Adjunctive Brief Behavioral Treatment of Insomnia (BBTI) for Sleep Intervention (SI): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Adjunctive Brief Behavioral Treatment of Insomnia (BBTI) for Sleep Intervention (SI) (NCT02248675), a randomized na study. The primary outcome is Columbia Suicide Severity Rating Scale (C-SSRS).

2. Clinical Question

The study rationale connects sleep disturbance with suicidal thoughts and behaviors, and notes its co-occurrence with depression and posttraumatic stress disorder. The proposed strategy was to improve sleep as an adjunct to usual care, with suicidal ideation as the primary outcome. The trial also recorded perceived treatment beliefs, depression symptoms, and insomnia severity. These questions concern measured outcomes over the study’s post-treatment period; they do not by themselves establish effects on suicide attempts, deaths, longer-term outcomes, or use of other treatments. Those outcomes are not reported as results in the registry record.

3. Trial Design

The registry describes a randomized, double-masked, parallel-group trial with an enrollment of 54. The study is marked completed. Its listed start was 2015-02, and primary completion was 2016-01. The phase is listed as NA. The sponsor was the VA Office of Research and Development. The registry does not report the method used to generate or conceal random assignments, the details of masking, or the original sample-size justification.

One group received CBT-I plus TAU; the comparator group received TAU. CBT-I was described as four individual sessions added to usual care. For TAU, participants were encouraged to begin or continue treatment for co-occurring conditions as recommended by treatment providers. The description notes that care could involve primary care teams, behavioral telehealth, or specialty outpatient mental health; it does not report which services were actually received by each participant.

The primary endpoint was the C-SSRS Suicidal Ideation Intensity scale at post-treatment, approximately 6-8 weeks. The registry reports a mean and standard deviation for each group. The posted primary outcome summaries include 23 participants in CBT-I plus TAU and 27 in TAU Control. The registry also lists three secondary outcomes: Perceived Treatment Beliefs (PTS), Patient Health Questionnaire-9 (PHQ-9), and Insomnia Severity Index (ISI). Their summaries are also for the post-treatment timeframe.

The record reports linear regression as the analysis method for the primary and secondary comparisons, but does not provide regression estimates, confidence intervals, or detailed model specifications. The posted group summaries permit simple, unadjusted comparisons; those are distinct from the sponsor-reported regression analyses.

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 Columbia Suicide Severity Rating Scale (C-SSRS), measured over post-treatment (~6-8 weeks).

The primary endpoint is the Suicidal Ideation Intensity scale from the Columbia Suicide Severity Rating Scale (C-SSRS), assessed post-treatment, approximately 6-8 weeks. Although the entire C-SSRS was to be administered, the investigators identified this intensity scale as the primary outcome. It is formed by summing five items, has a stated score range of 0-25, and higher scores indicate more severe suicidal ideation.

A continuous score preserves gradations in reported ideation intensity rather than reducing each participant to a yes-or-no category. Comparing group means therefore describes the average score in each arm at the specified assessment. This outcome is relevant to the study’s stated question because the trial aimed to examine whether adding a sleep intervention to usual care might reduce suicidal ideation in Veterans with other conditions.

Use of this measure does not make it a measure of every aspect of suicide risk or behavior. The posted endpoint concerns the specified intensity scale at post-treatment. The registry record does not report results for suicide attempts, deaths, or longer-term suicidal ideation. Nor does the score alone establish why group averages differ. Interpretation depends on the estimated contrast, its uncertainty, the analysis method, and the limitations of the posted trial record.

7. Results Posted in the Registry

• CBT-I + TAU: mean 5.40 versus 8.40 on control; difference -3.00 (95% CI -6.66 to 0.66), standardized effect d = -0.46; z-test p = 0.108.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Perceived Treatment Beliefs (PTS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • CBT-I + TAU: mean 135.70 versus 133.04 on control; difference 2.66 (95% CI -10.46 to 15.78), standardized effect d = 0.11; z-test p = 0.691.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CBT-I + TAU and the control. The interval gives the range of effects that remain compatible with the data.

Patient Health Questionnaire-9 (PHQ-9)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • CBT-I + TAU: mean 6.40 versus 12.30 on control; difference -5.90 (95% CI -8.67 to -3.13), standardized effect d = -1.17; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for CBT-I + TAU than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Insomnia Severity Index (ISI)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • CBT-I + TAU: mean 7.90 versus 17.80 on control; difference -9.90 (95% CI -12.88 to -6.92), standardized effect d = -1.85; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for CBT-I + TAU than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes offer a mixed but coherent descriptive pattern across domains. Post-treatment average scores for depression symptoms and insomnia severity are lower in the CBT-I plus TAU group than in TAU Control, whereas average treatment-belief scores are similar. These outcomes address related but distinct constructs; their pattern does not directly answer the primary question about suicidal ideation intensity.

