Feasibility of Bright Light Therapy on Fatigue, Sleep and Circadian Activity Rhythms in Lung Cancer Survivors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Feasibility of Bright Light Therapy on Fatigue, Sleep and Circadian Activity Rhythms in Lung Cancer Survivors (NCT02954809), a randomized na study. The primary outcome is Study Completion.

2. Clinical Question

The study's stated purpose was to determine the feasibility of a morning bright light therapy intervention for fatigue, sleep disturbances, and circadian activity rhythms in lung cancer survivors. That purpose is different from establishing whether the intervention changes fatigue, sleep, or circadian outcomes. Although the record lists 10 outcome measures, the primary outcome reported in these results is study completion. No secondary outcomes are posted in the registry record.

The practical question addressed by the posted result is therefore whether participants completed the study under the two intervention routines, not whether one light condition produced a clinical benefit. The completion counts provide a basic account of participant retention during the study. They do not explain why participants did or did not complete, and they do not by themselves establish how acceptable or feasible the routines were to participants. Reasons for noncompletion and participant experience are not reported in the registry record.

3. Trial Design

This was a randomized, parallel-group study with single masking. The registry record lists the study phase as NA and reports an enrollment of 14. The study status is completed. Its recorded start was 2016-10, and its primary completion was 2018-01. The sponsor was State University of New York at Buffalo.

Participants were assigned to morning bright light therapy or an attention-control dim light condition. The experimental group used Green-Blue Re-Timer glasses for 30 minutes in the morning during one week. The attention-control group used Red-Yellow Re-Timer glasses for the same stated duration and morning schedule during one week. Both arms therefore involved a light device and a morning routine, while the described light conditions differed.

The registry record says the study was randomized, but details of sequence generation, allocation concealment, and the randomization procedure are not reported in the registry record. Single masking is reported, but the record does not identify which participants, staff, assessors, or analysts were masked. The parallel model indicates that the study compared groups assigned to separate interventions; the record does not describe a crossover design.

The posted arm sizes are 7 participants in the experimental group and 7 in the attention-control group. For the primary endpoint, the recorded result is 7 completions in the experimental arm and 5 completions in the attention-control arm. The record describes the primary endpoint timeframe as baseline, while also defining the measure as completion of the study. Additional detail about when completion was assessed or how the endpoint was operationalized is not reported in the registry record.

This design can describe completion under the two study routines in this enrolled group. It is a feasibility study, and the registry record does not report that it was designed to estimate a treatment effect on fatigue, sleep, or circadian activity rhythms. Protocol details, including eligibility requirements, recruitment methods, and the planned statistical analysis, are not reported in the registry record.

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

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 Study Completion, measured over Baseline.

The primary endpoint is Study Completion. Its unit is participants, and the registry description defines it as the percentage of participants recruited who completed the study. The posted result, however, is expressed as a count in each arm: 7 participants in the experimental group and 5 in the attention-control group. A completion percentage is not posted in the registry record.

Completion is relevant to a feasibility study because it describes whether enrolled participants remain through the study process. Retention can inform whether an intervention routine and the associated study procedures can be carried out in a participant group. In this trial, the endpoint gives a direct count of participants recorded as completing, but it does not reveal why other participants did not complete or which study procedures posed difficulty.

The endpoint is not a direct measure of fatigue, sleep quality, or circadian activity rhythms. Although these topics are part of the stated study purpose and the registry lists 10 outcome measures, the reported primary result concerns study completion. No secondary outcome results are posted in the registry record. The reported counts therefore support a limited description of completion in the two groups; they do not quantify change in the symptoms or rhythms named in the study purpose.

The record does not report the criteria used to classify a participant as having completed the study, a completion percentage, or an analysis comparing completion between groups. Without those details, the counts should be read as posted participant totals rather than as a complete statistical evaluation of feasibility.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted primary endpoint is binary in nature: a participant either completed the study or did not. The registry presents the outcome as participant counts, rather than as a continuous measurement such as a symptom score. For binary outcomes, analyses commonly describe the number of participants with the event and the number at risk in each group. The meaning of the event depends on the endpoint definition; in this study, the event is study completion.

What does a group comparison estimate?

A comparative analysis of a binary endpoint can estimate how often the event occurred in each arm and quantify the contrast between arms. A confidence interval expresses the uncertainty around an estimated quantity, such as a group difference or ratio. A narrow interval would indicate greater precision, while a wide interval would indicate that the data are compatible with a broader range of effects. No comparative estimate or confidence interval for study completion is reported in the registry record.

The record explicitly indicates that values are missing or not estimable for a two-arm comparison of this binary endpoint. That means the posted results do not support a reported inferential contrast to interpret. The arm counts remain descriptive information, but they are not a substitute for a documented estimate with its uncertainty. No p-value is reported for the primary completion result.

