Amitiza in Constipation Associated With PD (Parkinson's Disease): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Amitiza in Constipation Associated With PD (Parkinson's Disease) (NCT00908076), a randomized phase 4 study. The primary outcome is Change From Baseline to End of Study.

2. Clinical Question

3. Trial Design

The registry record describes NCT00908076 as a randomized, double-blind, parallel-group phase 4 trial of lubiprostone for constipation associated with Parkinson’s disease. Baylor College of Medicine is listed as sponsor. The record gives a total enrollment of 78, with the study starting in 2009-02 and primary completion in 2010-10; its status is completed. The posted primary outcome results include 27 participants in each arm. The enrollment figure and the arm denominators describe different record entries, and the registry does not explain the relationship between them.

Participants were assigned to Amitiza or matching placebo. Randomization supports a comparison between assigned groups, while double masking is intended to limit differences in expectations or outcome assessment related to knowledge of assignment. The record does not state how the random sequence was generated, whether allocation was concealed, which people were masked, or whether masking was assessed. Those details are not reported in the registry record.

The stated timeframe is baseline to end of study. The trial description refers to global impression of change, a stool diary, a visual analog scale of improvement, an UPDRS rating scale, and constipation questionnaires. However, the posted primary result is expressed as a percentage of participants with an event, specifically marked or very marked clinical global improvement. The analysis description also refers to a difference from baseline in spontaneous bowel movements. These descriptions do not identify the same outcome in exactly the same way, so the record leaves some uncertainty about the link between the stated primary endpoint, its planned analysis, and the binary result.

The registry record does not report further details about treatment duration, dose, adherence, eligibility criteria, or follow-up procedures. Such details should not be inferred from the trial title or the brief summary.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: A marked or very marked clinical global improvement. Ondo WG et al. Placebo-controlled trial of lubiprostone for constipation associated with Parkinson disease. Neurology 2012 May 22; 78:1650. - See more at: http://www.jwatch.org/jn201205290000006/2012/05/29/lubiprostone-constipation-parkinson-disease#sthash.ggWrS7Vq.dpuf

5. Baseline Characteristics

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

Population: Patients were included if they had PD using standard criteria, were aged 35 to 85 years, and met the ROME II criteria19 and scored at least 10 on the ROME II constipation assessment. Excluded if they had any other identifiable cause of constipation, use of opioids, anticholinergic agents, or antacids containing magnesium or aluminum salts.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline to End of Study, measured over Baseline to end of study.

The posted primary result is the percentage of participants who achieved a marked or very marked clinical global improvement. This is a binary endpoint: each participant either meets the stated improvement threshold or does not. It gives a direct summary of the proportion reaching a defined level of perceived improvement, rather than describing the amount of change on a continuous scale.

The registry labels the timeframe as baseline to end of study and describes the endpoint broadly in connection with global impression of change, stool diaries, a visual analog scale, the UPDRS rating scale, and constipation questionnaires. Its description of the analysis, however, says that primary efficacy data would be analyzed using a Student’s t-test with unequal variances on the difference from baseline in spontaneous bowel movements. That is a continuous change measure, not the binary percentage presented in the posted result. The record does not explain this mismatch.

For a binary endpoint, the percentage is useful because it reports how often the defined outcome occurred in each arm. It does not show the range or magnitude of changes among participants who did not meet the threshold, nor does it convey all aspects of bowel function or quality of life. The registry’s description of additional measures provides context about the intended clinical assessment, but it does not report results for those measures. The stated reason for using the endpoint beyond the description of clinical improvement is not reported in the registry record.

7. Results Posted in the Registry

• Amitiza: 63.0% (17/27) versus 18.5% (5/27) on control; risk difference 44.4 percentage points (95% CI 21.1 to 67.8), risk ratio 3.40 (1.46–7.89); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary percentage represent?

A binary endpoint classifies each participant according to whether a defined event occurred. In this trial, the event is a marked or very marked clinical global improvement. The arm-specific percentages describe the fraction meeting that definition among the participants included in the posted outcome. This is different from a mean change score: a mean uses the amount of change across observations, while a binary percentage reduces each observation to an event or no event.

That simplification can make an outcome easier to interpret, but it discards information. Participants just below the threshold and participants with no improvement are both counted as not meeting the event definition, even if their experiences differ. The threshold and the way clinical global improvement was assessed therefore matter. The registry names the threshold but does not report further details about its assessment or validation.

What does the risk difference say?

The risk difference is an absolute comparison of the event proportions in two arms. Its estimate is 0.444, and the reported confidence interval runs from 0.211 to 0.678. The estimate summarizes the observed contrast on the proportion scale; it does not account for other participant characteristics unless an adjusted analysis is specified. The record does not report an adjusted risk-difference analysis.

