Comparing Internet and In-Person Brief Cognitive Behavioral Therapy of Insomnia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparing Internet and In-Person Brief Cognitive Behavioral Therapy of Insomnia (NCT01549899), a randomized phase 3 study. The primary outcome is Sleep Diary Sleep Efficiency.

2. Clinical Question

The study objective was to compare in-person and Internet-administered CBTi with minimal contact in relation to sleep improvement. The brief summary also identifies depression, substance use, and PTSD symptoms as areas of interest for comparisons involving the in-person CBTi and MC groups. The primary endpoint is self-reported sleep efficiency; the stated objective also refers to subjective and objective sleep measures. The registry record does not report an estimate showing how the groups differed on the primary endpoint, or whether any difference was statistically precise.

3. Trial Design

The registry describes a randomized, parallel-group, phase 3 study with single masking. It lists the study status as completed and enrollment as 185. The study start is recorded as 2012-03 and primary completion as 2015-09. The record does not identify who was masked, how the randomization sequence was generated, or how allocation was concealed.

The brief summary describes three groups: in-person CBTi, Internet CBTi, and minimal contact. It describes six CBTi sessions over six weeks. The in-person treatment is specified as six weekly 60-minute sessions and includes identical informational material. The Internet treatment is described as having identical information and instructions to the in-person treatment, administered through the afterdeployment.org website. Details of the intervention descriptions are incomplete in the registry record, so the remaining treatment content cannot be characterized from that record.

The MC group was asked to avoid other therapist contact or additional insomnia treatment during the six-week period, and was contacted every other week. These details describe the intended control condition, not a measured level of adherence. The registry record does not report how many participants followed each assigned treatment, completed treatment, or remained in follow-up.

The enrollment figure and the brief summary do not agree: the record lists 185, while the summary describes a total of 189 and planned group allocations of 77, 35, and 77. The registry does not explain the difference, so neither figure should be treated as a reconciled account of the analyzed population. The record also does not report a statistical analysis plan or explain the rationale for the enrollment target.

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 Sleep Diary Sleep Efficiency, measured over Change from Baseline to Post-Treatment and 6 month follow-up.

The primary endpoint is Sleep Diary Sleep Efficiency, expressed as Percentage Total Sleep Time/Time in Bed. Participants’ daily sleep diaries record bedtime, time to fall asleep, awakenings during the night, and time out of bed. The record says these reports are aggregated to estimate sleep efficiency using total sleep time divided by time in bed and multiplied by 100. Total sleep time and total wake time are also identified as variables to be extracted.

Sleep efficiency is used to describe the portion of time in bed spent asleep. As a diary-based measure, it captures participants’ reported sleep experience rather than a direct instrument-based measurement. The study’s broader description also refers to objective sleep measures, but this primary endpoint is explicitly based on daily self-reports.

The planned endpoint timeframe is change from baseline to post-treatment and six-month follow-up. Those are separate assessment points, so a result at one point would not automatically describe results at another. The registry record does not provide the values, group contrasts, uncertainty estimates, or analysis details for these assessments. It therefore does not support a conclusion about the size or precision of change in sleep efficiency.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The design is described as randomized and parallel. In general, randomization assigns participants to study groups using a chance-based procedure, which can reduce systematic differences between groups at the start of a trial. A parallel design means groups receive their assigned approaches during the same study period rather than switching between approaches. The registry records these design labels but does not report the sequence-generation method, allocation concealment, or procedures used to maintain masking. “Single” masking is recorded, but which people were masked is not reported in the registry record.

What does the primary measure represent?

The registry identifies the primary outcome as a continuous measure summarized by a mean. A mean describes the average value in a group; it does not show how widely individual participant values vary. Sleep efficiency is calculated from diary-derived total sleep time and time in bed, expressed as a percentage. The stated assessment is change from baseline to post-treatment and six-month follow-up. The record does not report the exact statistical model used to estimate changes or compare groups.

The registry lists standard error as the dispersion measure. In general, a standard error describes the precision of an estimated mean, rather than the spread of individual observations. It can be used to construct an interval estimate when the necessary method and assumptions are specified. Standard errors alone do not show the full distribution of participants’ sleep-efficiency values. The record does not include the actual estimates or standard errors for the primary endpoint.

Why does the timeframe matter?

