Lubiprostone (Amitiza®) Vs. Standard Care in Opioid-induced Constipation After Surgery in Inpatient Rehabilitation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Lubiprostone (Amitiza®) Vs. Standard Care in Opioid-induced Constipation After Surgery in Inpatient Rehabilitation (NCT00662363), a randomized na study. The primary outcome is Change in Patient Assessment of Constipation (PAC) - Symptoms (Sym).

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group trial with quadruple masking. Participants were assigned to lubiprostone or senna. The masking label indicates that the registry records four parties as masked, but it does not identify those parties in the record. Additional details about how randomization and masking were implemented are not reported in the registry record.

The study concerned postoperative opioid-induced constipation among patients receiving inpatient rehabilitation following orthopedic procedures. The brief description specifies patients requiring opioids for postoperative pain control who had constipation symptoms. The listed enrollment is 64, and the trial status is completed. The record gives a start of 2008-04 and primary completion of 2011-10.

The active treatment period lasted 6 days. The lubiprostone regimen is described as 24 µg by mouth twice daily with meals, alongside two placebo senna tablets at noon. The senna regimen is described as two tablets at noon daily, alongside placebo lubiprostone capsules twice daily. The registry does not report further information about adherence, treatment changes, or other care received during the trial.

The primary outcome was assessed at baseline and Day 7, after treatment was completed. The analysis population is described as all randomized participants who had baseline and endpoint data. The posted outcome lists 28 participants in each group. The reasons why the outcome sample sizes differ from the enrollment figure, if any, are not reported in the registry record. The record lists two outcome measures and one analysis, but no secondary outcomes are posted. Further information about the original sample-size rationale, the randomization procedure, or the planned analysis is 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: 72 patients were recruited. There were eight who did not meet inclusion and/or exclusion criteria following consent and screening visit. Reasons for exclusion were: history of Crohn's disease (1 person), history of irritable bowel disease (1 person), diarrhea on baseline visit day (1 person), or no associated bowel symptom (5 persons).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects randomized and who had baseline and endpoint data were included in the intention to treat analysis.

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Change in Patient Assessment of Constipation (PAC) - Symptoms (Sym), measured over Baseline and Day 7, after treatment completed (6 days of treatment).

The primary endpoint is change in the Patient Assessment of Constipation—Symptoms (PAC-SYM) score from baseline to Day 7, after the 6-day treatment period. PAC-SYM is a questionnaire intended to measure constipation symptoms and their course. The registry describes its symptom component as 12 items, with scores from 0 to 4; lower scores indicate improvement. The symptom score is formed by averaging scores within the domains.

The registry identifies abdominal, rectal, and stool symptoms as the three symptom domains. These cover different aspects of constipation, so the measure is intended to capture more than one type of symptom. The record states that PAC-SYM has previously been found valid and reliable for measuring constipation symptoms and clinical course, and describes evidence of validity and responsiveness in opioid-induced constipation among pain populations. Those statements provide the registry’s rationale for using the questionnaire; they do not supply trial-specific evidence about how well it performed in this study.

A change score expresses the difference between a participant’s follow-up and baseline measurement. Because lower PAC-SYM scores indicate improvement, a negative mean change corresponds to a decrease in symptoms on average. The posted group means are negative in both arms. The score is a symptom measure, not a direct count of bowel movements or a measure of every possible consequence of constipation. The registry’s primary outcome description does not report a threshold for interpreting the clinical importance of a change.

7. Results Posted in the Registry

• Senna: mean -0.29 versus -0.22 on control; difference -0.07 (95% CI -0.37 to 0.23), standardized effect d = -0.12; z-test p = 0.649.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: each participant has a PAC-SYM change score, and the posted group results are means with standard deviations. A mean summarizes the average change in a group. A standard deviation describes how dispersed individual changes are around that group average. They answer different questions: the mean describes the location of the group’s results, while the standard deviation describes variability among participants.

What does a mean difference represent?

