Cognitive Behavioral Therapy and Multimodal Therapy in Treating Sleep Disturbance in Patients With Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Cognitive Behavioral Therapy and Multimodal Therapy in Treating Sleep Disturbance in Patients With Cancer (NCT01628029), a randomized phase 2 study. The primary outcome is Pittsburgh Sleep Quality Index (PSQI) Score Change.

2. Clinical Question

3. Trial Design

This is a randomized phase II trial with double masking and a factorial model. The registry record lists an enrollment of 68 and identifies M.D. Anderson Cancer Center as the sponsor. Its status is active, not recruiting. The listed start date is 2014-01-15, and the primary completion date is 2025-12-22.

The arm descriptions combine three treatment components: methylphenidate hydrochloride or placebo, therapeutic melatonin or placebo, and bright white light therapy or sham light intervention. Cognitive behavioral therapy is included in every arm. The control is titled Dim Red Light + Placebo + Placebo. The other listed arms vary the medication and light components, allowing combinations of active and placebo or sham interventions to appear in the randomized design.

The descriptions specify that cognitive behavioral therapy comprises 3 30-minute counseling sessions between baseline and day 14. Medication schedules are described as methylphenidate hydrochloride PO BID or placebo, and therapeutic melatonin PO QD or placebo. Bright white light therapy or sham light therapy is described as occurring over 30 minutes for 15 days. The registry also lists quality-of-life assessment among the interventions.

The factorial label describes the trial model, but the registry record does not report the statistical plan for estimating individual component effects or interactions. The record does not report allocation ratios, sequence generation, concealment procedures, or how masking was maintained. Although the record identifies double masking, it does not specify which participants or study personnel were masked. The reported primary endpoint is assessed at baseline and Day 15, and its analysis population is described as those who completed the study assessment and treatment.

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.

Only the first 6 of 8 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Those who completed the study assessment and treatment

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.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Pittsburgh Sleep Quality Index (PSQI) Score Change, measured over Baseline and Day 15.

The primary endpoint is change in Pittsburgh Sleep Quality Index score between baseline and Day 15. The PSQI is a self-report measure covering several aspects of sleep. Its 19 items contribute to seven component scores: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction.

Each component ranges from 0, indicating no difficulty or good status, to 3, indicating severe difficulty or poor status. The seven component scores are totaled to form a global score ranging from 0 to 21. A score of 0 indicates very good sleep quality, while 21 indicates very poor sleep quality. The registry notes that a global score greater than 5 often suggests significant sleep problems.

Using change in the global score gives a summary of whether reported sleep quality moved over the assessment period. Because lower scores indicate better sleep quality, a negative change is in the direction of a lower score. That direction alone does not establish how meaningful a change is to patients or whether the change differs between treatment combinations. The registry identifies the endpoint parameter as a mean and the dispersion as an inter-quartile range, and states that a Wilcoxon signed-rank test was used to determine the p value. The specific inferential result is not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information in the registry record is limited. It names a Wilcoxon signed-rank test as the method used to determine the p value for the PSQI endpoint. It also describes the endpoint as a mean with inter-quartile range. However, the specific test result, analysis details, and an estimable two-arm comparison are not reported in the registry record. The arm-level change summaries therefore require a descriptive interpretation rather than a reconstruction of an unreported inferential analysis.

What does the Wilcoxon signed-rank test address?

In general, the Wilcoxon signed-rank test is a nonparametric procedure for paired measurements. It uses the direction and ranking of within-person differences to assess whether those differences are centered around zero under the test's assumptions. It is often used when paired outcome differences may not be well represented by a normal distribution. In this trial, the registry names the test but does not specify precisely which paired observations entered it, whether a separate test was conducted for each arm, or the null hypothesis as implemented. Those details are not reported in the registry record.

A within-person test and a comparison between randomized groups answer different questions. A test of change within an arm asks whether the observed paired changes are consistent with a specified reference such as no change. A randomized comparison asks whether change differs between assigned groups. A result from one kind of test cannot be substituted for the other. Since the registry record does not post a p value or an estimable two-arm comparison for this endpoint, the arm summaries cannot establish whether any differences between combinations are statistically distinguishable.

How should the factorial structure be understood?

In general, a factorial design organizes treatment combinations so that more than one intervention component can be evaluated within a common randomized study. Estimating a component effect requires an analysis that compares relevant assignments and states how combinations and possible interactions are handled. The registry labels this trial factorial and lists combinations of methylphenidate, melatonin, and bright white or sham light intervention, with cognitive behavioral therapy common to all arms. It does not report a component-effect model, interaction analysis, or the contrasts used. It is therefore not appropriate to infer a factorial effect from the individual arm summaries alone.

Randomization can support comparisons of assigned groups when the analysis preserves that assignment and accounts for the design. The posted endpoint is restricted to participants who completed both study assessment and treatment. The registry record does not state how this restriction was handled analytically, whether all randomized participants were analyzed, or how missing outcomes were addressed. These details matter because a completer-only summary can differ from an analysis that includes all randomized participants under a prespecified missing-data approach.

