This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Tolvaptan to Reduce Length of Stay in Hospitalized Patients With Cirrhosis and Hyponatremia (NCT01890694), a randomized phase 4 study. The primary outcome is Length of Hospital Stay.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01890694 |
| Phase | Phase 4 |
| Condition | Hyponatremia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 2 |
| Arms in the primary outcome | n/a |
| Lead sponsor | NYU Langone Health |
| Results posted | Yes |
2. Clinical Question
Population
Hospitalized patients with cirrhosis and hyponatremia were the intended population. The study description focused on people admitted for treatment of low blood sodium and proposed starting study treatment within 2 days of admission. The registry record does not report characteristics of enrolled participants, eligibility details, or a collected analysis population. The primary outcome description explicitly states that data were not collected because the trial ended prematurely.
Intervention
The active-treatment group was to receive tolvaptan once daily. The study description proposed its use for treating hyponatremia during the hospitalization. The registry record does not report a dose, treatment duration, adherence, or participant-level exposure.
Comparator
The comparator group was to receive placebo once daily. Both arm descriptions also specify assessments involving vital signs, laboratory tests, and ascites. The record does not report whether these procedures were completed for enrolled participants.
Primary question
The primary question was whether assignment to tolvaptan, compared with placebo, was associated with shorter length of hospital stay. The study also aimed to assess other complications of cirrhosis, including hepatic encephalopathy and ascites. Because primary outcome data were not collected, the registry record cannot answer the treatment-comparison question.
3. Trial Design
NCT01890694 is listed as a Phase 4, randomized, double-blind, parallel-group study sponsored by NYU Langone Health. Its registered title describes a study of tolvaptan to reduce length of stay in hospitalized patients with cirrhosis and hyponatremia. The start is listed as 2012-03 and primary completion as 2015-03. The status is terminated.
The planned comparison was once-daily tolvaptan versus once-daily placebo. The arm descriptions list vital signs and laboratory testing, along with ascites assessment, as procedures during visits. Listed laboratory assessments include chemistry, hematology, liver function, urine electrolytes, renin, and copeptin. The registry also lists hepatic encephalopathy assessment, quality-of-life assessment, and edema among the interventions or assessments. These registrations describe intended study activities; they do not establish which assessments were completed.
The enrollment recorded is 2. The primary outcome description explains that the study ended before adequate enrollment because the performance period expired and the funding sponsor did not continue support. It also states that data were not collected due to premature termination. Thus, the design describes a planned randomized comparison, but the record does not contain a completed efficacy analysis.
Details such as the randomization procedure, allocation concealment, exact masking procedures, treatment dose, and treatment duration are not reported in the registry record. The record also does not report a protocol-defined sample-size target or the statistical plan for the primary comparison. The stated expected average hospital stay of 2 weeks is a timeframe description, not a result from observed participants.
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
| Stage | Placebo | Tolvaptan |
|---|---|---|
| STARTED | 1 | 1 |
| COMPLETED | 0 | 0 |
| NOT COMPLETED | 1 | 1 |
Analysis population for the primary outcome, as defined in the registry: Data were not collected due to premature termination of the trial
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.
| Characteristic | Unit | Placebo | Tolvaptan | Total |
|---|---|---|---|---|
| Participants analysed | 1 | 1 | 2 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 1 | 1 | 2 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Sex: Female, Male: Female | Participants | 0 | 0 | 0 |
| Sex: Female, Male: Male | Participants | 1 | 1 | 2 |
6. Primary Endpoint
The registry lists the primary outcome as Length of Hospital Stay, measured over participants will be followed for the duration of hospital stay, an expected average of 2 weeks.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was length of hospital stay. It was intended to measure how long participants remained hospitalized, with follow-up described as continuing for the duration of the hospital stay. Length of stay is a direct measure of inpatient time and can be relevant to a study asking whether an intervention is associated with a shorter admission. It does not, by itself, explain why a stay is longer or shorter, nor does it capture every health outcome that may matter to participants.
For this trial, length of stay was a planned endpoint rather than an observed result. The registry record says data were not collected because the trial terminated prematurely. No group-specific lengths of stay, treatment-effect estimate, uncertainty interval, or test of the primary comparison is reported in the registry record. The stated expected average of 2 weeks is part of the outcome timeframe description; it is not an observed average or an estimate of the difference between groups.