The registry-derived comparison reports a mean difference of -0.07 for Senna. A mean difference contrasts the average change in one group with the average change in the other, using a particular subtraction order. The comparison is labelled Senna, and the posted arm means are -0.22 for lubiprostone and -0.29 for senna. Since lower PAC-SYM scores indicate improvement, a more negative change represents a larger average decrease in symptoms. The direction of a contrast depends on the group order used; it should be read alongside the arm means rather than treated as an independent outcome.

The mean difference is an unadjusted comparison based on the posted summary statistics. It does not account for covariates or other factors unless an analysis explicitly includes them. The registry does not report an adjusted model for this outcome. In general, an adjusted analysis can estimate a contrast conditional on specified variables, whereas an unadjusted contrast compares the group summaries without such adjustment. No unreported adjustment should be inferred.

What does the independent-sample t test assess?

The registry names an independent-sample t test for the comparison between the two groups. In general, this test evaluates whether the group means are compatible with a null hypothesis of no mean difference, under the assumptions of the chosen t-test procedure. “Independent-sample” refers to comparing separate groups of participants rather than paired measurements as if each participant belonged to both groups. The trial also measured participants at baseline and follow-up, but the posted primary result is a change score comparison; details of the sponsor’s exact calculation are not reported.

A p-value describes how unusual the observed test statistic, or a more extreme one, would be under the null model and its assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The registry reports p <0.05 for its t test. A separate simple, unadjusted re-calculation from the posted summary statistics gives p = 0.649. These values conflict, and the registry flags a discrepancy. The sponsor’s own analysis takes precedence over the re-calculation; the record does not include enough detail to identify why the results differ.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the data and the specified analysis procedure, while communicating uncertainty around an estimate. In general, a narrower interval indicates greater precision than a wider interval. The re-calculated interval for the mean difference crosses zero, so it includes both a contrast in one direction and a contrast in the other, as well as no difference. Its span signals uncertainty in the unadjusted estimate. The sponsor’s confidence interval is not reported in the registry record, so precision for the sponsor’s analysis cannot be assessed from that analysis’s p-value alone.

The re-calculation uses the posted means and standard deviations for 28 participants in each group. It is a simple, unadjusted calculation from summary statistics, not a replacement for the sponsor’s analysis. A reported sample standard deviation describes observed spread, and the standard error describes uncertainty in an estimated mean under a statistical model. The posted arm-specific standard errors are 0.108 and 0.110. A standard error is not the same as a standard deviation: the former concerns the precision of an estimated mean, while the latter describes variation among individual scores.

Why does the outcome’s scale matter?

PAC-SYM is described as a symptom questionnaire with scores from 0 to 4 and lower scores indicating improvement. Its change scores and group means are therefore interpreted in the units of that scale. The registry does not report a minimum clinically important difference or another threshold that would classify a particular change as meaningful to patients. Statistical evidence about an average difference should not be converted into a claim about clinical importance without such context.

In general, continuous outcomes can be summarized by means when the mean is an informative description of the group and the analysis assumptions are suitable. The registry names a t test but does not report diagnostic checks, distributional information, or alternative methods. Whether those assumptions were examined is not reported in the registry record. The posted standard deviations give a summary of spread but cannot show the complete score distributions or reveal whether unusual observations influenced the means.

Finally, the analysis population description matters. The record says that randomized participants with both baseline and endpoint data were included in the intention-to-treat analysis. In general, an intention-to-treat analysis aims to preserve randomized group assignment, but requiring endpoint data means that the record’s stated inclusion rule depends on data availability. The registry does not describe imputation, other missing-data methods, or sensitivity analyses. The analysis should therefore be interpreted as the comparison among participants meeting the stated data requirement, without assuming that unobserved outcomes were handled in a particular way.

10. Confidence Intervals and Precision

The re-calculated mean difference is -0.07, and its 95% confidence interval extends from -0.371 to 0.231. Because the interval includes zero, the posted summary statistics are compatible with no average difference as well as with differences in either direction under this unadjusted calculation. The interval is wide relative to the point estimate, so the estimate does not pin down a precise between-group contrast. Its endpoints do not identify which possible value is most likely.

