A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction Therapy in Patients With Moderately to Severely Active Crohn's Disease (UNITI-2): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction Therapy in Patients With Moderately to Severely Active Crohn's Disease (UNITI-2) (NCT01369342), a randomized phase 3 study. The primary outcome is Number of Participants With Clinical Response at Week 6.

2. Clinical Question

The primary endpoint was a count of participants with clinical response at Week 6. Clinical response was defined mainly as a reduction from baseline in the Crohn’s Disease Activity Index (CDAI) score of at least 100 points. The record also specifies a different response condition for participants whose baseline CDAI score was from 220 to 248: they were considered in response if they reached a CDAI score below 150.

The study compared each ustekinumab dose group with placebo. The posted primary outcome counts were 60.0 in the placebo group, 108.0 in the 130 mg group, and 116.0 in the approximately 6 mg/kg group; each group had 209.0 participants in the efficacy analysis. These are counts, not continuous measures of the amount of change in CDAI. The comparisons therefore address the proportion meeting a defined response threshold, rather than the average change in disease activity.

3. Trial Design

The registry record describes UNITI-2 as a Phase 3, randomized, double-blind, placebo-controlled, parallel-group, multicenter study. The masking designation is quadruple. The study was completed. Its stated purpose was to compare the effects of an initial ustekinumab treatment with placebo over 8 weeks in patients with moderately to severely active Crohn’s disease.

The three groups received placebo, ustekinumab 130 mg, or ustekinumab approximately 6 mg/kg. Each intervention is described as a single intravenous dose. The dose in the approximately 6 mg/kg group was weight-based; the record lists doses of 260 mg for participants weighing 55 kg or less and 390 mg for participants weighing more than 55 kg and 85 kg or less, then gives an incomplete description of the remaining weight range. The full dosing details beyond that incomplete text are not reported in the registry record.

The registry lists enrollment of 640. The primary efficacy analysis set is described as including participants randomized after the study was restarted. The posted primary outcome counts are based on 209.0 participants in each group; the registry record does not explain in the outcome description how that analysis-set size relates to overall enrollment. The study start is recorded as 2011-07 and primary completion as 2014-08.

The primary outcome was clinical response at Week 6. Secondary outcomes included clinical remission and clinical response at Week 8, as well as CDAI 70-point response at Week 6 and Week 3. The original protocol’s sample-size justification, detailed randomization procedures, and full 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.

Before assignment: In November 2011, due to stability issue with the batch of the IV drug (130 mg Ustekinumab) sponsor temporarily suspended dosing of participants with ustekinumab. Study was restarted with 90 mg/ml on 17 February 2012.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Efficacy analyses set included all the participants who were randomized after the study was restarted.

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 Number of Participants With Clinical Response at Week 6, measured over Week 6.

The primary endpoint counts participants who met the study’s definition of clinical response at Week 6. The main definition required a decrease from baseline in CDAI score of at least 100 points. The registry record adds a special criterion for participants with baseline CDAI scores from 220 to 248: they were considered to have responded if they attained a score below 150.

The CDAI is a disease-activity index. The registry describes its range as 0 to approximately 600 and says that a higher score indicates higher disease activity. A decrease over time indicates improvement in disease activity. The endpoint converts this score change into a binary classification: each participant either meets the response definition or does not. It therefore summarizes how many participants crossed a prespecified threshold, rather than how large each participant’s score change was.

A threshold-based endpoint can make a group comparison easier to interpret in terms of the number of participants reaching a defined response state. It also discards information about differences in score change among participants on the same side of the threshold. A participant just meeting the definition and one with a larger decrease both contribute one response to the count. The registry reports counts for the primary endpoint but does not report individual CDAI changes or a continuous group-level CDAI summary in this outcome.

The Week 6 timing is the stated primary timeframe. The endpoint definition and population are important to the interpretation: the posted counts describe participants in the efficacy analysis set who were randomized after the study restarted, not an unspecified broader population.