The unadjusted intervals for the depression and insomnia measures exclude zero, while the interval for treatment beliefs crosses zero. The regression analyses reported by the registry likewise give different p-value patterns across the secondary measures. These are results for separate endpoints, not multiple measurements of the same outcome, and they should not be combined into one overall claim.

Secondary outcomes can help describe whether the observed group pattern extends to measures related to the intervention’s rationale. They do not confirm the primary result, particularly when the primary estimate remains imprecise. Nor do post-treatment symptom scores establish changes in treatment use, longer-term status, or suicide-related behavior; those outcomes are not reported in the registry record.

The registry record does not report a multiplicity adjustment for the secondary tests or identify a hierarchy governing their interpretation. When multiple outcomes are tested, the chance of at least one apparently notable result can increase even if there is no true effect on any outcome. That general concern means unadjusted secondary p-values and intervals carry less evidential weight than a prespecified, multiplicity-aware analysis would. They are best treated as exploratory context and as questions for future research, not as independent confirmation that the intervention changes all related outcomes.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Perceived Treatment Beliefs (PTS)
• Patient Health Questionnaire-9 (PHQ-9)
• Insomnia Severity Index (ISI)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference describe?

For a continuous outcome, a mean is the arithmetic average of the recorded scores in a group. A mean difference compares those group averages. In the unadjusted recalculation for the primary endpoint, the contrast is expressed as CBT-I plus TAU relative to TAU Control, so the negative estimate indicates a lower average score in the intervention arm. Because higher C-SSRS intensity scores indicate more severe suicidal ideation, the sign describes the observed direction of the score contrast; it does not by itself establish causation, clinical importance, or certainty.

The arm-level standard deviations describe how dispersed individual scores were around each group mean. They are not confidence intervals for the mean and do not show the range of individual participant outcomes. The reported standard deviations are substantial relative to the group means, indicating that individual scores varied within both groups. A difference between averages can coexist with considerable overlap among individual scores.

What is a confidence interval?

A confidence interval expresses uncertainty around an estimated contrast under the statistical method used. In general, it is a range of values compatible with the data and that method; it is not a guarantee that the true effect lies within the particular interval, nor does it describe the range of individual participant scores. For the primary recalculation, the 95% interval crosses zero. Zero represents no mean difference on the scale, so the interval does not exclude no difference. The data are also compatible with contrasts in either direction within the interval.

Confidence intervals depend on the estimate, variability, sample information, and calculation method. They do not account automatically for every possible source of bias, such as missing outcomes, departures from the intended analysis, or problems with measurement. The registry record does not provide a confidence interval for the sponsor-reported regression result, so uncertainty around that model estimate cannot be assessed from the posted regression result alone.

How should the p-values be read?

A p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the assumptions used for the test were correct. It is not the probability that the null hypothesis is true, and it is not a measure of the size or importance of a difference. The recalculated unadjusted primary p-value is 0.108; the registry’s reported linear regression p-value is .153. Both describe evidence under their respective analyses, not the probability that the intervention did or did not work.

The registry does not report enough regression detail to determine precisely which covariates or other model specifications were used. The unadjusted recalculation uses only posted summary statistics, so it should not be described as a reproduction of the sponsor’s regression model. When analyses differ, their estimates and p-values need not be identical.

Why can small trials yield uncertain estimates?

In general, estimates from limited samples tend to be less precise, particularly when individual outcomes vary. Here, the primary group standard deviations indicate appreciable variability, and the posted group summaries include 23 and 27 participants. The confidence interval is correspondingly broad enough to include no difference and contrasts in either direction. This is a statement about precision, not a claim that a particular study is underpowered or that an observed effect is real.

Randomization is intended to make treatment groups comparable on average, but it does not ensure exact balance in every characteristic in a particular trial. The registry record does not report baseline characteristics or allocation details sufficient to assess balance, nor does it report the sample-size justification. The posted results therefore do not permit a full evaluation of how design choices affected precision or comparability.

Why can over-dispersion matter?