What do the arm summaries and bounds convey?

The outcome entries report a change value for each arm, a group size, and lower and upper accompanying values. The endpoint description labels the parameter as a mean and the dispersion as an inter-quartile range, but the posted dispersion field is empty. The record does not identify the lower and upper values as confidence limits or provide a definition that would establish their interpretation. They should therefore be read as reported accompanying values, not as confidence intervals for either an arm-specific estimate or a between-arm effect.

In general, a confidence interval expresses uncertainty around an estimated quantity under a specified method. Its interpretation depends on what quantity was estimated and how the interval was calculated. An interval around one group's change is not automatically an interval for the difference between groups. Likewise, a range describing the distribution of observations is not a confidence interval. Because the registry does not label the posted bounds as confidence intervals, they do not establish precision for a treatment contrast.

Small arm-level group sizes also mean that a descriptive estimate may be sensitive to individual observations. That is a general statistical concern, not a claim about the unreported data distribution in this trial. The listed group sizes identify the participants contributing to each posted endpoint summary; they do not establish how many participants were randomized to each arm or explain the difference between the enrollment figure and endpoint group sizes.

What does the planning illustration mean?

The posted sample-size section is described as a planning illustration computed from the observed effect, not as the trial's own design justification. In general, such illustrations estimate the participants per arm needed under stated assumptions to detect a chosen fraction of an observed effect. They are planning quantities, not observed power, and they do not show the probability that the completed trial detected a true effect. The section states assumptions of a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It also states that the effect to detect is a stated fraction of the observed effect.

These assumptions do not convert the arm-level summaries into a valid randomized contrast. The record does not report the trial's own sample-size justification, its planned primary analysis, or interim analyses. Because no two-arm comparison is estimable for this endpoint in the registry record, the planning illustration should be understood only as a conditional calculation based on its stated assumptions, not as evidence that a treatment effect was established.

10. Multiplicity, Missing Data and Analysis Populations

The trial lists a primary PSQI change endpoint and three secondary outcomes: FACIT-F Subscale Score Change, Insomnia Severity Index Score Change, and FACT-G Score Change. The registry record does not report a multiplicity strategy, such as a hierarchy of hypotheses, an adjustment across outcomes, or a prespecified approach to interpreting multiple tests. It also does not report how multiple treatment components, arms, or contrasts were handled for inferential purposes. Without those details and the endpoint p value, the risk of interpreting a selected finding among several outcomes or comparisons cannot be assessed from the record.

The primary endpoint population is described as those who completed the study assessment and treatment. The registry record does not report the number of participants with missing primary endpoint data, reasons for missingness, imputation procedures, sensitivity analyses, or a missing-data model. It does not state whether the primary analysis was intention-to-treat, per-protocol, or another defined population. The endpoint summaries should therefore be interpreted as applying to the stated completer population, rather than assumed to represent all randomized participants.

The registry record lists enrollment of 68, while the outcome entries have arm-specific group sizes. It does not provide a participant flow account that explains the relation between enrollment, assignment, completion, and inclusion in the outcome summaries. No conclusion about attrition patterns or their effects can be drawn without that information. These omissions limit assessment of the analysis population and of how robust the descriptive results are to missing outcomes.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

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 adverse-event record lists serious adverse events by treatment group. Each of the listed groups has 0 serious adverse events among participants at risk. The registry also lists one death in Dim Red Light + Placebo + Placebo, with 0 deaths in each of the other listed groups. The record does not state whether that death was considered related to treatment, nor does it provide a cause. A death entry should not be treated as evidence of treatment attribution without that information.

Other adverse events are reported separately from serious adverse events. The listed affected counts are 4 in Dim Red Light + Placebo + Placebo, 0 in Bright White Light Therpay + Melatonin + Placebo, 1 in Dim Red Light + Melatonin + Methylphenidate, 2 in Bright White Light Therapy + Placebo + Methylphenidate, 1 in Bright White Light Therapy + Placebo + Placebo, 2 in Dim Red Light + Melatonin + Placebo, 3 in Dim Red Light + Placebo + Methylphenidate, and 0 in Bright White Light Therapy + Melatonin + Methylphenidate. The corresponding groups at risk are listed in the safety results.

These are descriptive group counts, not a comparison adjusted for exposure or other factors. The registry does not report the definitions used for adverse-event collection, event terms for serious events, timing, attribution, or statistical comparisons. The absence of listed serious adverse events in a group does not establish that the intervention is free of risk.

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

The registry results describe changes in PSQI scores among participants who completed the study assessment and treatment. The reported arm-level values are negative, and the PSQI scoring direction means a negative change corresponds to a lower score. The record therefore describes downward score changes in the listed combinations over the baseline-to-Day 15 assessment period. It does not, on its own, show that a particular combination caused the change or that the changes differ between combinations.