7. Results Posted in the Registry

• Ustekinumab 130 mg: 51.7% (108/209) versus 28.7% (60/209) on control; risk difference 23.0 percentage points (95% CI 13.8 to 32.1), risk ratio 1.80 (1.40–2.31); unadjusted two-proportion z-test p = <0.001.
• Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg): 55.5% (116/209) versus 28.7% (60/209) on control; risk difference 26.8 percentage points (95% CI 17.7 to 35.9), risk ratio 1.93 (1.51–2.47); unadjusted two-proportion z-test p = <0.001.

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.

Number of Participants in Clinical Remission at Week 8

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ustekinumab 130 mg: 30.6% (64/209) versus 19.6% (41/209) on control; risk difference 11.0 percentage points (95% CI 2.8 to 19.3), risk ratio 1.56 (1.11–2.20); unadjusted two-proportion z-test p = 0.009.
  • Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg): 40.2% (84/209) versus 19.6% (41/209) on control; risk difference 20.6 percentage points (95% CI 12.0 to 29.1), risk ratio 2.05 (1.49–2.82); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab 130 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Participants in Clinical Response at Week 8

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ustekinumab 130 mg: 47.4% (99/209) versus 32.1% (67/209) on control; risk difference 15.3 percentage points (95% CI 6.0 to 24.6), risk ratio 1.48 (1.16–1.89); unadjusted two-proportion z-test p = 0.001.
  • Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg): 57.9% (121/209) versus 32.1% (67/209) on control; risk difference 25.8 percentage points (95% CI 16.6 to 35.0), risk ratio 1.81 (1.44–2.27); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab 130 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Number of Participants With Crohn's Disease Activity Index (CDAI) 70 Point Response at Week 6

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Ustekinumab 130 mg: 58.9% (123/209) versus 38.8% (81/209) on control; risk difference 20.1 percentage points (95% CI 10.7 to 29.5), risk ratio 1.52 (1.24–1.86); unadjusted two-proportion z-test p = <0.001.
  • Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg): 64.6% (135/209) versus 38.8% (81/209) on control; risk difference 25.8 percentage points (95% CI 16.6 to 35.1), risk ratio 1.67 (1.37–2.03); unadjusted two-proportion z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab 130 mg than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Ustekinumab Approximately (~) 6 Milligram Per Kilogram (mg/kg) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Across the four posted secondary endpoints, the outcome counts were higher in both ustekinumab groups than in the placebo group. These endpoints assess clinical remission and clinical response at Week 8, and CDAI 70-point response at Week 6 and Week 3. The broad direction is consistent with the Week 6 primary endpoint: more participants in each active group met the reported endpoint definitions than in placebo.

The secondary outcomes broaden the description of response across different definitions and timeframes. They do not replace the primary outcome, which was designated as clinical response at Week 6. Nor do they establish a comparison between the two active doses: the reported comparisons are each against placebo. The registry does not report a direct statistical test of one ustekinumab dose against the other.

The registry reports Cochran-Mantel-Haenszel chi-square tests for the secondary placebo comparisons. For Week 8 remission, the reported p-values are 0.009 for 130 mg and < 0.001 for approximately 6 mg/kg. For Week 8 clinical response and CDAI 70-point response at Week 6, the reported p-values are < 0.001 for both active groups. These unadjusted values do not account for the number of secondary outcomes and comparisons unless a procedure is specified, and a multiplicity procedure is not reported in the registry record.

General: testing multiple outcomes can increase the chance of obtaining one or more small p-values even when the underlying effects are absent or smaller than estimated. This does not mean that a particular secondary result is false; it means that unadjusted secondary evidence is less definitive when considered across a family of tests. The consistency of direction is useful descriptive context, but it does not substitute for a prespecified hierarchy or multiplicity adjustment. The registry record does not state whether these secondary analyses were intended as confirmatory or exploratory.

