Sleep to Prevent Evolving Affecting Disorders: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sleep to Prevent Evolving Affecting Disorders (NCT02988375), a randomized na study. The primary outcome is Severity of Depression as Measured by the Quick Inventory of Depressive Symptomatology (QIDS) Pre-Treatment.

2. Clinical Question

The registry’s broad research objective and the posted primary measure are not the same thing. The objective describes intended questions about depression and possible moderators, while the primary outcome actually reported is pre-treatment QIDS severity. This distinction matters when reading the results: a baseline comparison can describe groups before treatment, but it cannot establish a treatment effect on a later outcome. The registry record does not report a post-treatment depression result among the measures described in this record.

3. Trial Design

The registry identifies NCT02988375 as a completed, randomized, parallel-group trial sponsored by Henry Ford Health System. The recorded enrollment is 1385. The start date is 2016-03-08, and primary completion is dated 2018-04-26. The phase is listed as NA. The comparison was between dCBTI, delivered through online access to Sleepio, and Sleep Education, delivered as weekly email messages with sleep hygiene recommendations. Both are categorized as behavioral interventions.

The trial is listed as having no masking. That describes the masking status recorded in the registry; the record does not explain how participants, intervention staff, or outcome assessors were handled beyond that designation. It also does not report further details about allocation procedures, stratification, concealment, adherence, treatment exposure, or the precise sequence of follow-up assessments. Those details should not be inferred from the randomized and parallel-group labels.

For the posted primary outcome, the recorded analysis groups have 358 participants in dCBTI and 300 in Sleep Education. The registry’s adverse-event record lists different at-risk counts for its safety summaries. These counts have different reporting contexts, and the registry does not explain how they relate to enrollment or to the outcome-specific groups. They should not be treated as interchangeable denominators.

The primary outcome is recorded at pre-treatment, while the two posted secondary outcomes concern insomnia symptoms after treatment and at one-year follow-up. The record describes the post-treatment timing as approximately six weeks after screening and the longer follow-up as 58 weeks after screening. The registry lists six outcome measures overall, but only two secondary outcomes are described in the posted results represented in this record. The record does not report the trial’s own sample-size justification, a prespecified statistical analysis plan, or an account of interim 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

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 Severity of Depression as Measured by the Quick Inventory of Depressive Symptomatology (QIDS) Pre-Treatment, measured over Pre-treatment (Day 1).

The primary measure is severity of depression using the Quick Inventory of Depressive Symptomatology, or QIDS. The registry describes the instrument as a scale ranging from 0-27, with higher scores meaning greater depression severity. The posted measure is the mean score before treatment, on Day 1, and the registry reports a standard deviation for each arm.

A depression symptom scale can provide a structured way to summarize symptom severity across participants. Reporting a mean and standard deviation describes the group’s average score and the spread of individual scores around that average. This is useful for characterizing the groups at baseline and for understanding the scale on which a later change or between-group result might be assessed.

However, the timing is essential to interpretation. A pre-treatment measure records status before the interventions have had an opportunity to affect the measured outcome. It can therefore describe the groups’ starting scores, but it cannot by itself quantify a treatment-associated improvement, worsening, or difference in later depression severity. Although the registry’s objective also refers to depression incidence and relapse, the posted primary measure is a baseline severity score, not an incidence or relapse event measure.

The registry record does not report a post-treatment QIDS result or a depression change estimate in the results represented. The observed equality of the reported baseline means is a descriptive feature of these posted data, not evidence that the interventions had equivalent effects on depression. It also does not show that every participant had the same score or that the groups were identical on other characteristics.

7. Results Posted in the Registry

• Sleep Education: mean 10.80 versus 10.80 on control; difference 0.00 (95% CI -0.70 to 0.70), standardized effect d = 0.00; z-test p = 1.000.

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.

Severity of Symptoms of Insomnia - Post Treatment

• 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:
  • Sleep Education: mean 17.70 versus 17.90 on control; difference -0.20 (95% CI -0.87 to 0.47), standardized effect d = -0.05; z-test p = 0.557.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sleep Education and the control. The interval gives the range of effects that remain compatible with the data.

