Gao Bipolar Spectrum Lithium/Quetiapine Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Gao Bipolar Spectrum Lithium/Quetiapine Study (NCT01526148), a randomized phase 4 study. The primary outcome is Time to Study Discontinuation.

2. Clinical Question

This is a comparative question about how long participants remained on study medication before discontinuation for any cause, not a question about discontinuation for a particular reason. The registry description says the trial enrolled people who currently required a medication change, so the comparison concerns those participants in that treatment context. It does not establish how the results would apply to people with different clinical circumstances.

The record lists 42 participants as enrolled, while its summary describes an enrollment goal of 60. It does not explain the difference or report the reasons for termination. The study is listed as terminated. These details matter when interpreting what the trial could establish: the posted results describe the enrolled groups, but they do not by themselves show that the planned comparison was completed or that the primary question was answered with a formal comparative analysis.

3. Trial Design

The registry identifies this as a Phase 4, randomized, parallel-group trial comparing lithium with quetiapine. The listed allocation is randomized. The summary describes a four-month comparative safety, tolerability, and effectiveness trial and says randomization was to be stratified to reduce imbalance across bipolar type I, bipolar type II, and sub-threshold categories. The record does not report the randomization sequence, allocation concealment, or the details of stratification.

There is an inconsistency in the description of masking. The registry lists the study as single-masked, while its brief summary calls it open-label. The registry record does not clarify which description applied in practice, who, if anyone, was masked, or whether masking was maintained. This makes it difficult to characterize the trial’s protection against knowledge of assigned treatment from the record alone.

The planned enrollment goal was 60 participants, and the summary describes an intended enrollment period of 24 months from initial regulatory approval. Actual enrollment is listed as 42. The study start is recorded as 2012-01, primary completion as 2015-01, and status as terminated. The reason for termination and whether all intended follow-up was completed are not reported in the registry record.

The primary outcome is titled Time to Study Discontinuation, is measured in days, and has a Week 16 timeframe. The registry posts a mean and 95% confidence interval for each arm. It does not post a reported primary analysis, a between-arm estimate, or an analysis population for that outcome. The secondary outcome concerns change in HOMA-IR from Screening to Week 16. The record lists one secondary outcome and no reported secondary analyses, although arm-level summary statistics and a recalculated comparison are available for that measure.

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:

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 Time to Study Discontinuation, measured over Week 16.

The primary endpoint is time to study discontinuation, measured in days, with discontinuation described as occurring for all causes. In a time-to-event framework, this kind of endpoint focuses on when an event occurs rather than only whether it occurred by a fixed point. It can therefore retain information about event timing. That is a general statistical feature of time-to-event outcomes, not a description of the analysis used in this trial.

The registry states that the time to all-cause discontinuation would be used as the primary outcome and gives Week 16 as the timeframe. All-cause discontinuation combines reasons for stopping rather than isolating one particular reason. This may be useful for a broad comparison of treatment persistence, but it does not indicate why a participant discontinued. Reason-specific results are not reported in the registry record.

The posted primary summaries are means with 95% confidence intervals for lithium and quetiapine. These are arm-level descriptions; they are not themselves an estimate of the difference between arms. The registry record does not report a comparative estimate, a test result, or the method used to handle participants who had not discontinued by the end of follow-up. It also does not report how censoring or missing follow-up was handled. Those details are important for interpreting a time-to-event endpoint, because participants with incomplete observation may not have a known event time.

Consequently, the available summaries do not establish how the groups compared on the primary question. They describe the posted arm-level outcome summaries, while the inferential comparison and key analysis details remain unreported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation depends on matching each reported estimate to the question it answers. The primary endpoint is time to all-cause discontinuation, but the posted primary results are means and confidence intervals by arm. An arm-specific mean describes the central value for that group; it does not directly quantify the contrast between the randomized groups. A treatment comparison generally requires an estimate of the between-arm contrast, its uncertainty, and a description of the analysis used. Those primary comparative results are not reported in the registry record.

What does a confidence interval communicate?

In general, a confidence interval communicates the uncertainty associated with an estimate under the statistical method’s assumptions. A narrower interval indicates greater precision than a wider interval, but precision is not the same as clinical importance. An interval does not establish that all values it contains are equally plausible, and it is not a probability statement about the true value after the data have been observed. For the primary outcome, the intervals are attached to each arm’s mean. They do not provide the confidence interval for a difference between arms.

The secondary outcome has a recalculated comparison. Its mean difference is defined as the quetiapine mean relative to the lithium mean, and the reported estimate is 5.7. The 95% confidence interval, from -1.673 to 13.07, includes zero. In general, zero represents no difference on a difference scale. The interval therefore does not rule out no mean difference under the calculation used, and it also permits effects in either direction. The unadjusted p-value is 0.13. A p-value is a measure of how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions held. It is not the probability that the null hypothesis is true, nor a measure of the size or importance of an effect.