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

• Number of Participants in Clinical Remission at Week 8
• Number of Participants in Clinical Response at Week 8
• Number of Participants With Crohn's Disease Activity Index (CDAI) 70 Point Response at Week 6
• Number of Participants With CDAI 70 Point Response at Week 3

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: a participant either met the clinical-response definition at Week 6 or did not. For a binary endpoint, the number of participants meeting the definition is interpreted in relation to the number analyzed in each group. Here the posted primary analysis gives counts of 60.0, 108.0, and 116.0, with 209.0 participants in each group. These counts are not means, and they do not describe the distribution of individual CDAI score changes.

What does a risk difference express?

General: a risk difference is the difference between the proportions experiencing an event in two groups. For these results, the event is meeting the clinical-response definition. A positive risk difference indicates a higher observed response proportion in the ustekinumab group than in placebo. The recalculated estimates are 0.230 for 130 mg versus placebo and 0.268 for approximately 6 mg/kg versus placebo. They are unadjusted calculations from the posted summary counts, not estimates adjusted for other variables.

The confidence intervals attached to these differences describe uncertainty in the estimated risk difference. Their interpretation depends on the statistical method and its assumptions. In general, a confidence interval gives a range of values reasonably compatible with the observed data under the method; it is not a guarantee that the true value lies within the interval, nor is it a probability statement about a fixed parameter. The precision of the estimate is reflected in the interval’s width as well as its location.

What does the registry’s test contribute?

The registry reports Cochran-Mantel-Haenszel chi-square tests for the two primary placebo comparisons. In general, this family of tests evaluates association between group and a categorical outcome, with a method that can account for stratification when strata are part of the analysis. The stratification variables used for the reported analyses are not reported in the registry record. The registry gives p < 0.001 for both comparisons but does not give a test statistic, degrees of freedom, or a detailed account of any adjustment variables.

A p-value summarizes how incompatible the observed data are with a specified null hypothesis, under the test assumptions. It does not express the probability that the null hypothesis is true, the probability that a result is due to chance, or the magnitude of the group difference. The sponsor’s reported test and the unadjusted risk-difference recalculation answer related but not identical statistical questions. The test is the analysis reported in the registry; the risk difference and its interval are simple calculations from posted summary counts.

Why is the endpoint definition important?

The clinical-response endpoint depends on a threshold applied to CDAI change, with a special rule for baseline scores from 220 to 248. Thresholding makes the outcome categorical and permits a direct comparison of response counts. As a general statistical consequence, categorization can hide variation in the size of score changes and can make results sensitive to where the threshold is set. The registry describes the threshold but does not report analyses using alternative thresholds or continuous CDAI change.

The efficacy population is specified as participants randomized after the study was restarted. The reasons for the restart, how participants were handled if outcome data were missing, and whether any imputation or other missing-data procedure was used are not reported in the registry record. Those details matter because a binary outcome analysis depends on how the analysis set and missing observations are defined.

How should multiple outcomes be considered?

The primary endpoint has two placebo comparisons, and the registry also reports four secondary endpoint comparisons for each active group. When many statistical tests are considered, the chance of at least one result appearing notable can increase, even if no individual comparison is especially persuasive. This is a general multiplicity concern. The registry record reports the analyses and p-values but does not report a multiplicity-control procedure, a hierarchy among outcomes, or whether secondary outcomes were confirmatory or exploratory. The primary endpoint should therefore be distinguished from the secondary results, and the latter should be interpreted with appropriate caution.

In general, a statistical plan specifies the principal endpoint, analysis population, model or test, handling of missing data, and approach to multiple comparisons before results are examined. The trial’s full original plan and sample-size justification are not reported in the registry record, so those features cannot be inferred from the outcome results alone.

11. Confidence Intervals and Precision

The primary risk-difference estimate for ustekinumab 130 mg compared with placebo is 0.230, with a 95% confidence interval from 0.138 to 0.321. For approximately 6 mg/kg compared with placebo, the estimate is 0.268, with a 95% confidence interval from 0.177 to 0.359. Both intervals lie above zero. Under the recalculation method, the posted data are therefore compatible with positive differences in response proportions for both comparisons.