Precision is distinct from the direction of the point estimate. The senna group’s mean change is -0.29, compared with -0.22 in the lubiprostone group, but the confidence interval indicates substantial uncertainty around the between-group contrast. The interval is a property of the re-calculation, not a confidence interval reported for the sponsor’s own analysis. The sponsor’s confidence limits are not reported in the registry record.

In general, a confidence interval is constructed by a statistical procedure intended to capture the target quantity at a stated rate over repeated samples when the procedure’s assumptions hold. It is not a range containing a specified probability that the true value lies within it after observing the data. A confidence interval’s usefulness also depends on the suitability of the analysis, the completeness of the data, and the relevance of the analyzed participants to the intended population.

The posted standard deviations, 0.57 for lubiprostone and 0.58 for senna, indicate variation in individual change scores. The outcome sample size is 28 in each arm. Together, spread and sample size contribute to uncertainty in the group means and their contrast. The record does not provide the sponsor’s interval or enough analysis detail to determine whether its reported p-value corresponds to the same contrast or calculation as the re-calculation.

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 sample-size figures are participants per arm, not total events. They are planning illustrations based on the observed effect, not the trial’s own original design assumptions. The assumptions listed for these illustrations 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 to the stated fraction of the observed effect. The original sample-size justification and design assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm in a hypothetical design that would have an 80% chance of detecting the assumed effect under the stated planning assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding planning number for a 90% chance. These are prospective design calculations, not estimates of the chance that the completed trial detected an effect. They are not observed power.

The figures increase as the assumed effect becomes smaller: planning for 75% of the observed effect requires more participants per arm than planning for 100%, and planning for 50% requires more still. This is a general feature of power calculations: smaller differences are harder to distinguish from random variation, so more information, commonly obtained through a larger sample, is needed. The displayed curve represents that relationship between the assumed effect and the planned sample size; it should be read as a planning aid rather than as an account of the number enrolled in this trial.

For planning for 100% of the observed effect, the figures are 1,060 participants per arm for 80% power and 1,419 per arm for 90% power. For planning for 75% of the observed effect, they are 1,884 and 2,522 participants per arm, respectively. For planning for 50% of the observed effect, they are 4,238 and 5,673 participants per arm, respectively. These illustrations depend on the observed effect used as their starting point, so they should not be treated as a definitive sample-size recommendation for a future study.

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 = -0.070; pooled standard deviation σ = 0.575.
2. n = 2 × (1.96 + 0.842)² × 0.575² / -0.070² = 1,059.3, rounded up to 1,060 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 <- -0.07                      # observed difference in means
sigma <- 0.575021738719503                     # 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(662363)
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 = 1,060 (80% power), 1,419 (90% power)
 75% of observed effect: n per arm = 1,884 (80% power), 2,522 (90% power)
 50% of observed effect: n per arm = 4,238 (80% power), 5,673 (90% power)
power.t.test: 1060.2 
Simulated power at n = 1,060 per arm: 0.797 (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 that the sponsor’s analysis used an independent-sample t test and gives p <0.05. It does not provide the test statistic, the sponsor’s estimate of the mean difference, or confidence limits for that analysis. The posted arm means are -0.22 for lubiprostone and -0.29 for senna, while the simple unadjusted re-calculation gives a mean difference of -0.07 and p = 0.649.

Those results do not agree. The registry flags the discrepancy, so the sponsor’s own analysis takes precedence over the recalculation. The recalculation is a transparent, unadjusted check from the posted group summaries; it is not a substitute for the sponsor’s analysis. Without the sponsor’s estimate and interval, the registry record does not permit a detailed reconciliation of the difference between the reported p-values, nor a direct assessment of the precision of the sponsor’s estimated contrast.