Why can small samples limit precision?

When an endpoint is based on participant counts and the groups are small, estimates can be sensitive to the outcome of individual participants. This is a general statistical feature of sparse binary data. It can make estimates uncertain and make standard approximations less reliable. The record does not report a method for handling sparse data or an adjusted analysis, so no particular method can be attributed to this study.

The study's enrollment was 14, with 7 participants listed in each arm. These are the total arm sizes reported for the study, while the completion results are counts of participants completing. An analysis of completion would ordinarily need to account for the number of participants in each group who were eligible for the endpoint and the number who met its definition. The registry record gives the arm totals and completion counts, but does not provide a fuller analysis population definition or endpoint-level handling rules.

What do planning assumptions mean?

General sample-size planning asks how many participants may be needed to detect a chosen effect under specified assumptions. The planning illustration associated with this study is described as based on the observed effect, equal allocation, a two-sided significance level of 0.05, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These assumptions describe an illustrative calculation, not the trial's own design rationale. The registry record does not report the trial's sample-size justification or establish that the study was designed around those assumptions.

Planning calculations are not observed power. Observed power is not a useful way to resolve uncertainty after results are known; the observed effect and its uncertainty are more directly relevant. In this record, a comparative completion effect is not estimable, so no conclusion about power to detect a completion difference can be drawn from the posted primary result. Planning sample sizes should be understood as hypothetical requirements under their stated assumptions, not as participants observed in the trial.

Why can over-dispersion matter?

Over-dispersion is a general concern for count outcomes when observed variability exceeds what a simple statistical model assumes. It matters because a model that understates variability can produce uncertainty estimates that are too narrow. The primary endpoint is a participant completion count with a binary event definition, rather than a repeatedly observed rate or a count of recurrent events. The registry record does not report a model for over-dispersion, and there is no basis for asserting that it was assessed in the trial's analysis.

What can be recalculated?

A recalculated result would be a simple, unadjusted calculation based on posted summary statistics; it would not replace an analysis conducted by the study sponsor. In this case, the registry record says a two-arm comparison is not estimable for the binary endpoint, and no recalculated comparative effect is available. The sponsor's analysis is not reported in the registry record. The record indicates no discrepancy between a sponsor result and a recalculation, but that does not create an estimate where none is posted.

In general, statistical results should be read alongside their endpoint definition, analysis population, missing-data approach, and uncertainty interval. Those details affect what an estimate represents. Here, the registry identifies the endpoint and provides arm counts, but does not report a comparative analysis, confidence interval, p-value, or detailed analysis plan. Statistical interpretation must therefore remain descriptive.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Experimental", "Attention Control")
n      <- c(7, 7)
events <- c(7, 5)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Experimental  7  7  100.0  59.0  100.0  64.6  100.0  63.1  99.7
ARM  Attention Control  7  5  71.4  29.0  96.3  35.9  91.8  34.9  91.5

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 10 outcome measures, but no secondary outcome results are posted. The primary result reported is study completion. The statistical plan for choosing outcomes, controlling multiplicity, or interpreting multiple comparisons is not reported in the registry record. Without that plan and the secondary results, it is not possible to determine how any family of outcome tests was defined or whether adjustments were intended.

Missing-data methods are also not reported in the registry record. The completion counts indicate that fewer participants completed in the attention-control arm than were listed in that arm, but the record does not state why participants did not complete, whether their outcome data were missing, or whether they contributed data to other measures. The endpoint itself is completion, so noncompletion is not necessarily missingness for that endpoint; it is a possible endpoint outcome. The record does not explain the classification procedures or report reasons for noncompletion.

The analysis population is not reported in the registry record. It is therefore unclear whether the trial used an intention-to-treat population, a per-protocol population, or another set of participants for any outcome analysis. No detailed account of exclusions, protocol deviations, or outcome-specific denominators is posted. The primary result is best read as the arm-level completion counts recorded by the registry, without assuming an unreported analysis population or missing-data strategy.

12. 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 0% reporting threshold.

The serious adverse event record lists 7 participants at risk in each arm and 0 serious adverse events in each arm. It also lists 0 other adverse events and 0 deaths in both the experimental and attention-control groups. No serious adverse event terms are listed.

These are recorded counts for the participants included in the registry's safety summary. A count of 0 means that no events of the specified category were recorded in that summary; it does not establish that an intervention carries no risk. The number of participants observed was limited, and the registry record does not provide event narratives, exposure-adjusted rates, or further detail about safety monitoring.

The figures also do not show whether an event was actively sought, how events were defined, or the period over which participants were monitored. Those safety methods are not reported in the registry record. The appropriate reading is limited to the recorded counts: no serious adverse events, other adverse events, or deaths are listed for either arm in the posted summary.