The intervals also show uncertainty about the size of each difference: the data do not identify a single exact value. An interval’s width reflects both the observed data and the method used to estimate uncertainty. In general, narrower intervals indicate greater precision, while wider intervals leave more uncertainty about the plausible magnitude. The two intervals overlap; that overlap is not itself a formal test comparing the ustekinumab doses. The registry’s reported primary tests compare each dose group with placebo, not one dose with the other.

A confidence interval should not be read as the range containing most individual participant outcomes, nor as a guarantee about future studies. It concerns uncertainty in a group-level estimate under the specified calculation. These intervals are unadjusted recalculations from the posted summary counts. The registry’s own reported primary analyses are Cochran-Mantel-Haenszel chi-square tests, for which p < 0.001 is reported for each comparison. The confidence intervals and p-values are distinct summaries; neither conveys the full clinical context or resolves limitations such as unreported missing-data handling and multiplicity procedures.

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 sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each illustration asks how many participants per arm would be needed to detect the stated fraction of the observed effect at the stated power level. The original trial sample-size justification and assumptions are not reported in the registry record.

For the primary comparison of 130 mg with placebo, planning for 100% of the observed effect gives 71 participants per arm for 80% power and 94 participants per arm for 90% power. Planning for 75% of the observed effect gives 123 participants per arm for 80% power and 164 participants per arm for 90% power. These are illustrations, not a reconstruction of the trial’s enrollment plan.

In plain language, 80% power means that, if the assumed effect and other planning assumptions were correct, a study of the stated size would have an 80% chance of producing a statistically significant result under the calculation. The 90% column describes the corresponding planning target with a higher chance of detecting that assumed effect. The calculation is not observed power; it does not describe the probability that the completed trial’s result is true or will replicate.

The listed figures for these non-time-to-event outcomes are participants per arm, not total events. The stated assumptions apply to two-arm comparisons with equal allocation. Smaller assumed effects require more participants to detect them at the same power and significance level, which is why planning for 75% or 50% of an observed effect yields larger requirements than planning for the full observed effect. The figures depend on the observed effect used as their starting point, so they should be treated as planning illustrations rather than independent evidence about the size of a future effect.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.287; treatment proportion p₁ = 0.517; average p̄ = 0.402.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6934 = 1.3590.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6741 = 0.5673.
4. n = (1.3590 + 0.5673)² / (0.2297)² = 70.3, rounded up to 71 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
p0 <- 0.287081339712919                          # control proportion
p1 <- 0.516746411483254                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1369342)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 71 (80% power), 94 (90% power)
 75% of observed effect: n per arm = 123 (80% power), 164 (90% power)
 50% of observed effect: n per arm = 268 (80% power), 358 (90% power)
power.prop.test: 70.3 
Simulated power at n = 71 per arm: 0.804 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Analyses Reported by the Sponsor

The registry’s own primary analyses compare placebo separately with ustekinumab 130 mg and approximately 6 mg/kg using a Cochran-Mantel-Haenszel chi-square test. It reports p < 0.001 for both comparisons. The unadjusted recalculations from the posted counts also give p < 0.001 for both primary comparisons, and their risk-difference confidence intervals are above zero. No discrepancy is flagged between the registry’s reported primary results and the recalculated summaries.

The sponsor’s reported analysis should be distinguished from the simple recalculation. The sponsor’s method is the test recorded in the registry, while the recalculated risk differences and confidence intervals are unadjusted estimates derived from the summary counts. They offer a transparent description of the absolute group contrasts, but they do not reproduce any stratification or other features of the reported test. The stratification details and full analysis specification are not reported in the registry record.

The same general correspondence appears in the secondary results: the reported analyses use the same named test for placebo comparisons, and the posted p-values match the unadjusted recalculations at their displayed precision. Agreement between summaries does not remove the need to consider multiple outcomes, missing-data methods, or the limits of the registry record. It indicates that the simple comparisons are directionally consistent with the sponsor’s reported analyses, not that every design or analysis question has been resolved.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary endpoint and multiple secondary outcomes, with comparisons of each ustekinumab group against placebo. That creates multiple opportunities to obtain small p-values. A statistical plan can address this by specifying a hierarchy, allocating error across tests, or using another multiplicity-control approach. Whether any such method was used is not reported in the registry record. The reported p-values should therefore be read as the values for the listed analyses, not as evidence that family-wise multiplicity has been controlled.

