A Safety and Efficacy Study of LiRIS® in Females With Interstitial Cystitis/Bladder Pain Syndrome: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Safety and Efficacy Study of LiRIS® in Females With Interstitial Cystitis/Bladder Pain Syndrome (NCT02411110), a randomized phase 2 study. The primary outcome is Change From Baseline in Daily Average Bladder Pain to Treatment 1 Week 4 Follow-up.

2. Clinical Question

The clinical question is about a short-term patient-reported pain outcome in the first treatment period. Baseline was the average pain over Days -7 to 0, and the endpoint was measured through Treatment 1 Week 4. The primary outcome description specifies an analysis of covariance model that adjusted for baseline pain and included treatment group and stratification factors. The registry record identifies those factors as age group (< 40 years or ≥ 40 years) and baseline bladder pain NRS (≤ 6 or > 6).

This question concerns the difference between randomized treatment groups, not simply whether pain scores changed within either group. Changes in each arm describe the observed direction and precision of its estimated outcome; the between-group analysis addresses the comparative question. The registry record reports an adjusted least squares mean difference and a p-value, but does not report a confidence interval for that difference. The record does not state an explicit contrast coding for the reported difference.

3. Trial Design

NCT02411110 is listed as a Phase 2, randomized, double-blind, parallel study. The official title describes it as multicenter and placebo-controlled. The sponsor is Allergan, and the status is completed. The registry record lists enrollment of 131 participants. The recorded start date is 2015-05-21 and the primary completion date is 2016-10-12.

The Treatment Period 1 comparison was LiRIS® versus a matching placebo device. The LiRIS® device was inserted into the bladder by cystoscopy on Day 0 and removed on Day 14. The placebo device was inserted and removed on the same stated schedule. The arm descriptions also include an optional Treatment Period 2. Participants assigned to LiRIS® in Treatment Period 1 could receive LiRIS® in Treatment Period 2; participants assigned to placebo in Treatment Period 1 could receive LiRIS® in the optional second period. The primary endpoint, however, concerns Treatment 1 Week 4 follow-up.

The primary outcome lists 65 participants in the LiRIS® arm and 66 in the placebo arm. The stated analysis population includes randomized participants who received Treatment 1. The registry record describes an ANCOVA model with baseline value as a covariate and treatment group and the stated stratification factors as factors. Further details of sequence generation, allocation concealment, the conduct of masking, and the sample-size justification are not reported in the registry record.

The design supports a randomized comparison of the first treatment period, but the posted primary result is limited to the specified pain outcome and follow-up. The record lists one outcome measure and one analysis. It does not report secondary outcomes in the posted results. Because the second treatment period was optional and included active treatment after either first-period assignment, it should not be conflated with the primary randomized comparison.

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.

Treatment Period 1

Reasons for not completing this period:

Treatment Period 2

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Modified Intent-to-Treat (mITT) Analysis Population included all participants who were randomized and received Treatment 1.

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 From Baseline in Daily Average Bladder Pain to Treatment 1 Week 4 Follow-up, measured over Baseline (Days -7 to 0) to Treatment 1 Week 4.

The primary endpoint is change from baseline in daily average bladder pain through Treatment 1 Week 4. Participants recorded pain over the previous 24-hour period using a 7-day pain assessment tool and an 11-point Numeric Rating Scale, ranging from 0, no pain, to 10, worst pain. The daily reports over the 7-day period were averaged. Baseline is specified as Days -7 to 0.

Change scores retain the direction of the scale: a negative change from baseline indicates improvement. The outcome therefore describes how the average reported pain score changed from baseline by the stated follow-up, rather than a single pain report on one day. Averaging reports across the stated period can represent the participant’s pain over that period more fully than relying on a single daily entry. That is a general measurement rationale, not a claim about the completeness of reporting in this trial.

A patient-reported pain measure is relevant to a trial whose stated condition includes bladder pain syndrome because it directly captures participants’ reported symptom experience. The result is still specific to this instrument, baseline definition, treatment period, and follow-up. It does not, on its own, describe other symptoms, functioning, longer-term outcomes, or safety. Those outcomes are not part of the stated primary endpoint. The registry record does not report a clinically important difference threshold for this scale, so the numerical changes cannot be classified against such a threshold from the posted record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint was analyzed using analysis of covariance, or ANCOVA. In general, ANCOVA compares outcome means between groups while accounting for one or more covariates. Here, the registry record says that baseline value was used as a covariate and that treatment group and the stratification factors were included as factors. The stated factors were age group (< 40 years or ≥ 40 years) and baseline bladder pain NRS (≤ 6 or > 6). Including baseline in the model can improve the precision of a comparison when baseline outcome values are associated with follow-up outcome values. That is a general statistical rationale; the registry record does not report the model diagnostics or the actual association between baseline and follow-up in this trial.