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

The result documents study completion in a randomized feasibility study comparing morning bright light therapy with a dim light attention control. The registry reports 7 completions in the experimental arm and 5 in the attention-control arm, with 7 participants listed in each arm. This is a descriptive account of completion in the enrolled study group.

The result does not establish that the light conditions differ in feasibility. No statistical comparison or uncertainty interval is reported, and the registry indicates that a two-arm comparison of the binary endpoint is not estimable. The completion counts cannot be treated as a precise estimate of what would happen in a larger or different group of lung cancer survivors.

The result also does not establish whether morning bright light therapy changes fatigue, sleep disturbance, or circadian activity rhythms. Those outcomes are part of the study's stated purpose, but the posted primary result measures completion. Although the registry lists 10 outcome measures, no secondary outcome results are posted. There is therefore no posted result on which to base an interpretation of changes in those outcomes.

Nor does completion alone provide a full assessment of feasibility. The record does not report adherence, participant acceptability, reasons for noncompletion, recruitment experience, or implementation barriers. A completed study status is not equivalent to complete retention of all enrolled participants. The available evidence supports a narrow conclusion: the registry records the completion counts and safety counts described, while broader feasibility and clinical outcome questions remain unresolved by the posted results.

14. Why This Trial Matters Statistically

Feasibility studies address whether an intervention and its research procedures can be carried out, a question that differs from estimating clinical effectiveness. This distinction matters when interpreting a trial whose stated purpose concerns fatigue, sleep, and circadian activity rhythms but whose posted primary result is study completion. Keeping the endpoint in view helps prevent a retention count from being mistaken for evidence about symptom change.

The record also illustrates the importance of separating descriptive findings from comparative inference. Counts can be useful for describing what happened in each arm, yet they do not automatically provide a precise estimate of the contrast between groups. Here, no comparative estimate or confidence interval is reported, and the record says the binary comparison is not estimable. A careful interpretation therefore preserves the observed counts while acknowledging the absence of a reported inferential result.

For biostatistical practice, this trial highlights the value of transparent endpoint definitions, analysis populations, missing-data procedures, and outcome reporting. Those details allow readers to understand what a result represents and how much uncertainty surrounds it. In this record, the endpoint description and counts are available, but analysis details and secondary results are not reported. The resulting boundary on interpretation is itself important: registry completion and a posted count should not be used to answer questions the reported data do not address.

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 small enrollment of 14 limits the amount of information available about completion and safety. The reported completion counts are 7 participants in the experimental group and 5 in the attention-control group. With few participants, these counts cannot characterize the range of experiences likely across a broader population, and no comparative uncertainty estimate is posted.

The registry record provides limited information about participant characteristics, eligibility criteria, recruitment, and study procedures. It does not identify which individuals were masked or explain the randomization process. These omissions make it difficult to assess how participants entered the study, how assignments were implemented, and how broadly the enrolled group may represent lung cancer survivors.

The primary endpoint description refers to a percentage of recruited participants who completed the study, but the posted results are counts. The record does not report a completion percentage, the rules for classifying completion, or reasons for noncompletion. It also does not report adherence to the light-device routine or participant views about the intervention. These gaps restrict the assessment of feasibility beyond the simple completion counts.

Although 10 outcome measures are listed, no secondary outcome results are posted. The statistical analysis plan, analysis population, and missing-data approach are not reported in the registry record. As a result, the available information cannot establish how fatigue, sleep, or circadian activity outcomes were evaluated, or whether any outcome analyses were adjusted for other factors.

The study is marked completed, but this status does not resolve the limitations of the posted results. The record offers a narrow view of completion and safety counts rather than a full account of intervention implementation or outcome effects. Conclusions should remain specific to the information reported for this study.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Feasibility of Bright Light Therapy on Fatigue, Sleep and Circadian Activity Rhythms in Lung Cancer Survivors — NCT02954809.
• Registry record last updated 2025-05-01; status last verified 2025-04.
• Results first posted 2025-05-01.
• 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

NCT02954809 is a completed randomized, single-masked, parallel-group feasibility study in lung cancer survivors. Enrollment was 14, with 7 participants listed in each arm. The experimental group received morning bright light therapy using Green-Blue Re-Timer glasses; the attention-control group received dim light using Red-Yellow Re-Timer glasses. Both routines were described as 30 minutes in the morning during one week.

The primary endpoint was study completion. The posted counts are 7 completions in the experimental group and 5 in the attention-control group. The registry record does not report a completion percentage or an estimable two-arm comparison. No secondary outcome results are posted. The safety summary lists 0 serious adverse events, 0 other adverse events, and 0 deaths in each arm.

The record's findings describe completion and recorded safety counts, not effects on fatigue, sleep, or circadian activity rhythms. Details of the statistical analysis plan, analysis population, and missing-data methods are not reported in the registry record.