The primary endpoint summary is not accompanied by a reported p value, a documented inferential comparison, or bounds identified as confidence intervals. Consequently, the posted values do not quantify the precision of a between-arm treatment effect. The registry record indicates that values are missing or not estimable for a two-arm comparison. The summaries are useful as descriptions of the observed outcome entries, but they do not support a claim that one randomized regimen is better than another.

The treatment design also includes cognitive behavioral therapy in every listed arm. The results cannot answer whether cognitive behavioral therapy is more effective than no cognitive behavioral therapy, because that contrast is not represented in the listed arms. Nor can the arm summaries separately establish effects of methylphenidate, melatonin, or light therapy without reported factorial contrasts and an analysis plan.

The endpoint measures self-reported sleep quality, not every aspect of cancer-related symptoms or wellbeing. The registry lists fatigue, insomnia severity, and general quality of life as secondary outcomes, but comparative results for those outcomes are not reported in the record. The result does not establish durability after Day 15, effects on clinical outcomes outside the measured scores, or applicability to people who did not complete treatment and assessment.

In general, a clinical trial result should be interpreted in light of the study population, analysis population, uncertainty, and prespecified comparisons. Here, the record supports a limited descriptive interpretation. It does not provide enough inferential detail to determine whether the observed score changes reflect differences between assigned combinations, and it should not be read as a definitive assessment of efficacy or safety.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of how a factorial treatment question can be difficult to interpret when only arm-level outcome summaries are posted. The design combines medication and light components while keeping cognitive behavioral therapy common to the listed arms. To learn about individual components or interactions, readers need clearly specified contrasts and an analysis that corresponds to the factorial design. Those methods are not reported in the registry record.

The endpoint also illustrates the distinction between change within a group and a randomized comparison between groups. A negative PSQI change describes movement toward a lower score for an arm. It does not establish that the change differs from another arm, or that any one component accounts for it. The record identifies a Wilcoxon signed-rank test but does not post its p value or make the two-arm comparison estimable. This leaves the descriptive summaries without a reported inferential bridge to comparative conclusions.

Interpretation also depends on who contributed data. The primary outcome is described for participants who completed study assessment and treatment, while the record does not report missing-data methods or an intention-to-treat analysis. Such details are central to understanding whether the posted summaries represent the randomized groups as assigned or a subset shaped by completion.

Finally, the trial record distinguishes the primary sleep-quality outcome from secondary fatigue, insomnia, and quality-of-life measures, and separates serious from other adverse events. Keeping these questions distinct avoids treating one score or one safety count as a complete account of the intervention. In general, transparent reporting of analysis populations, contrasts, uncertainty, and multiplicity is essential for making a factorial trial's results interpretable.

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 primary outcome summaries are reported for participants who completed study assessment and treatment, and the registry record does not describe how missing observations were handled. It does not report participant flow, reasons for noncompletion, an intention-to-treat analysis, or sensitivity analyses. This limits understanding of how the outcome summaries relate to the randomized population.

The record identifies a Wilcoxon signed-rank test but does not report its p value, the exact hypothesis tested, or whether tests were performed separately within arms. No estimable two-arm comparison is available for the primary endpoint. The posted lower and upper values are not identified as confidence intervals, and the dispersion field is empty despite the endpoint description naming inter-quartile range. As a result, the precision of the arm-level estimates and any comparative effect cannot be assessed from the registry entry.

Although the trial is labeled factorial, the record does not describe the statistical model, component contrasts, interaction terms, or treatment of the common cognitive behavioral therapy component. It also does not report a multiplicity plan across the primary and secondary outcomes or among the treatment comparisons. These omissions constrain interpretation of component-specific effects and the false-positive risk associated with multiple testing.

The primary assessment is at baseline and Day 15. Longer-term outcomes are not described in the posted primary results, so persistence cannot be evaluated from them. The registry record also does not report outcome results for the listed secondary measures in this entry. Safety reporting provides group counts but not event attribution, timing, or serious-event terms. Finally, the record does not give the trial's sample-size justification or planned interim-analysis procedures.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Cognitive Behavioral Therapy and Multimodal Therapy in Treating Sleep Disturbance in Patients With Cancer — NCT01628029.
• Registry record last updated 2026-05-22; status last verified 2026-05.
• Results first posted 2026-02-04.
• 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.

18. Record Summary

The registry record describes a randomized, double-masked phase II factorial trial sponsored by M.D. Anderson Cancer Center, with enrollment listed as 68 and status active, not recruiting. Cognitive behavioral therapy is included in every described arm, alongside combinations of methylphenidate or placebo, melatonin or placebo, and bright white or sham light intervention.

The primary outcome is PSQI score change from baseline to Day 15 among those who completed study assessment and treatment. The record posts arm-level change summaries but no p value or estimable two-arm comparison. It identifies a Wilcoxon signed-rank test, but does not report the specific inferential result or analysis details. The record lists three secondary outcomes and adverse-event counts by group. Missing-data handling, multiplicity procedures, the detailed factorial analysis, and interpretation of the posted bounds are not reported in the registry record.