Why does the analysis source matter?

The secondary comparison is identified as recalculated from posted summary statistics, and no reported secondary analysis is listed. Such a recalculation is unadjusted: it compares the available arm-level summaries without incorporating covariates or other model features that might have appeared in a sponsor analysis. The registry record does not report a sponsor analysis for this measure. The recalculated result should therefore be understood as a simple comparison of posted summaries, rather than as a substitute for an unreported prespecified analysis.

The registry does not report the method used for the primary time-to-discontinuation outcome. In general, time-to-event analyses can account for participants whose event time is not observed during follow-up, using censoring methods under stated assumptions. The posted primary means do not reveal whether or how such methods were used. The registry also does not report an analysis population, missing-data rules, or handling of participants who stopped attending follow-up. These omissions prevent readers from assessing whether the primary summary reflects all randomized participants or another subset.

How should the planning illustrations be read?

The sample-size illustrations are based on the observed secondary effect, not on the trial’s own design assumptions. The registry record describes the illustration as using a two-sided significance level of 0.05, equal allocation, and a normal approximation. It lists participants per arm for 80% power and participants per arm for 90% power under planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For example, the listed planning values are 53 and 71 participants per arm for 80% and 90% power when planning for 100% of the observed effect; for 75%, they are 94 and 126; for 50%, they are 211 and 282.

These figures are planning illustrations, not observed power and not evidence that the trial was designed around the observed effect. They show how required group size would change under the stated planning assumptions and effect scenarios. The record does not report the trial’s own sample-size justification, target effect, or design calculation. The stated enrollment goal of 60 does not, by itself, reveal what assumptions informed that goal.

General statistical interpretation also requires attention to multiplicity. If several outcomes or analyses are examined, the chance of obtaining at least one apparently unusual result can increase unless the analysis plan accounts for multiple testing. The registry lists one secondary outcome, but it does not report a multiplicity procedure or a complete analysis plan. The unadjusted secondary p-value should therefore be interpreted with the limited analysis information available, rather than as an isolated confirmatory finding.

10. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical analysis plan describing which analyses were prespecified, how outcomes were prioritized, or how multiple comparisons would be handled. It lists one secondary outcome and does not list reported analyses for it. Without a documented plan, readers cannot determine whether the posted secondary comparison represents the intended analysis or whether other analyses were conducted but not reported.

Multiplicity is a general concern when investigators examine more than one outcome, subgroup, time point, or model. Each additional analysis can create another opportunity for an apparently notable result to arise by chance. Methods such as adjustment or a prespecified hierarchy can address this concern, but no multiplicity procedure is reported in the registry record. The secondary p-value is explicitly unadjusted and should be interpreted accordingly.

Missing-data methods and analysis populations are also not reported. The record does not state whether analyses included all randomized participants, only participants with observed outcomes, or another population. For the time-to-discontinuation endpoint, it does not describe censoring, handling of incomplete follow-up, or how participants without a recorded event by the end of observation were treated. For HOMA-IR, the record does not describe missing measurements, exclusions, or imputation.

These omissions matter because analysis choices can affect both an estimate and its uncertainty. They do not prove that the results are biased, but they limit assessment of robustness and reproducibility. The registry record also does not report interim analyses, stopping rules, or the reason for termination. No inference about those matters should be made from the status alone.

11. 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 registry reports serious adverse events for 0 of 18 participants at risk in the lithium arm and 0 of 24 participants at risk in the quetiapine arm. These are the posted counts of participants affected, not estimates of comparative risk with uncertainty. The registry also lists no serious adverse event terms. The record does not provide additional detail about event definitions, monitoring, or the period covered by these counts.

Other adverse events are reported for 16 of 18 participants at risk in the lithium arm and 21 of 24 participants at risk in the quetiapine arm. These are raw participant counts. They do not indicate event severity, duration, number of events per participant, or whether events were considered related to treatment. The registry does not provide those details in the posted figures.

Deaths are not reported in the registry record. The absence of a reported death count should not be interpreted as evidence that no deaths occurred. Likewise, zero reported serious adverse events in the posted group counts does not establish that serious events were impossible or that the treatments have equivalent safety. The study enrolled 42 participants overall, and the registry status is terminated; the record does not explain whether safety follow-up was complete.

These figures support a limited description of what was posted. They do not provide a formal safety comparison or enough detail to establish differences in safety between the arms.

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

The registry results describe a randomized study of lithium versus quetiapine in participants with bipolar disorder who were described as needing a medication change. For the primary outcome, the record posts arm-level mean times to study discontinuation and confidence intervals. It does not report a between-arm estimate or formal primary analysis. The posted summaries therefore cannot, on their own, answer whether time to discontinuation differed between the assigned treatments.