In general, over-dispersion means that outcomes vary more than a model assumes. When a statistical model understates variability, standard errors can be too small and confidence intervals too narrow, making results appear more precise than warranted. For continuous outcomes, a suitable model should represent the observed variation and any relevant structure in the data. The registry identifies linear regression, but does not report model diagnostics or details that would establish whether over-dispersion or other departures from assumptions were assessed. No conclusion about model adequacy can be drawn from the posted p-value alone.

The group standard deviations describe the observed spread but do not themselves diagnose a regression model. The recalculation uses a pooled standard deviation for a simple standardized contrast and standard errors for the arm means; these are summary-statistic calculations, not checks of the sponsor’s model assumptions.

What does standardization add?

A standardized mean difference expresses a mean contrast relative to a measure of outcome variability. It can help compare contrasts measured on different scales, but it is less directly interpretable than the original score difference. A standardized effect does not make outcomes clinically interchangeable, and it is sensitive to the variability in the study population. The planning illustrations use the observed effect as a basis for hypothetical future sample sizes, but that observed effect is uncertain and should not be treated as a reliable forecast.

For planning, smaller target effects require more participants to distinguish them from random variation under otherwise fixed assumptions. This general relationship is reflected in the reported planning scenarios. The sample-size assumptions are an illustration based on the observed effect, not the trial’s original design plan. The registry record does not report the trial’s own sample-size justification or the assumptions used when it was designed.

11. Confidence Intervals and Precision

For the primary outcome, the unadjusted recalculated mean difference is -3.0 points, with a 95% confidence interval from -6.656 to 0.656. The interval crosses zero, so the estimate is compatible with no mean difference as well as with contrasts on either side of zero. Its width reflects substantial uncertainty about the size and direction of the mean contrast. This uncertainty matters even though the point estimate is on the side of a lower average score in the CBT-I plus TAU arm.

The group standard deviations, 6.8 and 6.3, show that scores varied within both arms. They are not uncertainty limits around the treatment contrast. Confidence intervals around a mean difference address uncertainty in that contrast, whereas standard deviations describe variation among observed scores. Neither statistic describes every participant’s experience.

The secondary unadjusted intervals also differ in how clearly they separate from zero. The interval for Perceived Treatment Beliefs spans zero, while the posted intervals for PHQ-9 and ISI do not. Those intervals concern separate secondary outcomes and do not resolve uncertainty in the primary endpoint. Because several outcomes were examined and multiplicity handling is not reported, apparent precision on a secondary measure should be interpreted cautiously.

The sponsor-reported primary analysis has p = .153 but no regression estimate or confidence interval in the registry record. Its precision cannot be judged from that p-value alone. In general, a p-value does not reveal the range of plausible effect sizes, and a confidence interval does not remove concerns about model specification, missing data, or study design. The record’s incomplete reporting on analysis details limits how fully the uncertainty can be assessed.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are hypothetical planning illustrations based on the observed effect, not the trial’s own design assumptions. The listed assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is the stated fraction of the observed effect. The original trial’s sample-size justification is not reported in the registry record.

The 80% power column represents the number of participants per arm estimated to give an 80% chance of detecting the specified effect under those assumptions, if that effect and the model assumptions hold. The 90% power column represents a higher planned chance of detection, which generally requires more participants. These are planning probabilities under an assumed effect and statistical model, not probabilities that the trial result is correct.

For the primary outcome, planning for 100% of the observed effect gives 75 participants per arm for 80% power and 100 participants per arm for 90% power. Planning for 75% of the observed effect gives 133 participants per arm for 80% power and 178 participants per arm for 90% power. Planning for 50% of the observed effect gives 298 participants per arm for 80% power and 399 participants per arm for 90% power. The pattern illustrates that smaller effects require more participants to detect under the same assumptions.

These are not observed power calculations for the completed trial. They are forward-looking illustrations derived from an observed estimate, which is uncertain, and should not be read as evidence that the trial did or did not have adequate power. Different assumptions about the target effect, allocation, outcome variability, or analysis could yield different planning requirements.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -3.000; pooled standard deviation σ = 6.534.
2. n = 2 × (1.96 + 0.842)² × 6.534² / -3.000² = 74.5, rounded up to 75 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -3                      # observed difference in means
sigma <- 6.53391791602353                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(2248675)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 75 (80% power), 100 (90% power)
 75% of observed effect: n per arm = 133 (80% power), 178 (90% power)
 50% of observed effect: n per arm = 298 (80% power), 399 (90% power)
power.t.test: 75.4 
Simulated power at n = 75 per arm: 0.797 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry reports that the sponsor used linear regression for the primary comparison and gives p = .153. It does not report the regression coefficient, confidence interval, covariates, or fuller model specification. The result therefore identifies the stated method and p-value but does not provide a model-based estimate or its precision.