What does a least squares mean represent?

A least squares mean is a model-based estimate of an arm’s mean outcome after accounting for the covariates and factors included in the model. It is not necessarily the same as a simple arithmetic average of all observed scores. The posted arm estimates are -1.2 for LiRIS® and -1.5 for placebo, with 90% confidence intervals of -1.77 to -0.58 and -2.14 to -0.87, respectively. These adjusted arm estimates describe change from baseline under the fitted model. They should not be mistaken for the treatment contrast itself.

The comparative estimate is reported separately as a least squares mean difference of 0.34. A difference is formed within the fitted model, using a stated contrast between groups. The registry record does not report the contrast coding explicitly, so the direction represented by the positive sign should not be assigned beyond the record’s wording. A confidence interval for this difference is also not reported. The difference is the quantity relevant to the randomized comparison; its uncertainty is important for judging the range of effects compatible with the result.

How should the confidence intervals be read?

A confidence interval describes uncertainty around an estimate under the statistical model and its assumptions. The posted 90% confidence intervals accompany the two arm-specific least squares means. They show that those estimated mean changes are not known with exact precision. They do not provide the interval for the difference between arms. A confidence interval for a contrast is generally calculated using the uncertainty and covariance of the model estimates; it cannot be recovered reliably just by comparing the endpoints of separate arm intervals.

The use of a 90% interval is reported for the arm estimates. The registry record does not state why this interval level was selected, nor does it report a confidence interval for the ANCOVA difference. It is therefore not possible to use the posted arm intervals as a substitute for the missing contrast interval or to infer the trial’s decision rule from them.

What does the p-value say?

The reported p-value is 0.505 for the ANCOVA analysis. In general, a p-value describes how compatible the observed data are with a specified null hypothesis under the model; it is not the probability that the null hypothesis is true, nor is it a measure of clinical importance. A p-value does not tell readers the plausible range of the treatment difference, which is why an interval estimate for the contrast is useful. The registry record does not report the decision threshold for this analysis, so the p-value should be presented as reported rather than converted into a claim about a prespecified significance decision.

The record lists the method as ANCOVA and the result as a least squares mean difference. It does not report further details about model assumptions, residual checks, transformations, robust variance estimation, or sensitivity analyses. Such details matter because model-based estimates rely on assumptions about how the outcome relates to the covariates and how the residual variation is represented. Those assumptions cannot be assessed from the summary result alone.

Why can over-dispersion matter?

Over-dispersion is a general concern for outcomes whose variation is greater than a model assumes, especially certain count outcomes. If it is ignored, standard errors can be too small and intervals or tests can be misleading. The primary endpoint in this trial is a change in an averaged pain rating, not a count outcome, and the registry does not report an over-dispersion assessment. This general concept should not be read as evidence that over-dispersion occurred in this analysis.

The sample-size assumptions recorded for planning illustrations describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with effects expressed as stated fractions of the observed effect. These are planning illustrations based on the observed effect, not the trial’s own design justification and not observed power. The registry record does not report the trial’s sample-size justification, target effect, or interim analyses. No planning result should be treated as a property of the trial’s actual conduct.

Finally, the analysis population is defined as randomized participants who received Treatment 1. The registry record does not describe how missing pain assessments were handled, whether all eligible assessments were included, or whether other analysis populations were examined. These details affect how readers assess the estimates and their generalizability. They are not resolved by knowing that ANCOVA was used.

10. Analyses Reported by the Sponsor

The sponsor’s posted analysis is the ANCOVA result: a least squares mean difference of 0.34 with p = 0.505. The registry record does not report a confidence interval for this difference or explicit contrast coding. The arm-level least squares means and their 90% confidence intervals provide context for the outcome in each group, but they are not a separate estimate of the randomized between-group contrast.

The posted information does not give a separate recalculated two-arm estimate for this endpoint. It therefore does not support a numerical comparison between the sponsor’s adjusted result and an unadjusted re-calculation. No discrepancy is identified in the registry record. The sponsor’s analysis should be described as the available comparative analysis, without implying that an absent re-calculation independently confirms it.