Repeated assessments can provide information about whether measured values change over time and whether group patterns differ across assessments. In general, measurements on the same participant tend to be related, so an analysis of repeated observations should account for that dependence. The registry identifies baseline, post-treatment, and six-month follow-up timing, but it does not report whether a repeated-measures model was used, how within-participant correlation was handled, or whether separate contrasts were estimated at each time point.

The endpoint is based on daily self-reports aggregated into sleep measures. In general, daily observations can vary within a participant, and the way diary records are aggregated can affect the resulting summary. The registry describes which diary items are collected and the sleep-efficiency calculation, but the exact aggregation rules, completeness requirements, and handling of missing diary entries are not reported.

What can the design labels establish?

Randomization and masking are features of the intended design, not proof that all potential sources of bias were eliminated. In general, masking can limit the influence of expectations on care or assessment, but its value depends on who is masked and whether masking is maintained. Those operational details are not reported. The record also does not report adherence, crossover to other treatment, or departures from assigned intervention. The design description alone therefore cannot establish how treatment exposure unfolded in practice.

The study includes more than one active intervention and a minimal-contact control. The brief summary describes intended comparisons of the groups on subjective and objective sleep measures, and identifies additional areas of interest. That description does not establish that every possible pairwise comparison was performed. The registry records zero analyses and does not include a statistical analysis plan, so the tested contrasts, model covariates, and decision rules are not reported.

How should planning calculations be read?

The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The hypothetical effect is expressed as a stated fraction of the observed effect. In general, a planning calculation estimates the sample size needed under specified assumptions to detect an effect at a chosen significance level and power. It does not show the probability that the completed trial detected an effect, and it is not observed power. These illustrations should not be treated as the trial’s original design justification; that justification is not reported in the registry record.

In general, over-dispersion means that observed variability is greater than expected under a simpler statistical model. If present, it can make standard errors too small unless the analysis accounts for it. The registry record does not report whether over-dispersion was relevant to the primary outcome or whether any adjustment was made. More broadly, the absence of a reported model and results prevents assessment of assumptions, model fit, or robustness. The design and endpoint definitions clarify the intended scientific question, but they are not enough to reconstruct the analysis or estimate its findings.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint has assessments at baseline, post-treatment, and six-month follow-up. The registry flags that the primary outcome involves several categories or time points and was not reduced to one comparison. It does not report which time point or contrast was designated as the principal test, or how any multiple comparisons were handled. When several outcomes, groups, or time points are examined, the chance of at least one apparently unusual result can increase; in general, analysis plans may address this through prespecified testing strategies or adjustments. Whether such procedures were used is not reported in the registry record.

The record lists 11 outcome measures and names secondary measures including the Beck Depression Inventory II, Substance Use, Actigraphy, PTSD Check List-Military, Beck Anxiety Inventory, Insomnia Severity Index, Epworth Sleepiness Scale, and Dysfunctional Beliefs and Attitudes About Sleep Scale. The brief summary describes some additional domains as interests of the CBTi and MC comparison. The registry record does not report results or a multiplicity strategy for these outcomes.

Missing-data methods are also not reported. The record does not give the amount or pattern of missing outcome data, how incomplete diaries or follow-up assessments were treated, or whether imputation or another method was used. It does not state whether the analysis followed an intention-to-treat principle, used a per-protocol population, or applied another analysis-population definition. The randomized design label does not establish which participants contributed to a particular analysis. Without these details, the potential influence of missing observations and analysis-population choices cannot be evaluated from the registry record.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: accident (4.0%), hypomania (1.3%), alcohol-detox (1.3%).

The registry lists serious adverse events by group as affected counts and numbers at risk. For in-person CBTi, it reports 5 serious adverse events among 75 at risk; for Internet CBTi, 0 among 34 at risk; and for minimal contact, 1 among 76 at risk. These are descriptive registry figures. They do not by themselves establish whether an event was related to treatment, how events were assessed, or whether the groups differed statistically.

For other adverse events, the record lists 47 affected among 75 at risk in the in-person group, 8 among 34 at risk in the Internet group, and 9 among 76 at risk in the minimal-contact group. The record does not provide event-level descriptions for these figures in the material. The listed top serious-event terms are Accident at 4.0%, Hypomania at 1.3%, and Alcohol-Detox at 1.3%. The record does not assign these terms to specific groups.

Counts and percentages should be read in light of their definitions, observation periods, and ascertainment procedures. Those details are not fully reported in the registry record. Serious-event counts are not evidence of causation, and differences in raw counts alone do not account for uncertainty or exposure. The registry also contains a separate deaths field, but its interpretation is not explained in the record.