Severity of Symptoms of Insomnia - 1 Year Post Treatment

• 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:
  • Sleep Education: mean 12.80 versus 8.20 on control; difference 4.60 (95% CI 3.66 to 5.54), standardized effect d = 0.78; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Sleep Education than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary results concern insomnia symptom severity at two follow-up times, while the posted primary measure concerns depression severity before treatment. These outcomes therefore address related but distinct questions. The secondary results cannot be used as substitutes for a later QIDS comparison, and their pattern does not directly answer the registry’s stated questions about depression incidence or relapse.

Across the two follow-ups, the registry reports a near-zero post-treatment mean difference with an interval that includes zero, and a larger positive mean difference at one-year follow-up whose interval does not include zero. The direction of a mean difference is determined by the registry’s subtraction order; the arm-specific means are needed to understand which group had the higher average score. These summaries describe observed insomnia scores at each assessment, rather than within-person trajectories or changes from baseline. The registry record does not provide a reported analysis of change over time or a model comparing trajectories.

The secondary comparisons are unadjusted, and no reported analyses are listed for them in the record. Their p-values should consequently be interpreted as individual, unadjusted results, not as a complete confirmatory testing strategy. When multiple outcomes or time points are examined, the chance of at least one apparently notable result can increase. Whether and how the trial addressed that issue is not reported in the registry record. This is one reason a secondary result carries less weight than a prespecified primary analysis with a clearly documented analysis plan, especially when the primary measure posted is not a post-treatment outcome.

The registry’s planning illustrations for these outcomes are based on observed effects and stated assumptions. They can show how planning estimates vary with assumed effect size, but they do not establish the study’s original target, the power actually achieved, or the reliability of the observed effect. Overall, the secondary outcomes offer follow-up descriptions of insomnia symptoms, but the limited analysis detail and the distinction from the primary depression measure constrain how broadly they can be interpreted.

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

• Severity of Symptoms of Insomnia - Post Treatment
• Severity of Symptoms of Insomnia - 1 Year Post Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference represent? For a continuous outcome, each arm’s mean summarizes the average observed score among the participants included in that outcome’s summary. A mean difference compares those averages. Its sign depends on the order used for subtraction, so it must be interpreted together with the arm means and the registry’s comparison convention. The comparison is unadjusted in the posted statistics: it does not account for other participant characteristics through a regression or other adjusted model.

For the primary outcome, the two reported means are both 10.8, and the mean difference is 0.0. This is a comparison of pre-treatment scores. It is not a comparison of changes over time, and it is not a post-treatment contrast. A baseline difference can be useful descriptively, but a baseline measure cannot indicate what happened after the intervention. In a randomized trial, random allocation is intended to support a fair comparison, but the registry’s reported baseline means alone do not demonstrate balance on every measured or unmeasured characteristic.

What does the standard deviation describe? A standard deviation describes the spread of observed scores around the mean within an arm. It is different from the standard error, which describes the sampling uncertainty in an estimate such as a mean. The primary arm standard deviations are 4.5 and 4.6. These values indicate that individual scores vary around the same reported mean; equal means do not imply identical score distributions or identical participants.

What does a confidence interval say? A confidence interval gives a range of values compatible with the observed data under the statistical procedure and its assumptions. It communicates uncertainty around an estimate, rather than only whether a null hypothesis was rejected. For the primary result, the interval spans zero, which is consistent with the absence of a detected baseline mean difference and also allows differences in either direction. The interval concerns the baseline comparison, not a treatment effect on later depression severity.

The interval for the post-treatment insomnia comparison also spans zero, while the interval for the one-year insomnia comparison lies above zero in the registry’s displayed subtraction order. The latter interval is relatively more separated from zero, and its unadjusted p-value is <0.001. These statements describe statistical evidence for the particular posted comparisons; they do not establish the importance of the outcome to participants or settle whether the analysis was prespecified.

How should a p-value be read? A p-value is calculated under a null hypothesis and statistical model. It indicates how unusual results at least as discrepant as the observed result would be if that null model were correct. It is not the probability that the null hypothesis is true, and it does not estimate the probability that the result will replicate. A p-value of 1.0 for the primary baseline comparison is consistent with the equal reported means. It does not prove equivalence; equivalence requires a prespecified margin and an analysis designed to assess whether uncertainty lies within that margin. No such margin is reported in the registry record.