Absolute contrasts are distinct from relative measures such as a risk ratio or odds ratio. They answer a different question: how far apart are the observed proportions on the scale of the outcome? The registry also lists relative estimates, but the binary proportion and risk difference are sufficient to describe the primary posted comparison without treating a relative measure as interchangeable with an absolute one.

What can a p-value establish?

A p-value evaluates how compatible the observed data are with a specified null model, conditional on the test’s assumptions. The registry reports an unadjusted p-value of <0.001. “Unadjusted” indicates that the value is presented without an adjustment for multiple comparisons; it does not by itself identify the test or its assumptions. The record does not report the calculation method for this p-value. The endpoint description mentions a Student’s t-test with unequal variances for change in spontaneous bowel movements, but that is not clearly the analysis underlying the posted binary result.

General explanation: a small p-value is not a measure of effect magnitude, clinical relevance, or certainty that the alternative hypothesis is true. It also does not repair missing information about the endpoint, analysis population, or missing-data handling. The estimate and its interval should be considered alongside the p-value and the registry’s documentation.

Why might unequal variances matter?

The registry’s description names a Student’s t-test with unequal variances for the difference from baseline in spontaneous bowel movements. General explanation: a t-test that allows unequal variances does not require the two groups to have the same variability. This can be relevant when group spreads differ, because a test that assumes equal variances may represent uncertainty poorly if that assumption is not reasonable. The registry does not post the continuous outcome’s group means, standard deviations, test statistic, or result, so the suitability and consequences of that planned test cannot be evaluated from the posted primary result.

The mention of a t-test also highlights the distinction between the described analysis and the posted binary endpoint. A t-test ordinarily compares means, whereas the displayed primary result is a percentage meeting a threshold. The record does not explain whether the binary result is a separate summary, whether another analysis was used, or whether the endpoint description was updated. It is therefore safer to explain each as recorded rather than assume they describe an identical analysis.

What does last observation carried forward assume?

The analysis description specifies last observation carried forward for missing data in the intent-to-treat population. In this approach, a participant’s last observed value is used in place of a missing later value. General explanation: this procedure can preserve observations in an analysis, but it relies on the assumption that carrying the earlier measurement forward is a reasonable representation of the missing outcome. If outcomes change after the last measurement, the substituted value may misrepresent the later state. The method does not remove uncertainty caused by missing observations.

The registry does not report how many outcome observations were missing, why they were missing, when the last available observations occurred, or whether other missing-data analyses were conducted. It also does not explain how last observation carried forward was applied to the binary percentage shown in the results. The stated approach should therefore be treated as the registry’s analysis description, not as enough information to reconstruct the posted result.

What is reported about the analysis population?

The primary description names an intent-to-treat population. General explanation: an intent-to-treat approach usually aims to analyze participants according to their randomized assignment, regardless of later treatment adherence. The precise operational definition, exclusions, and handling of participants without outcome data are not reported in the registry record. The arm denominators for the posted outcome are 27 and 27; the record does not explain how these relate to the stated enrollment of 78 or to the intent-to-treat population.

The registry lists no posted analyses and no posted secondary outcome results. This does not establish that no analysis was performed; it means the record does not present additional analyses for readers to inspect. No multiplicity procedure, interim analysis plan, or detailed statistical analysis plan is reported in the registry record.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.444, with a reported confidence interval from 0.211 to 0.678. A confidence interval represents uncertainty around an estimated contrast under the method used to construct it. General explanation: the interval is not a range containing every plausible value with certainty, and its interpretation depends on the statistical procedure and confidence level. The confidence level is not reported in the registry record.

The interval’s endpoints are both above zero, so the reported interval does not include a zero risk difference. Its width also indicates that the estimate is not pinpointed to a single value. The interval allows a range of contrasts rather than supporting a claim that the true contrast is exactly the point estimate. The posted p-value, <0.001, is consistent with a strong departure from a zero-difference null under the test used, but the registry does not report the method used for that p-value.

Precision depends partly on the amount of information in the outcome data. The posted outcome denominators are 27 in each arm, while enrollment is recorded as 78. The registry does not explain this difference, report the number of missing observations, or state how many randomized participants were included in the posted result. These gaps make it difficult to assess the contribution of missing data or exclusions to precision.

General explanation: a confidence interval concerns statistical uncertainty, not every source of uncertainty. It does not account automatically for differences in outcome measurement, potential bias, incomplete reporting, or limitations in how participants were selected. Since the registry does not report the interval level or full method for the binary estimate, the numerical interval can be described as posted, but its construction cannot be independently assessed from the record.