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

The record establishes that a randomized, parallel-group study was conducted to compare in-person and Internet CBTi with minimal contact in an active duty military population with chronic insomnia. It identifies sleep-diary sleep efficiency as the primary endpoint and specifies change from baseline to post-treatment and six-month follow-up. The intervention descriptions and selected safety counts provide context for the intended comparison.

It does not establish the size or direction of any treatment difference in sleep efficiency. No primary outcome estimates, confidence intervals, or p-values are reported in the registry record, and the record flags that the primary outcome includes multiple time points rather than a single comparison. It therefore cannot establish whether any observed difference was statistically precise, how it varied over time, or how it compared with a threshold for practical importance.

The record’s study objective is not itself a result. The named secondary outcomes likewise do not establish changes in depression, substance use, PTSD symptoms, anxiety, sleepiness, or other measured domains. The safety counts describe reported events in groups but do not establish that an intervention caused or prevented them. No medical decision should be based on these registry entries alone.

Interpretation is further limited by unreported analysis methods, missing-data procedures, analysis populations, and details of masking. The record’s enrollment figure also differs from the total described in the brief summary, without an explanation. These gaps do not imply that a particular result is favorable or unfavorable; they indicate that the registry record does not contain enough information to characterize the study’s estimated effects or their reliability.

13. Why This Trial Matters Statistically

This trial is statistically notable because its primary measure is collected repeatedly and because the study compares more than one intervention with a control condition. A change-from-baseline endpoint assessed after treatment and at follow-up raises questions about which time point and contrast should anchor interpretation. If multiple comparisons are made, a prespecified strategy is important for distinguishing primary tests from exploratory ones. The registry record does not report that strategy.

The outcome also combines diary observations into a participant-level sleep-efficiency measure. That makes measurement and aggregation choices relevant: the definition of a complete diary, treatment of missing days, and aggregation method can affect the estimate. The record describes the diary items and the calculation but does not report those implementation details or the analysis model.

Finally, the registry illustrates the difference between a planned comparison and a reported result. It documents randomization, intervention groups, an endpoint, and a study objective, yet does not report the primary analysis estimates. Statistical interpretation depends not just on a study’s question but also on transparent estimates, uncertainty, analysis choices, and participant flow. The record leaves those elements unavailable, so it supports understanding of the design more than evaluation of comparative effects.

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 specific to this registry record constrain interpretation. The primary endpoint is assessed at more than one time point, but no single comparison is reported. The record lists zero analyses and does not provide primary outcome estimates, uncertainty intervals, or p-values. Without these, the magnitude, direction, and precision of any group difference cannot be assessed.

The record does not report the statistical analysis plan, the method used to handle repeated observations, missing-data procedures, analysis populations, or multiplicity control. It also does not identify who was masked or describe the randomization and allocation procedures. These omissions prevent evaluation of important design and analysis choices.

The enrollment information is inconsistent within the record: enrollment is listed as 185, while the brief summary describes 189 personnel and planned group allocations. No reconciliation is reported. Intervention descriptions are also incomplete, and adherence, treatment exposure, and follow-up completion are not reported. The safety figures lack details about event ascertainment, observation periods, and attribution. The record does not explain the separate deaths field.

These are limitations of what the registry record reports; they do not establish that the underlying study lacked these procedures or analyses. They do mean that the reported record alone cannot support a detailed assessment of the results’ robustness or generalizability.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Comparing Internet and In-Person Brief Cognitive Behavioral Therapy of Insomnia — NCT01549899.
• Registry record last updated 2025-04-17; status last verified 2025-04.
• Results first posted 2025-04-17.
• 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 completed, randomized, single-masked, parallel, phase 3 study in active duty military personnel with chronic insomnia. It compares in-person CBTi and Internet CBTi with a minimal-contact control. The primary endpoint is sleep-diary sleep efficiency, defined from total sleep time and time in bed, with change assessed from baseline to post-treatment and six-month follow-up.

The record lists enrollment as 185, while the brief summary describes 189 personnel and planned allocations of 77, 35, and 77. It does not reconcile the difference. It records zero analyses and does not report a primary endpoint estimate or a single comparison. Statistical methods for missing data, multiplicity, and analysis populations are not reported. Safety counts and selected serious-event terms are listed, but their attribution and ascertainment details are not explained. The record documents the intended question and design, but not enough result information to judge comparative effects.