The record identifies the primary efficacy analysis set as all participants randomized after the study was restarted. It does not report the reason for the restart, the number of participants excluded from the efficacy set, or the reasons for any exclusions. The outcome entries show 209.0 participants per group for the primary analysis. The relationship between that analysis set and the registry’s overall enrollment of 640 is not explained in the outcome description.

How missing endpoint information was handled is not reported in the registry record. The record does not state whether participants with missing Week 6 or Week 8 data were counted as nonresponders, excluded, imputed, or addressed through another method. This matters because different missing-data rules can change the analyzed response counts and the uncertainty around group comparisons.

The registry reports the Cochran-Mantel-Haenszel chi-square test but does not provide the stratification factors, detailed model specification, or enough information to reconstruct the sponsor’s analysis fully. The original statistical analysis plan, including any prespecified sensitivity analyses, is not reported in the registry record. These omissions limit how completely the posted results can be audited and how confidently the reported tests can be interpreted beyond their stated comparisons. They do not alter the fact that the registry reports higher response counts in the active groups, but they qualify the strength and scope of conclusions drawn from the analyses.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: crohn's disease (2.4%), anal abscess (1.4%), small intestinal obstruction (0.9%), colitis (0.5%), chest pain (0.5%).

The registry lists serious adverse events for the safety groups. Serious adverse events affected 15 participants among 212 in the placebo group, 10 among 216 in the ustekinumab 130 mg group, and 9 among 211 in the approximately 6 mg/kg group. These are counts with the corresponding at-risk group sizes; they are not adjusted comparisons between groups. The registry does not report a statistical test or confidence interval for these safety figures.

The record also lists other adverse events affecting 44 participants among 212 in the placebo group, 40 among 216 in the 130 mg group, and 44 among 211 in the approximately 6 mg/kg group. These are separate from the serious adverse-event counts. The most frequently listed serious-event terms include Crohn’s disease at 2.4%, anal abscess at 1.4%, and small intestinal obstruction at 0.9%; colitis, chest pain, and gastroenteritis are each listed at 0.5%. The record does not provide arm-specific values for these terms in the posted summary.

Deaths are not reported in the registry record. The safety counts describe events during the period covered by the registry entries; they do not by themselves establish that an intervention caused or prevented an event. The registry summary also does not provide enough detail to assess event timing, duration, severity beyond the serious classification, or statistical uncertainty for the between-group safety contrasts.

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

The posted results show that more participants in each ustekinumab group met the Week 6 clinical-response definition than in the placebo group. The recalculated risk differences are positive, their confidence intervals are above zero, and the registry reports p < 0.001 for both primary placebo comparisons. These are findings about the specified endpoint in the stated efficacy analysis set.

They do not mean that every participant responded, that each participant experienced the same change, or that a response was sustained beyond the reported timeframes. The primary endpoint classifies participants using a CDAI threshold; it does not summarize the size of individual changes. The record also gives secondary results at Week 3 and Week 8, but those endpoints do not provide information about longer-term outcomes.

The analyses do not establish that one ustekinumab dose is better than the other. The registry reports each dose compared with placebo, not a direct head-to-head dose comparison. Likewise, the statistical results do not establish the clinical importance of the estimated differences by themselves. That judgment requires context beyond a p-value or risk-difference estimate and should not be inferred solely from the registry statistics.

The results also do not resolve questions about missing data, multiplicity, the trial’s original sample-size rationale, or all details of the analysis plan; those are not reported in the registry record. Safety counts are descriptive and do not establish causation. Overall, the posted data document higher response counts in the two active groups for the reported endpoints, with quantified uncertainty for the primary comparisons. They do not establish outcomes for people outside the analyzed population or beyond the trial’s reported endpoint timeframes.