In general, a simple re-calculation from summary statistics would not reproduce an ANCOVA estimate because the sponsor’s model adjusts for baseline and stratification factors. Even when summary statistics allow an unadjusted comparison, that calculation answers a different question from a covariate-adjusted analysis and can differ for legitimate reasons. In this case, the record does not provide the arm-level details needed to present such a separate comparison. The reported 0.34 and p = 0.505 should therefore remain the focus, alongside the stated limits on its precision and interpretation.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and one analysis. It does not list secondary outcomes in the posted results. This limits the multiplicity questions that can be assessed from the available outcome record, but it does not establish that no additional outcomes were measured or analyzed outside what is posted. The protocol’s complete outcome hierarchy and any multiplicity procedure are not reported in the registry record.

Multiplicity arises when several hypotheses or outcomes are tested, because the chance of at least one apparently unusual result can increase as the number of tests grows. In general, prespecified testing strategies or adjustment procedures can address that issue. The record does not report a multiplicity adjustment, a testing sequence, or whether the primary analysis was intended to be confirmatory. The single listed analysis should not be taken as proof that no other analyses were conducted.

The analysis population is the modified Intent-to-Treat population: randomized participants who received Treatment 1. The arm counts for the primary outcome are 65 for LiRIS® and 66 for placebo. The registry record does not explain how randomized participants who did not receive Treatment 1 would be handled, nor does it report a full Intent-to-Treat analysis.

Missing-data handling is not reported in the registry record. In general, missing outcome data can affect an estimate when the probability of missingness is related to outcomes or participant characteristics. Relevant details could include how incomplete 7-day pain records were scored, whether partial records were accepted, how missing follow-up values were treated, and whether sensitivity analyses were used. None of those procedures is described in the posted primary outcome information, so the potential effect of missing data cannot be assessed.

Interim analyses and stopping rules are also not reported in the registry record. Their absence from the record is not evidence that they did or did not occur. The available information is therefore insufficient to assess how the analysis population, missing observations, or any multiple testing were handled beyond the stated primary ANCOVA model.

12. 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: urinary tract infection (2.1%), urinary tract stoma complication (2.1%), kidney infection (1.5%), nephrolithiasis (1.5%), pelvic congestion (1.5%).

The registry lists serious adverse events by treatment-period group. In Treatment Period 1, the placebo group had 1 serious adverse event among 66 participants at risk, and the LiRIS® group had 2 among 65 participants at risk. For the groups with optional Treatment Period 2, the registry lists 1 among 48 participants in the placebo-then-LiRIS® group and 0 among 50 in the LiRIS®-then-LiRIS® group. These are counts within the registry’s stated groupings; they should not be combined into a single comparison because the groups refer to different treatment-period histories.

The registry also lists serious-event terms with percentages: urinary tract infection, 2.1%; urinary tract stoma complication, 2.1%; kidney infection, 1.5%; nephrolithiasis, 1.5%; pelvic congestion, 1.5%; and procedural pain, 1.5%. The record does not provide enough context in this summary to assign these terms to a particular randomized contrast or to infer causality. An event recorded during a trial is not, by itself, proof that the intervention caused it.

Deaths are not reported in the registry record. The meaning of “other” adverse events is not further characterized in the posted information. The Treatment Period 2 groups are optional-period groups, so their counts should not be read as a direct randomized comparison of the original assignments. These figures describe recorded events, not a complete assessment of safety or risk.

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

The primary result describes adjusted changes in self-reported average bladder pain through Treatment 1 Week 4 among participants who were randomized and received Treatment 1. The registry reports negative arm-specific least squares mean changes in both the LiRIS® and placebo groups. It also reports an ANCOVA least squares mean difference of 0.34 and p = 0.505. The result is thus a short-term comparison on a specific pain scale and within a defined analysis population.

The posted estimates do not establish a precise range for the between-group difference because its confidence interval is not reported. The p-value is not a measure of the size or importance of a difference. The record also does not report a threshold for a clinically important change on the pain scale. As a result, the summary does not support a conclusion about whether the estimated contrast would be meaningful to participants, even apart from statistical uncertainty.

The result does not answer whether there was a difference in outcomes beyond Treatment 1 Week 4, whether effects persisted, or what happened for other symptoms or outcomes not listed in the posted results. It does not provide a comparative estimate for optional Treatment Period 2. Nor does the primary pain outcome summarize serious adverse events or other safety outcomes.

Randomization and masking are features of the stated design, but details of their implementation are not reported. The endpoint is patient-reported, and the record does not describe missing-data procedures or sensitivity analyses. Those omissions matter when considering how representative the reported estimate is of all randomized participants and how robust it is to incomplete observations.