The secondary HOMA-IR outcome has a recalculated mean difference and uncertainty interval, but the interval includes zero and the unadjusted p-value is 0.13. This is not evidence that the treatments are identical, and it does not establish that a difference exists. The result is uncertain and comes with limited information about the analysis plan, missing data, and multiplicity.

In general, a randomized design can support comparisons between assigned groups, but interpretation also depends on how the trial was conducted and analyzed. Here, the record contains an unresolved difference between its single-masked design listing and its open-label summary description. It does not report allocation concealment, analysis populations, or handling of incomplete outcome data. These gaps make it harder to assess potential sources of bias.

The record also lists an enrollment of 42 against a stated goal of 60 and identifies the study as terminated. It does not report why it was terminated or whether follow-up and planned analyses were completed. The posted data should therefore be treated as a limited account of the trial’s results, rather than as a complete resolution of the comparative question.

Nothing in these results alone establishes that either treatment is preferable, safer, or more effective for an individual or for all people with bipolar disorder. They describe what the registry reports, with substantial limitations in the primary comparative analysis and the methodological detail available for interpretation.

13. Why This Trial Matters Statistically

This trial illustrates an important distinction in biostatistics: posting summary values for each randomized group is not the same as reporting the comparison that addresses the primary research question. The primary question concerns time to all-cause discontinuation between lithium and quetiapine, yet the registry reports arm-level means and confidence intervals without a comparative estimate. Readers need the contrast and its uncertainty to judge how the groups compare statistically.

The record also demonstrates why endpoint type and analysis method should be considered together. Time to discontinuation is framed as an event-timing outcome, while the posted summaries are means. The registry does not explain how incomplete follow-up or participants without a recorded discontinuation were handled. That missing information leaves a key part of the statistical interpretation unresolved.

The secondary outcome further shows why results should be distinguished by source and status. The posted comparison is a recalculation from summary statistics, not a reported analysis. Its uncertainty and unadjusted nature matter, as does the absence of a documented multiplicity approach. Planning illustrations based on an observed effect can be useful for considering future study sizes, but they do not describe the trial’s own design justification or its observed power.

Finally, the discrepancy between the masking descriptions and the difference between the stated enrollment goal and actual enrollment highlight the importance of transparent registry reporting. These are not reasons to assume a particular bias or outcome. They identify questions the registry does not answer. For researchers, the example underscores the value of reporting prespecified analyses, populations, missing-data procedures, and between-group estimates alongside descriptive summaries.

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.

Several limitations are specific to this trial record. First, the primary outcome lacks a reported between-arm analysis. The means and confidence intervals describe the arms separately, but the registry does not report the estimated contrast, its uncertainty, or a test result. That limits direct interpretation of the trial’s stated primary question.

Second, the design description is inconsistent: the registry lists single masking, while the brief summary calls the study open-label. The record does not resolve this discrepancy or say who was masked. It also does not report allocation concealment or detailed randomization procedures. These missing details prevent a complete assessment of safeguards against bias.

Third, the study is listed as terminated, with 42 participants enrolled against a stated goal of 60. The reason for termination, completeness of follow-up, and relationship between enrollment and the planned analysis are not reported in the registry record. The actual analysis population is also not identified.

Fourth, the primary endpoint’s handling of incomplete follow-up and censoring is not described. For the secondary endpoint, the registry does not report missing-data procedures, covariate adjustment, or a sponsor analysis. Its available comparison is unadjusted and recalculated from posted summary statistics. The record also does not report a multiplicity procedure, interim analyses, or the trial’s own sample-size justification.

Finally, safety counts are limited to the posted participant counts and do not describe event severity, relationship to treatment, timing, or complete follow-up. Deaths are not reported in the registry record. These limitations constrain what can be concluded from the record, but do not establish the direction or magnitude of any bias.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Gao Bipolar Spectrum Lithium/Quetiapine Study — NCT01526148.
• Registry record last updated 2017-08-16; status last verified 2017-07.
• Results first posted 2015-10-30.
• 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.

18. Record Summary

The registry describes a Phase 4 randomized parallel-group comparison of lithium and quetiapine in participants with bipolar disorder who required a medication change. The study is listed as single-masked, while its summary calls it open-label. Enrollment is 42, the stated goal was 60, and the status is terminated. The record does not report the reason for termination.

The primary outcome is time to study discontinuation in days at Week 16. Arm-level means and 95% confidence intervals are posted, but a primary between-arm analysis is not reported. One secondary outcome concerns change in HOMA-IR from Screening to Week 16. Its comparison is a recalculation from posted summaries, and no reported secondary analysis is listed.

The registry does not report the analysis population, missing-data methods, multiplicity procedure, interim analyses, or the trial’s own sample-size justification. The posted results permit a limited description of the outcomes but do not provide a complete inferential account of the primary question.