Why do outcome timing and analysis population matter? The clinical meaning of an estimate depends on when the measure was collected. Here, the primary QIDS score is explicitly pre-treatment. The insomnia outcomes are recorded at post-treatment and one-year follow-up. They answer different measurement questions and should not be blended into a single claim about depression. The numbers of participants contributing to summaries can also affect precision and representativeness. The record does not identify the analysis population, explain missing data, or state whether participants were analyzed according to randomized assignment.

What do planning illustrations mean? The registry includes planning illustrations based on observed effects for the two secondary insomnia outcomes. These illustrations use the assumptions stated in the record: a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as stated fractions of the observed effect. They are not the trial’s own design justification, because that justification is not reported in the registry record. They are also not observed power. Planning calculations based on an observed effect can be unstable, particularly when that effect is imprecise or small. A planning number should be understood as an illustration under its assumptions, not as a guarantee of achieving a desired result in a future study.

Why can over-dispersion matter? In general, over-dispersion means that the observed variability is greater than a statistical model assumes. When a model underestimates variability, standard errors can be too small, confidence intervals too narrow, and tests too optimistic. This is a general statistical consideration, not a finding reported for this trial. The posted outcomes are continuous scores summarized with means and standard deviations; the record does not report an over-dispersion assessment or a model that would permit a conclusion about it.

The registry’s displayed comparisons should therefore be read as summary-statistic comparisons with the stated uncertainty, not as a complete account of the trial’s analytical methods. The registry record does not report model specifications, covariate adjustment, diagnostic checks, missing-data procedures, or sensitivity analyses. Without those details, the posted estimates can be described, but the robustness of the estimates to alternative analytical choices cannot be assessed from this record.

11. Confidence Intervals and Precision

Precision concerns how tightly an estimate is determined by the observed data. The estimate alone is not enough: the confidence interval shows a range of values compatible with the data under the method used. A narrow interval indicates greater precision than a wide interval, although precision does not establish that an estimate is unbiased or clinically important.

For the primary baseline QIDS comparison, the estimated mean difference is 0.0. The confidence interval extends on both sides of zero, and the reported p-value is 1.0. The interval therefore does not rule out a difference in either direction. Because this is a pre-treatment comparison, its uncertainty concerns the difference in starting scores, not a later effect of dCBTI compared with Sleep Education. The record does not report an equivalence margin, so the result should not be described as proof that the groups are equivalent.

The secondary insomnia estimates have different levels of precision. At post-treatment, the mean difference is -0.2 and the interval includes zero. At one-year follow-up, the mean difference is 4.6 and the reported interval runs from 3.663 to 5.537. These intervals pertain to the particular outcomes and follow-up times recorded; they do not provide precision for the unreported post-treatment depression question.

Precision also depends on the number of observations contributing to an outcome and the variability in scores. The registry gives outcome-specific counts and standard deviations, but does not explain missing observations or the analysis population. It is therefore not possible to determine from this record why the contributing counts differ from total enrollment, or whether missingness could affect the estimates. The confidence intervals quantify sampling uncertainty under the reported calculation; they do not capture every possible source of bias or uncertainty about the conduct and analysis of the trial.

12. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, time points, subgroups, or analytical choices are tested. Each additional test can increase the chance of at least one statistically notable result even if all null hypotheses are true. Common approaches include prespecifying a hierarchy of outcomes, adjusting significance thresholds, or treating secondary analyses as exploratory. This is a general statistical issue; the registry record does not report which approach, if any, was used for this trial.

The record lists six outcome measures overall and describes two secondary insomnia outcomes in the results represented. It gives unadjusted comparisons for the secondary outcomes, but no reported analyses are listed. It does not report a multiplicity adjustment, a hierarchy among outcomes, or a statistical analysis plan defining confirmatory and exploratory tests. Consequently, the unadjusted p-values should not be read as if they necessarily controlled the overall false-positive risk across all outcomes and analyses.

Missing-data handling is also not reported in the registry record. The outcome summaries include 358 participants for dCBTI and 300 for Sleep Education for the primary outcome; the follow-up outcome counts differ from those baseline counts. These differences alone do not identify whether participants missed assessments, were excluded, or were handled in another way. The record does not report reasons for missing values, imputation, weighting, likelihood-based methods, or sensitivity analyses. It is not possible to evaluate from the posted information whether missingness could have affected the results.