18. Why This Trial Matters Statistically

This trial illustrates how a binary clinical endpoint can be analyzed and communicated. The primary outcome is a count of participants crossing a defined CDAI response threshold, which allows a direct comparison of response proportions across randomized groups. Reporting both the counts and the analyzed group sizes helps readers understand the basis of that comparison.

The risk differences add an absolute-scale estimate to the registry’s hypothesis tests. An absolute contrast can be useful because it describes how far apart the observed response proportions are, while a confidence interval communicates uncertainty around that estimate. The posted primary intervals lie above zero, and the registry’s chi-square tests report p < 0.001 for both placebo comparisons. These summaries are complementary rather than interchangeable: one focuses on estimated magnitude and precision, the other on evidence against a specified null hypothesis.

The record also demonstrates why analysis details matter. The registry identifies an efficacy population tied to randomization after a study restart, but does not report missing-data handling, multiplicity control, or the full original analysis plan. Those details affect how results should be interpreted and how easily an independent reader can reproduce the sponsor’s analyses.

Finally, the secondary outcomes show a direction consistent with the primary result, while also raising the usual issue of multiple testing. Statistical interpretation requires considering the endpoint hierarchy, uncertainty, analysis population, and limits of follow-up together. The trial’s posted results are therefore useful not just as a set of p-values, but as an example of why effect estimates, precision, population definitions, and unreported methods all belong in a careful reading of clinical evidence.

19. 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 primary efficacy results concern a defined analysis set: participants randomized after the study was restarted. The registry record does not explain the restart or detail how the efficacy set relates to overall enrollment. This limits a reader’s ability to reconstruct participant flow or assess the reasons for any exclusions.

The endpoint is based on a thresholded CDAI change. As a general limitation of binary endpoints, this can conceal variation in the size of change among participants and makes the count dependent on the chosen definition. The record does not report a continuous analysis of CDAI change alongside the primary count. It also does not report sensitivity analyses using alternative response thresholds.

The outcome timeframes are Week 3, Week 6, and Week 8. The registry’s stated study comparison covers an initial treatment over 8 weeks, so these posted results do not describe persistence beyond those timeframes. The record does not report longer-term efficacy results in the outcome information summarized for this trial.

Missing-data procedures, the full original statistical plan, and multiplicity control are not reported in the registry record. These omissions limit assessment of possible bias from missing outcomes and the evidential status of multiple secondary comparisons. The reported comparisons are each against placebo; a direct comparison between the two ustekinumab doses is not reported. For safety, the registry gives event counts and selected serious-event terms but does not provide uncertainty estimates or a statistical comparison in the posted summary. These limitations qualify the scope of interpretation without changing the descriptive pattern in the recorded counts.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Safety and Efficacy of Ustekinumab Induction Therapy in Patients With Moderately to Severely Active Crohn's Disease (UNITI-2) — NCT01369342.
• Registry record last updated 2017-01-06; status last verified 2016-11.
• Results first posted 2017-01-06.
• 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.

23. Record Summary

UNITI-2 is a completed Phase 3, randomized, double-blind, placebo-controlled, parallel-group, multicenter study of intravenous ustekinumab induction in moderately to severely active Crohn’s disease. The registry lists enrollment of 640 and records a study start of 2011-07 and primary completion of 2014-08.

The primary endpoint was clinical response at Week 6. Counts were 60.0 for placebo, 108.0 for ustekinumab 130 mg, and 116.0 for approximately 6 mg/kg, with 209.0 participants per group in the efficacy analysis. Unadjusted risk-difference recalculations were 0.230 and 0.268 for the two active-group comparisons, respectively; the registry reports p < 0.001 for each.

Secondary endpoint counts were higher in both active groups than in placebo across the posted outcomes. Interpretation is limited by unreported details about missing-data handling, multiplicity procedures, the original sample-size rationale, and the full statistical plan. The registry’s reported primary test is the Cochran-Mantel-Haenszel chi-square test.