A simple unadjusted recalculation from the posted group means and standard deviations gives a mean difference of -3.0 and a p-value of 0.108, with a 95% confidence interval from -6.656 to 0.656. This recalculation is not the sponsor’s regression analysis: it uses the arm-level summary statistics and does not reproduce any unreported model adjustments or details. Differences between these p-values should not be treated as a contradiction without knowing the regression specification and data handling.

The recalculated estimate is useful as a transparent description of the unadjusted group contrast, but it does not replace the sponsor-reported analysis. The registry record does not mark a discrepancy between the posted results and the recalculation. Interpretation should retain the distinction between the sponsor’s reported regression p-value and the unadjusted summary-statistic estimate, rather than selecting whichever appears more favorable.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and three secondary outcomes. It reports linear regression analyses for the primary and secondary comparisons, but does not report whether the secondary tests were adjusted for multiplicity, whether an outcome hierarchy was specified, or whether any outcomes were designated exploratory in the original analysis plan. Without that information, the family of hypotheses being tested and the intended interpretation of the multiple p-values are unclear.

In general, testing multiple outcomes can increase the chance of finding at least one apparently strong result by chance. Adjustment methods can address some of that concern, but they may also change the thresholds or intervals used for interpretation. No multiplicity correction should be assumed from the registry’s listing of outcome measures or from the reported regression method. The posted secondary results are therefore most cautiously read as unadjusted unless the record states otherwise.

Missing-data handling is not reported in the registry record. The primary group summaries include 23 participants in CBT-I plus TAU and 27 in TAU Control, while enrollment is listed as 54. The registry does not explain the relationship between those figures, state how many randomized participants contributed to each analysis, or describe any imputation, exclusion, or sensitivity analysis. It also does not report whether missing outcomes differed by arm or reason.

The analysis population is likewise not reported in enough detail to determine whether results were based on all randomized participants, participants who completed treatment, or another defined group. Although randomization is recorded, the registry does not state whether the sponsor used an intention-to-treat approach or how deviations from assigned treatment were handled. These omissions limit assessment of attrition bias and the extent to which the analyzed groups preserve the randomized comparison. They also make it difficult to evaluate whether the regression analysis and unadjusted summaries use the same observations.

16. 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 0% reporting threshold.

The adverse-event figures report zero serious adverse events in the CBT-I plus TAU group and zero in TAU Control. The figures for other adverse events are also zero in both groups. The registry lists 23 participants at risk for each event category in CBT-I plus TAU and 27 in TAU Control. The death counts are not reported in the registry record.

These entries describe what was posted for the trial; they do not establish that no adverse events of any kind occurred outside the registry’s reporting categories, or that there is no safety concern. The record does not provide event definitions, follow-up details, or a narrative explaining the event reporting. Zero recorded events also do not show that two interventions have equivalent safety, especially when the study is a pilot and the number of participants contributing to the safety summaries is limited.

Safety figures should be interpreted alongside the outcome results and the completeness of reporting. The registry’s zero counts are not evidence about unreported events or about longer-term safety. The information supports only the narrow statement that the posted serious and other adverse-event counts were zero in each group, with deaths not reported.

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

The posted results describe a randomized comparison of CBT-I added to usual care with usual care alone among Veterans. For the primary outcome, the adjunctive-intervention group had a lower average post-treatment suicidal ideation intensity score. The unadjusted estimate is uncertain: its confidence interval crosses zero, and the registry’s sponsor-reported regression result is p = .153. The results therefore do not provide a precise, definitive estimate of the primary contrast.

A lower mean score in one group is not, on its own, proof that the intervention caused the difference. Randomization supports a comparative design, but interpretation also depends on the analysis population, missing outcomes, model specification, and outcome measurement. The registry does not report key details about those issues. The posted results also do not establish whether a difference of the estimated size is clinically important; a threshold for meaningful change is not reported in the registry record.

The trial’s stated focus is suicidal ideation intensity at post-treatment. That endpoint should not be generalized to suicide attempts, deaths, longer-term risk, or every dimension of mental health. The registry record does not report these outcomes as findings. Likewise, the secondary measures of depression, insomnia, and treatment beliefs describe separate outcomes and cannot substitute for the primary endpoint.