The appropriate reading is bounded: the registry record reports the arm estimates and an adjusted comparison for the specified first-period pain outcome. The information does not justify extending that result to unreported outcomes, longer follow-up, or different populations. It should not be treated as medical advice, nor as a stand-alone determination that treatment does or does not work.

14. Why This Trial Matters Statistically

This trial illustrates why a within-arm change and a between-arm comparison answer different statistical questions. The registry reports a negative estimated pain change in each arm. Those changes describe outcomes relative to baseline, but the randomized comparison is represented by the ANCOVA least squares mean difference. Readers should not infer comparative benefit merely because an arm’s score changed in a favorable direction: changes can also occur in the placebo group, as the posted estimates demonstrate.

It also shows why the uncertainty around the contrast matters. Arm-level confidence intervals are reported, while the contrast confidence interval is not. The contrast is the quantity that most directly addresses the randomized question, and its interval would communicate its precision. A p-value alone cannot replace that information. This distinction is important in clinical research reporting, where estimates, uncertainty, and clinical context should be considered together.

The use of ANCOVA makes the model specification relevant to interpretation. The record identifies baseline adjustment and stratification factors, but does not report diagnostics, missing-data procedures, or sensitivity analyses. Clear reporting of these details helps readers judge whether the statistical approach matches the data and whether conclusions depend on modeling choices. Their absence does not show that the analysis was inappropriate; it limits independent assessment.

The optional second treatment period is another useful design feature to distinguish from the primary randomized comparison. The primary outcome concerns Treatment 1 Week 4, whereas the second period was optional and used LiRIS® in both described sequences. The first-period comparison and optional-period safety groupings therefore have different interpretive roles.

Finally, the single posted outcome and analysis do not settle questions about multiplicity, unreported outcomes, or analysis choices beyond the listed model. Careful interpretation stays close to the endpoint, population, timeframe, and comparative estimate that the registry actually reports.

15. 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 result is limited to one posted outcome measure: change from baseline in daily average bladder pain through Treatment 1 Week 4. The registry record does not provide secondary outcome results. It therefore cannot support conclusions about other symptoms, functioning, quality of life, or longer-term outcomes from the posted results.

The comparative ANCOVA estimate is reported without a confidence interval, and explicit contrast coding is not reported. This makes it harder to judge the estimate’s precision and direction from the record alone. The arm-level intervals do not resolve that limitation because they are intervals for separate estimates, not for the between-group difference. A clinically important difference threshold for the pain scale is also not reported.

The stated analysis population includes randomized participants who received Treatment 1, rather than all randomized participants. The record does not report missing-data methods, the handling of incomplete diary records, sensitivity analyses, or results from other analysis populations. These omissions limit assessment of possible bias from missing observations or from restricting the analysis population.

Although the official title describes a multicenter study, the registry summary does not report center-level results or whether center was included in the analysis model. Details of randomization implementation, allocation concealment, masking procedures, model diagnostics, sample-size justification, interim analyses, and multiplicity procedures are not reported in the registry record. The lack of these details prevents independent evaluation of those aspects; it does not establish that the procedures were absent.

Finally, the primary result addresses the first treatment period and a specified short-term follow-up. The optional second period has different treatment histories and is not a continuation of the same first-period randomized contrast. The posted summary should therefore be interpreted within its stated population, endpoint, and timeframe, rather than generalized to unreported outcomes or periods.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Safety and Efficacy Study of LiRIS® in Females With Interstitial Cystitis/Bladder Pain Syndrome — NCT02411110.
• Registry record last updated 2018-01-04; status last verified 2017-12.
• Results first posted 2018-01-04.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT02411110 is a completed Phase 2 randomized, double-blind, parallel study described as multicenter and placebo-controlled. It enrolled 131 participants and compared LiRIS® with a matching placebo device during Treatment Period 1. The primary endpoint was change from baseline in daily average bladder pain through Treatment 1 Week 4, measured with an 11-point Numeric Rating Scale.

The primary analysis population comprised randomized participants who received Treatment 1. The registry reports least squares mean changes of -1.2 for LiRIS® and -1.5 for placebo, each with a 90% confidence interval, and an ANCOVA least squares mean difference of 0.34 with p = 0.505. A confidence interval for the difference is not reported. The registry lists one outcome measure and one analysis; secondary results, missing-data procedures, and several design and analysis details are not reported in the registry record.