The sponsor’s p-value indicates that its stated test crossed the threshold recorded in the registry, but a p-value alone does not convey the magnitude or direction of an effect. It also does not establish clinical importance. The arm means provide descriptive information about average changes, but they do not explain the discrepancy between the sponsor’s result and the summary-statistic re-calculation. Details that might clarify the difference are not reported in the registry record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures and one analysis, but no secondary outcomes are posted. The primary endpoint is change in PAC-SYM symptoms. The registry does not report whether additional outcomes were collected but not posted, whether multiple comparisons were planned, or whether any adjustment for multiplicity was used. Therefore, the record does not establish whether the reported p-value was selected from a larger set of analyses or how any such multiplicity was handled.

In general, testing multiple outcomes or contrasts can increase the chance of obtaining at least one small p-value by chance when all null hypotheses are true. Statistical methods can account for multiplicity, but the appropriate approach depends on the planned questions and analysis hierarchy. No such plan is reported in the registry record, so multiplicity adjustment should not be assumed either to have occurred or to have been omitted in the sponsor’s analysis.

The stated analysis population includes all randomized participants who had both baseline and endpoint data, and the registry calls this an intention-to-treat analysis. The endpoint results list 28 participants in each arm. The enrollment figure is 64; reasons for any difference between enrollment and endpoint analysis counts are not reported in the registry record. The record does not state how many participants lacked endpoint data, why data might be missing, or whether missingness differed by group.

In general, missing outcomes can affect an estimate if the availability of data is related to the outcome or to participant characteristics. Common strategies include modeling or imputing missing values, but no specific strategy is described. The record does not report a sensitivity analysis testing how conclusions might change under alternative assumptions about missing data. It is therefore not possible to assess from the registry record how robust the posted comparison is to missing endpoint information.

The registry does not report the sponsor’s full statistical analysis plan, prespecified covariates, checks of test assumptions, subgroup analyses, interim analyses, or protocol deviations. Their use or absence should not be inferred. The reported test and the population description are the available basis for describing the analysis, while its unreported details remain unknown.

15. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cardiomyopathy (3.4%), patellar dislocation (3.2%).

The registry lists serious adverse events affecting 1 participant in the lubiprostone group among 31 at risk, and 1 participant in the senna group among 29 at risk. The listed counts are participants affected, not a comparison of event rates adjusted for follow-up or exposure. The record also lists cardiomyopathy at 3.4% and patellar dislocation at 3.2% among the top serious adverse event terms. It does not explain in this entry how those percentages relate to the group-level counts or provide further context for each term.

These figures should be read as descriptive safety entries. The registry record does not report enough information to determine whether the listed events were judged related to study treatment, when they occurred, or how they were distributed beyond the details stated. Death counts are not reported in the registry record. The available counts alone do not establish a comparative safety effect between the treatment groups.

In general, interpreting adverse-event counts requires attention to the number of participants at risk, the time over which they were observed, definitions of seriousness, and the method used to assign treatment relatedness. The record gives at-risk counts and affected-participant counts, but additional detail on these other aspects is not reported. A small number of events also provides limited information about uncommon harms.

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

The posted outcome describes average change in constipation symptoms after a 6-day treatment period among the participants included in the analysis. The mean PAC-SYM change is -0.22 in the lubiprostone group and -0.29 in the senna group. Since lower scores indicate improvement, both averages represent a decrease from baseline. The difference between those group summaries is small in the posted unadjusted re-calculation, and its confidence interval includes zero.

The registry also reports a sponsor-conducted independent-sample t test with p <0.05. That result conflicts with the recalculated p-value of 0.649, and the record flags the discrepancy. The sponsor’s analysis takes precedence, but its effect estimate and confidence interval are not reported. Thus, the sponsor’s p-value does not, by itself, tell readers the size or direction of the sponsor’s estimated difference. The posted means are descriptive, while the discrepancy limits how completely the statistical result can be reconstructed from the record.

The available result does not establish that one treatment caused a greater improvement, nor does it establish that the treatments have equivalent effects. In general, a nonzero point estimate with an interval crossing zero can reflect an uncertain difference; it is not proof that no difference exists. Conversely, a p-value below a conventional threshold does not, by itself, show that a difference is large or important to patients. The registry does not report a threshold for a clinically important PAC-SYM change.