The endpoint description does not specify how discharge was defined, whether transfers or deaths would affect the endpoint, or what summary measure and statistical model were intended. Those details are not reported in the registry record. As a result, the clinical intent of the endpoint is clear, but its measurement and analysis cannot be evaluated from the posted record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record does not provide a primary efficacy result. Although the study was designed to compare tolvaptan with placebo on length of hospital stay, its primary outcome description states that data were not collected because the trial ended prematurely. There is therefore no reported group comparison, effect estimate, confidence interval, or hypothesis-test result for the primary endpoint. The record cannot establish the direction or size of any effect on hospital stay.
The registered enrollment is 2, but that figure should not be mistaken for a completed analysis population. The record does not say that primary outcome data were available for those enrolled, and it explicitly says that such data were not collected. A planned randomized design does not yield an interpretable treatment-effect estimate when the endpoint observations needed for the comparison are absent.
The study status is terminated. The primary outcome description attributes this to the performance period expiring, inadequate enrollment, and the funding sponsor not continuing support. These circumstances explain why the planned question remained unanswered in the registry record; they do not provide evidence for or against an effect of tolvaptan.
The record lists secondary outcomes including hepatic encephalopathy severity and ascites, but no analysis results are reported. It records 14 outcome measures and 0 analyses. Repeated listings of hepatic encephalopathy severity do not supply outcome values or clarify the assessment schedule. No results are available to characterize changes in those outcomes or compare arms.
The adverse-event information is separate from the missing primary endpoint. It records 0 serious adverse events and 0 other adverse events for the reported at-risk entries in each arm. The listed at-risk counts are 1 for each category in each arm. These entries describe the registry’s recorded adverse-event counts; they do not replace an efficacy analysis, and they provide only limited safety information. Death counts are not reported in the registry record.
In statistical terms, the main issue is not an imprecise estimate but the absence of a primary outcome estimate. A confidence interval would ordinarily describe the uncertainty around an estimated treatment effect, but no such estimate is reported. No p-value is reported either. Neither a null result nor evidence of benefit or harm can be inferred from the lack of a result.
9. Statistical Methods Explained
What does randomization mean in this study?
Randomization means that assignment to the intervention groups is intended to use a chance-based process. In general, this can reduce systematic differences between groups at the point of assignment, helping make a treatment comparison interpretable. The registry lists this study as randomized and parallel, with tolvaptan and placebo arms. It does not describe the sequence-generation method, how assignment was concealed, or whether the intended process was completed. Those details are not reported in the registry record.
A parallel-group design assigns participants to a group and compares outcomes across groups rather than switching each participant between treatments. Double masking generally means that treatment assignment is concealed from participants and specified study personnel, reducing the risk that knowledge of assignment changes behavior or outcome assessment. The record labels the study double-blind but does not identify exactly who was masked or how masking was maintained.
How would a treatment comparison be summarized?
For a length-of-stay endpoint, an analysis could summarize the observed duration in each arm and estimate a contrast between them. The meaning of that contrast depends on the chosen summary, such as a difference in means or another measure, and on the analysis model. The registry record does not report the planned summary measure, model, adjustment variables, or test for the primary endpoint. Since primary outcome data were not collected, the record reports no treatment-effect estimate to interpret.
In general, a confidence interval gives a range of values compatible with the data and the assumptions of an analysis. Its width reflects statistical precision: wider intervals indicate more uncertainty about the magnitude of an estimated effect. A p-value, in general, measures how unusual data at least as incompatible with a specified null hypothesis would be under that hypothesis and the analysis assumptions. Neither statistic is reported for the primary outcome. Without the underlying outcome data and analysis, there is no basis for describing either precision or statistical evidence for a treatment difference.
Why can the distribution of hospital stay matter?
As a general statistical consideration, duration outcomes may have an asymmetric distribution: a large share of observations may be relatively short while a smaller share are much longer. In that setting, a mean can be influenced by long stays, and alternative summaries or models may answer different questions. If participants have incomplete follow-up or an event such as discharge is handled as a time-to-event outcome, the chosen analysis must also account for how follow-up ends. These are general considerations, not methods documented for this trial. The intended distributional assumptions and handling of incomplete or competing outcomes are not reported in the registry record.
Over-dispersion is a general issue in count or rate data when variation exceeds what a simple model assumes. It is not a reported feature of this endpoint, which is described as length of stay. More broadly, selecting a model requires knowing the outcome’s form, its observed distribution, the estimand of interest, and the data structure. The record does not provide these data or a documented model choice, so no assessment of model fit or assumptions is possible.
