reported in the registry record.