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 80% power and participants per arm for 90% power. They are not total events. For the planning illustration based on 100% of the observed effect, the figures are 18 per arm for 80% power and 24 per arm for 90% power. Planning for 75% of the observed effect gives 32 per arm for 80% power and 42 per arm for 90% power. Planning for 50% of the observed effect gives 66 per arm for 80% power and 87 per arm for 90% power.

In plain language, the 80% column gives the number of participants per arm estimated for a design with an 80% chance of detecting the stated effect under the listed assumptions, if that effect is present. The 90% column gives the corresponding planning number for 90% power. Higher target power generally calls for more participants under otherwise unchanged assumptions. The rows reduce the effect being planned for from the observed effect to a smaller fraction; smaller effects need more participants to detect under the same assumptions.

The listed 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 the stated fraction of the observed effect. These are planning illustrations computed from the observed effect, not the trial’s original design assumptions, which are not reported in the registry record. This is not observed 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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.185; treatment proportion p₁ = 0.630; average p̄ = 0.407.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6949 = 1.3619.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6197 = 0.5216.
4. n = (1.3619 + 0.5216)² / (0.4444)² = 18.0, rounded up to 18 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 <- 0.185185185185185                          # control proportion
p1 <- 0.62962962962963                          # 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(908076)
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 = 18 (80% power), 24 (90% power)
 75% of observed effect: n per arm = 32 (80% power), 42 (90% power)
 50% of observed effect: n per arm = 66 (80% power), 87 (90% power)
power.prop.test: 18 
Simulated power at n = 18 per arm: 0.830 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry reports an unadjusted p-value for the primary posted comparison. It does not report a procedure for controlling multiplicity across outcomes, repeated measurements, or other analyses. The trial description names several measures, including global impression of change, stool diary, visual analog scale of improvement, UPDRS rating scale, and constipation questionnaires. However, no secondary outcome results are posted, and the record does not explain how these measures were prioritized or whether multiple comparisons were planned.

Multiplicity matters because examining multiple outcomes or analyses can increase the chance that at least one result appears unusual by chance. General explanation: methods such as adjustment of significance thresholds or a prespecified hierarchy can address particular multiplicity concerns, but their use cannot be assumed. The registry does not report such a method. The unadjusted p-value should therefore be read as the reported result for its comparison, not as evidence that multiplicity has been addressed.

The analysis description states that missing data would be handled using last observation carried forward in an intent-to-treat population. It does not report the amount or pattern of missing data, reasons for missingness, or sensitivity analyses under alternative assumptions. It also does not explain whether this procedure was used for the posted binary outcome. The number of observations included in the primary result differs from the stated enrollment, but the registry does not describe participant flow or the reasons for that difference.

The registry record does not report an interim analysis plan, a detailed statistical analysis plan, or a prespecified approach to deviations from the planned analysis. It also does not report whether the result was adjusted for baseline characteristics. These omissions do not establish that such procedures were absent; they mean the posted record does not document them. The listed analysis description and the binary result should be kept distinct unless additional documentation clarifies their relationship.

14. 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 non-serious adverse event terms

The serious adverse-event entries report 0 events among 27 participants at risk in the Amitiza arm and 0 among 27 at risk in the placebo arm. The registry lists no serious adverse-event terms. The absence of a recorded serious event in these entries does not establish that no serious event occurred outside the registry’s reported information or that the interventions have no risk.

Other adverse events are recorded for 3 of 27 participants at risk in the Amitiza arm and 1 of 27 at risk in the placebo arm. These counts are descriptive entries in the registry record. The record does not provide event types, severity, timing, duration, attribution, or a statistical comparison for these other events. It also does not report deaths; the death entries are not populated with a numerical value.

Safety counts should be interpreted in light of what the record captures and the population followed. Definitions and observation periods for adverse events are not reported in the registry record. These figures therefore summarize the posted entries, rather than establishing a complete safety profile or supporting medical advice.

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

The posted results show that 64.0% of participants in the Amitiza arm and 18.5% in the placebo arm met the registry’s definition of marked or very marked clinical global improvement. The reported risk difference is 0.444, with a confidence interval from 0.211 to 0.678, and the unadjusted p-value is <0.001. These values describe the observed comparison recorded for the trial.

The results do not establish the size of an effect in every population or care setting. The registry does not report detailed eligibility criteria, treatment adherence, outcome-assessment procedures, missing-data counts, or the confidence interval level. The reported enrollment is 78, while the posted outcome denominators are 27 in each arm; the record does not explain how those numbers relate. Such limitations make it important not to treat the posted proportions as a complete account of the trial or as a guarantee of individual outcomes.