What do the planning assumptions tell us?
The record describes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It describes the target effect as a stated fraction of the observed effect. Such calculations are illustrations of sample sizes under assumptions; they are not observed power and are not evidence that a trial can detect an effect. In particular, the registry record reports no primary outcome data from which to interpret an observed length-of-stay effect. A trial-specific sample-size justification or a documented design target is not reported in the registry record.
More generally, sample-size calculations depend on the outcome measure, variability or event rate, target effect, error-rate convention, allocation, and analysis method. If those assumptions do not match the eventual study or population, the calculated sample size may not deliver the anticipated precision. The registry record does not report the values underlying a trial-specific calculation or a completed primary analysis. Its planning assumptions therefore should not be read as evidence of study adequacy or as an estimate of treatment performance.
Can the registered secondary outcomes be analyzed?
The record lists hepatic encephalopathy severity repeatedly and also lists ascites. Repeated assessments can require methods that account for correlation among observations from the same participant; treating such observations as independent may understate uncertainty. This is a general point, not a description of an analysis used in this trial. The registry does not specify assessment timepoints, scoring rules, repeated-measures methods, or results for these outcomes. The listed outcomes alone do not establish that data were collected or analyzed.
Finally, statistical interpretation depends on a defined analysis population and rules for missing observations. The record does not report an intention-to-treat population, a per-protocol population, or methods for missing data. The primary outcome description instead says data were not collected. With no reported primary observations and no documented analysis, the treatment comparison remains unanswered, regardless of the planned design labels.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 14 outcome measures and 0 analyses. Secondary outcomes include repeated entries for hepatic encephalopathy severity and an entry for ascites. The record does not explain whether the repeated severity entries correspond to different assessment times, scales, or other distinctions. The number of planned comparisons and their hierarchy are not reported in the registry record.
Multiplicity arises when multiple outcomes, timepoints, or statistical comparisons are examined. In general, conducting many tests can increase the chance of finding an apparently unusual result even when no underlying effect exists. Study plans may address this by defining a primary comparison, specifying a hierarchy, or adjusting inference across comparisons. The registry does identify length of hospital stay as the primary outcome, but a multiplicity strategy for secondary outcomes is not reported in the registry record. No outcome analyses are reported, so there are no secondary findings to interpret or adjust.
The registry record does not report a missing-data strategy, imputation procedure, or sensitivity analysis. More fundamentally, it states that primary outcome data were not collected because of premature termination. This is different from having a completed outcome dataset with some individual observations missing: there is no reported primary dataset from which to estimate the treatment comparison.
Analysis populations are also not reported in the registry record. The record does not state which participants, if any, were included in an efficacy analysis or whether the intended population was based on assignment, treatment received, or protocol adherence. The enrollment entry alone does not define an analysis population. These omissions, together with the absence of collected primary outcome data, prevent assessment of the planned estimand, missingness assumptions, and robustness of results.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo | 0 | 1 | 0.0% |
| Tolvaptan | 0 | 1 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo | 0 of 1 |
| Tolvaptan | 0 of 1 |
The registry records a serious-adverse-event at-risk count of 1 and 0 affected for placebo, and the same figures for tolvaptan. For other adverse events, it likewise records 1 at risk and 0 affected in each arm. These are the adverse-event counts in the registry record; they should not be read as a complete account of all possible harms.
The recorded counts describe what the registry reports for the at-risk entries, not a comparative safety estimate. With only 1 listed at risk in each arm for each category, the record offers little information about how frequently events might occur in a broader population. A count of 0 affected in these entries does not demonstrate that an event cannot occur or establish that the treatments have equivalent safety.
The deaths entries are null, so death counts are not reported in the registry record. The serious-adverse-event terms list is empty; no specific serious-event terms are listed. The record does not provide event timing, severity details beyond the serious/other categories, exposure duration, or a narrative explaining how the adverse-event data were collected. These limitations are especially important because the trial terminated early and the primary outcome data were not collected. The adverse-event entries should be interpreted as sparse registry counts, not as a comprehensive safety evaluation.
12. What the Result Does, and Does Not, Mean
The central result is the absence of a reported primary outcome result. The trial set out to compare tolvaptan with placebo on length of hospital stay in hospitalized patients with cirrhosis and hyponatremia, but the registry states that primary data were not collected because the study ended prematurely. There is no estimate of how much longer or shorter hospital stay was in either group and no uncertainty interval or statistical test for that comparison.