The outcome concerns questionnaire-based symptom change at Day 7. It should not be read as evidence about every aspect of constipation, longer-term outcomes, or patients outside the described inpatient rehabilitation setting after orthopedic procedures. The record’s analysis population requires baseline and endpoint data, and the handling of missing data is not reported. These limits matter when considering how broadly to interpret the posted comparison.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it places a sponsor-reported significance result alongside a simple re-calculation from posted group summaries that does not agree with it. The registry’s own analysis reports p <0.05, while the unadjusted re-calculation gives p = 0.649. The record flags the discrepancy, making it important to distinguish the sponsor’s analysis from a secondary calculation rather than silently choosing the result that appears more plausible.

The example also shows why p-values should not be read without estimates and uncertainty. The arm means are -0.22 and -0.29, and the re-calculated mean difference is -0.07. The confidence interval for that re-calculation includes zero and spans effects in both directions. These details convey the scale and uncertainty of the contrast more directly than a threshold label alone. The sponsor’s estimate and interval are not reported, so the sponsor’s p-value cannot supply the same information.

The sample-size illustrations further demonstrate how planned sample size depends on the effect a study is designed to detect. Smaller assumed effects correspond to larger participant requirements. Because the illustrations begin with the observed effect, they are planning examples rather than evidence about the completed trial’s power or the sample-size rationale used when it was designed.

Finally, the registry describes an intention-to-treat analysis using participants with both baseline and endpoint data. This wording highlights the importance of reporting the analysis population and missing-data handling: the label alone does not explain how participants without endpoint data were treated. Such distinctions are central to assessing what a trial’s estimate represents.

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.

The registry record does not provide the sponsor’s estimated between-group effect or confidence interval, despite reporting a p-value below 0.05. This prevents a full assessment of the sponsor’s effect size and precision and makes it difficult to reconcile that result with the re-calculation from the posted means and standard deviations. The registry flags the discrepancy but does not explain its source.

The available primary outcome data include 28 participants per arm, while enrollment is listed as 64. The record does not explain the difference, identify reasons for missing endpoint data, or describe how missingness was handled. Although the analysis population is called intention-to-treat, it is defined as randomized participants with baseline and endpoint data; the record does not establish that every randomized participant contributed to the endpoint analysis.

There is no reported threshold for a clinically important PAC-SYM change, so the clinical significance of the mean changes cannot be determined from the outcome description alone. The registry also does not provide the original sample-size justification, a detailed analysis plan, or information about checks of the t-test assumptions. The sample-size illustrations are based on the observed effect and should not be mistaken for the study’s original design rationale.

Further details about treatment adherence, protocol deviations, concomitant care, and the implementation of allocation and masking are not reported in the registry record. The adverse-event entry gives limited event counts but not enough context to establish treatment relatedness or comparative safety. The findings are specific to the described postoperative inpatient rehabilitation population and the Day 7 symptom outcome; the record does not establish how they apply to other settings, populations, or follow-up periods.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Lubiprostone (Amitiza®) Vs. Standard Care in Opioid-induced Constipation After Surgery in Inpatient Rehabilitation — NCT00662363.
• Registry record last updated 2013-03-06; status last verified 2013-02.
• Results first posted 2013-02-21.
• 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

The registry record describes a completed randomized, quadruple-masked, parallel-group comparison of lubiprostone and senna for postoperative opioid-induced constipation among patients in inpatient rehabilitation after orthopedic procedures. The listed enrollment is 64. Treatment lasted 6 days, and the primary outcome was change in PAC-SYM symptoms at Day 7.

The posted mean changes are -0.22 for lubiprostone and -0.29 for senna, with 28 participants in each group. The registry reports an independent-sample t test with p <0.05, while the simple unadjusted re-calculation gives p = 0.649 and a mean difference of -0.07. The record flags the discrepancy, so the sponsor’s analysis takes precedence. Its effect estimate and confidence interval are not reported.

The outcome is a symptom questionnaire for which lower scores indicate improvement. Missing-data handling, the original sample-size rationale, and several details needed to fully interpret the sponsor’s analysis are not reported in the registry record.