The secondary group summaries show lower average depression and insomnia scores in the adjunctive-intervention arm, while treatment-belief scores are similar. Because multiplicity handling is not reported, and because these are separate outcomes in a pilot trial, that pattern is exploratory. It does not establish a broad treatment effect across conditions or confirm the primary finding.

The appropriate reading is descriptive and cautious: the registry records a numerically lower primary mean in the CBT-I plus TAU group, but the available primary estimate is imprecise and the sponsor-reported analysis does not provide a coefficient or interval. These data alone do not support medical advice or a claim that the treatment works or does not work.

18. Why This Trial Matters Statistically

This trial is statistically interesting because its intervention targets sleep while its primary endpoint concerns suicidal ideation intensity. That makes the distinction between the intervention’s immediate target and the primary outcome important: a change in sleep symptoms would not automatically establish a change in suicidal ideation, and a pattern in secondary scores does not replace the primary comparison.

The randomized parallel-group design supports a direct comparison of an adjunctive behavioral sleep intervention with treatment as usual. The registry also reports both group summaries and a sponsor regression p-value, allowing readers to distinguish a transparent unadjusted contrast from the sponsor’s model-based analysis. This distinction is central to careful reading: a recalculation from summary statistics is informative, but it is not a reproduction of a regression whose specification is not reported.

The primary estimate demonstrates why uncertainty matters. The observed direction is clear in the group means, but the unadjusted interval includes zero and extends across a range of possible contrasts. A point estimate alone would obscure that imprecision. The secondary outcomes further illustrate why multiplicity and outcome hierarchy matter: several measures may be reported, but they do not all answer the same question or carry equal confirmatory weight.

Finally, the registry record leaves gaps in sample-size justification, missing-data handling, analysis populations, and model details. Those gaps are not merely administrative; they constrain how confidently the analysis can be evaluated and replicated. The trial can contribute a pilot estimate and clarify the kinds of outcomes that were studied, while its registry reporting also illustrates why effect estimates, uncertainty, and analysis details should be considered together.

19. 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.

This is a pilot trial with an enrollment of 54, and the registry reports primary group summaries based on 23 participants in CBT-I plus TAU and 27 in TAU Control. The registry does not explain the difference between enrollment and the posted group counts or state the analysis population. Missing-data procedures and reasons for any unavailable outcomes are not reported, limiting assessment of attrition and potential bias.

The record does not report the original sample-size justification, so it is not possible to determine the intended precision or target effect for the trial. Planning illustrations based on the observed effect are not a substitute for the original design rationale. The primary unadjusted confidence interval is broad and crosses zero, and the sponsor’s regression estimate and interval are not reported.

Other important analysis details are also absent from the registry record. The regression covariates and model specification are not given; model diagnostics are not reported; and it is not stated whether the analysis followed an intention-to-treat approach. Allocation concealment and the procedures used to maintain double masking are not described. These omissions make it difficult to assess potential sources of bias or reproduce the sponsor’s model.

The outcome timeframe is post-treatment, approximately 6-8 weeks. Longer-term outcomes, including sustained changes in ideation or suicide-related behavior, are not reported in the registry record. The registry also does not state a threshold for clinically meaningful change in the primary score. Finally, multiplicity adjustment for the secondary outcomes is not reported, so their statistical results should not be treated as confirmatory evidence independent of the primary analysis.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Adjunctive Brief Behavioral Treatment of Insomnia (BBTI) for Sleep Intervention (SI) — NCT02248675.
• Registry record last updated 2017-04-28; status last verified 2017-03.
• Results first posted 2017-04-28.
• 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.

23. Record Summary

The registry record describes a completed randomized, double-masked, parallel-group pilot trial sponsored by the VA Office of Research and Development, with enrollment of 54. It compared four individual sessions of CBT-I added to TAU with TAU alone in Veterans with insomnia and co-occurring conditions.

The primary endpoint was post-treatment C-SSRS Suicidal Ideation Intensity. The mean was 5.4 in CBT-I plus TAU and 8.4 in TAU Control. The unadjusted recalculated mean difference was -3.0, with a 95% confidence interval from -6.656 to 0.656 and p = 0.108. The sponsor-reported linear regression p-value was .153; its estimate and confidence interval are not reported.

Secondary outcomes concerned treatment beliefs, depression symptoms, and insomnia severity. The registry does not report the original sample-size rationale, multiplicity adjustment, missing-data approach, or analysis-population details. Serious and other adverse-event counts were zero in both groups; death counts are not reported.