This does not show that tolvaptan shortened hospital stay, nor does it show that it failed to do so. It also does not establish that the two groups had the same outcomes. A study without a reported primary outcome estimate cannot distinguish among those possibilities. The stated expected average of 2 weeks describes the planned follow-up timeframe, not an observed result.
The study’s termination and registered enrollment of 2 help explain why the record does not answer its question, but they do not substitute for outcome data. The recorded adverse-event counts are also not efficacy evidence. Their limited scope and the absence of reported death counts mean they cannot support broad conclusions about comparative safety.
For readers, the most appropriate interpretation is that this registry entry documents a planned randomized comparison that did not produce a reported primary analysis. The record does not report a trial-specific statistical plan, analysis population, missing-data method, or results for its listed secondary outcomes. Conclusions about treatment effects, precision, or consistency across outcomes cannot be drawn from the information recorded. Any planning illustrations should be understood as calculations under assumptions, rather than findings from participants or observed power.
13. Why This Trial Matters Statistically
This trial matters statistically because it illustrates the difference between a design and an answer. Randomization and masking can support a credible treatment comparison when the intended outcomes are collected and analyzed. Here, the registry records those design features, but also states that the primary endpoint data were not collected. The design labels alone cannot establish an effect or its precision.
Early termination can leave a clinically relevant question unresolved. In this record, the stated reasons include the performance period expiring, inadequate enrollment, and the funding sponsor not continuing support. These details help explain the lack of a completed primary analysis. They do not indicate what the treatment effect would have been had the study continued.
The record also demonstrates why outcome data, analysis methods, and transparent reporting all matter. A useful comparison requires more than an endpoint title: readers need the observed outcomes, a defined analysis, an estimate, and an account of uncertainty. When multiple secondary outcomes are listed, timepoints and a strategy for interpreting multiple comparisons are also important. Those elements are not reported for this trial.
Finally, sparse adverse-event entries should not be mistaken for a comprehensive evaluation. The registry’s recorded counts are informative as registry entries, but they do not offer a precise comparative safety assessment. In statistical education, this study is a clear example of why missing results must be described as missing rather than converted into evidence of no effect.
14. 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 main limitation is that the registry states that primary outcome data were not collected because of premature termination. Consequently, no length-of-stay estimate or comparison between tolvaptan and placebo is reported. The registry attributes termination to the performance period expiring, insufficient enrollment, and the funding sponsor not continuing support. The recorded enrollment is 2, but the record does not establish that those participants contributed primary outcome observations.
The registry does not report participant characteristics, eligibility details, treatment dose, treatment duration, adherence, or actual exposure. It describes an intended treatment start within 2 days of admission in the study summary, but does not report whether that occurred. Randomization and double masking are listed, while operational details of allocation, concealment, and masking are not reported in the registry record.
The primary endpoint description gives an expected average hospital stay of 2 weeks, but no observed duration, endpoint definition beyond length of stay, or analysis method. There is no reported statistical plan, sample-size justification, analysis population, or missing-data procedure. These omissions limit assessment of how the study would have estimated its primary effect even if outcome data had been available.
The registry lists 14 outcome measures and 0 analyses. Repeated hepatic encephalopathy severity entries are not explained by reported assessment times or methods, and no secondary outcome results are reported. Adverse-event entries show 0 affected in the listed serious and other categories, with 1 at risk in each arm for each category; death counts are not reported in the registry record. These sparse entries do not resolve the lack of an efficacy analysis or provide a broad safety assessment.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Tolvaptan to Reduce Length of Stay in Hospitalized Patients With Cirrhosis and Hyponatremia — NCT01890694.
- Registry record last updated 2017-12-11; status last verified 2017-11.
- Results first posted 2016-04-11.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01890694 is a Phase 4, randomized, double-blind, parallel-group trial sponsored by NYU Langone Health. It planned to compare once-daily tolvaptan with placebo in hospitalized patients with cirrhosis and hyponatremia, with length of hospital stay as the primary outcome. The listed start is 2012-03, primary completion is 2015-03, and status is terminated.
The registry records enrollment of 2 and states that the trial ended prematurely because the performance period expired, enrollment was inadequate, and the funding sponsor did not continue support. The primary outcome description says data were not collected. The record lists 14 outcome measures and 0 analyses; it reports no primary treatment-effect estimate or secondary outcome analysis. Serious and other adverse-event entries show 0 affected for each arm, with 1 at risk in each category in each arm. Death counts are not reported in the registry record.