The record also does not provide secondary outcome results, despite mentioning bowel movement consistency, quality of life, and motor symptoms in the study summary. Its primary analysis description refers to change in spontaneous bowel movements and a t-test, while the posted primary result is a binary clinical improvement percentage. The record does not clarify the relationship between these descriptions. This limits how specifically the statistical result can be tied to the described analysis plan.

The adverse-event entries report no serious events in either arm and record other adverse events in both. They do not give enough detail to determine the nature, severity, or timing of those events. Taken together, the registry documents a difference in the posted binary outcome and limited adverse-event counts; it does not provide a complete basis for claims about all outcomes, all possible harms, or an individual’s likely experience.

16. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions and reported results must be read together. A percentage achieving a threshold for clinical global improvement is not the same quantity as a mean change in spontaneous bowel movements. Both can be relevant ways to describe clinical outcomes, but they answer different statistical questions. When a registry’s analysis description refers to one measure and the posted primary result presents another form, readers need clarification before treating the result as a direct report of the stated analysis.

The record also shows how absolute estimates and uncertainty contribute different information. The risk difference of 0.444 expresses the observed contrast on an absolute scale, while the interval from 0.211 to 0.678 communicates uncertainty around that estimate. The p-value, <0.001, addresses compatibility with a null model under the method used; it does not indicate the magnitude, practical importance, or reliability of the finding by itself.

Planning assumptions matter as well. The registry’s sample-size illustrations use fractions of the observed effect and explicitly identify their assumptions. They cannot substitute for the original sample-size rationale, which is not reported. Because smaller planned effects require more participants under otherwise unchanged assumptions, planning directly from an observed effect should be treated cautiously; the illustration itself is not observed power.

Finally, transparent reporting of denominators, missing data, analysis populations, secondary outcomes, and safety-event definitions supports meaningful interpretation. The registry contains a primary result and some safety counts, but leaves several of these details unreported. Statistical reading is not only about interpreting a p-value; it also involves checking whether the endpoint, analysis, uncertainty, and available record align.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several trial-specific details that could affect interpretation are not reported in the registry record. These include eligibility criteria, the dose and duration of treatment, adherence, the method of sequence generation, allocation concealment, and which parties were masked. The record states that the study was double-blind but does not describe the practical masking procedures or whether masking was assessed.

The registry gives enrollment of 78 and posted primary outcome denominators of 27 in each arm, without a participant-flow account explaining the relationship. It does not report the number or timing of missing outcome measurements, reasons for missingness, or whether participants were excluded from the posted analysis. Although the primary description names an intent-to-treat population and last observation carried forward, the application of that approach to the binary result is not explained.

There is also a reporting ambiguity between the stated primary analysis of change from baseline in spontaneous bowel movements using a Student’s t-test with unequal variances and the posted percentage achieving marked or very marked clinical global improvement. The record does not clarify whether these refer to separate outcomes or how the posted binary estimate was analyzed. No secondary outcome analyses are posted, despite the study summary describing additional clinical measures.

The confidence interval level and the method used for the posted unadjusted p-value are not reported. The record does not provide a multiplicity approach, interim analysis plan, or detailed statistical analysis plan. For safety, it does not report event terms for other adverse events, definitions, timing, or attribution, and death entries are not numerically populated. These limitations concern what the registry documents; they do not establish that unreported procedures or data were absent from the trial.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Amitiza in Constipation Associated With PD (Parkinson's Disease) — NCT00908076.
• Registry record last updated 2022-12-29; status last verified 2022-12.
• Results first posted 2016-08-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT00908076 is a completed, randomized, double-blind, parallel phase 4 trial of Amitiza (lubiprostone) versus matching placebo for constipation associated with Parkinson’s disease. Baylor College of Medicine is listed as sponsor. The recorded enrollment is 78, with a start date of 2009-02 and primary completion in 2010-10.

The posted primary outcome reports marked or very marked clinical global improvement in 64.0% of the Amitiza arm and 18.5% of the placebo arm. The risk difference is 0.444, with a confidence interval from 0.211 to 0.678; the unadjusted p-value is <0.001. The registry’s analysis description refers to change in spontaneous bowel movements, leaving the relationship between that description and the posted binary result unclear.

The serious adverse-event entries report 0 in each arm. Other adverse events are recorded for 3 participants in the Amitiza arm and 1 in the placebo arm. Secondary analyses, further outcome results, and several details needed to assess the analysis and safety reporting are not reported in the registry record.