The analysis populations are not described. The record does not state whether the estimates follow randomized assignment for all eligible participants, include only participants with available outcome data, or use another population. It also does not report adherence-based exclusions. Randomization is a design feature, but the interpretability of an outcome comparison depends on how participants are included in the analysis and how missing outcomes are addressed. Those details are not reported in the registry record.

The same caution applies to interim analyses, covariate adjustment, and subgroup analyses. The registry’s brief summary mentions possible moderators, but the record does not report moderator estimates or a plan for testing them. It does not report interim analyses or the trial’s sample-size justification. Without those details, the posted estimates can be described as recorded, but the complete confirmatory framework and the potential influence of analytical choices cannot be assessed.

13. 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 registry’s adverse-event record lists zero serious adverse events, zero other adverse events, and zero deaths in each of its two safety groups. It records 946 participants at risk in the dCBTI safety group and 439 at risk in the Sleep Education safety group. These are the denominators stated for the safety summaries, not the outcome-specific participant counts used for the continuous score results.

Zero events in the recorded groups means that no events of those categories are listed in those registry summaries. It should not be taken to mean that the interventions are free of risk, or that no event could have occurred outside the recorded definitions, reporting period, or safety population. The record does not give exposure time, event ascertainment procedures, detailed definitions, or an explanation of how the at-risk groups relate to enrollment and outcome groups. No serious-event terms are listed. These reporting limits prevent a fuller comparison of safety rates or an assessment of how completely events were collected.

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

The posted primary result means that the two outcome groups had the same reported average pre-treatment QIDS score: 10.8 in each arm, with an unadjusted mean difference of 0.0. Its confidence interval spans zero, and its p-value is 1.0. This is a description of baseline depression severity in the participants included in that summary. It is not evidence about whether dCBTI changes depression severity relative to Sleep Education after treatment.

The registry’s stated objective includes reducing depression severity and preventing depression incidence and relapse among people with insomnia. Those are clinically distinct outcomes that require measurements after the intervention or over follow-up. The primary outcome posted in this record is a pre-treatment score. The registry record does not report a post-treatment depression comparison, a depression incidence estimate, or a relapse estimate in the results. Therefore, those intended questions cannot be answered from the reported primary result.

The secondary outcomes describe insomnia symptoms at post-treatment and one-year follow-up. Their estimates and uncertainty are specific to those measures and assessment times. They do not show that a change in insomnia symptoms caused a change in depression, nor do they establish the effect of treatment on depression prevention. The record does not report a causal mediation analysis or a joint analysis of insomnia and depression outcomes.

Statistical significance and clinical importance are also different concepts. A p-value reflects compatibility with a null model under assumptions; it does not determine whether a difference matters to participants. The registry record does not report a clinically important difference threshold for these outcomes. Nor does the baseline p-value of 1.0 demonstrate equivalence, since no equivalence margin or equivalence analysis is reported.

The most defensible reading is consequently limited: the posted baseline depression averages are equal, and the registry also contains follow-up summaries for insomnia symptoms. These results do not establish that either intervention succeeds or fails at the broader goals described in the trial summary. Further interpretation would require relevant post-treatment depression outcomes and transparent reporting of analysis populations, missing-data methods, and prespecified comparisons.

15. Why This Trial Matters Statistically

This trial illustrates why the name or broad objective of a study should not substitute for checking the outcome that was actually posted. The registry’s summary describes goals involving depression severity, incidence, and relapse, while the primary outcome reported in these results is pre-treatment depression severity. Recognizing that timing prevents a baseline comparison from being mistaken for a treatment-effect estimate.

It also demonstrates the value of reading means, uncertainty intervals, and outcome timing together. Identical means and a p-value of 1.0 describe the observed baseline averages, but do not prove equivalence or establish what happened later. The follow-up insomnia estimates answer a different question and have different precision. Keeping those outcomes distinct is essential to avoid transferring a result from one measure or time point to another.

The record also highlights how statistical interpretation depends on transparent reporting beyond an estimate and p-value. Analysis populations, missing-data methods, multiplicity plans, and the original sample-size rationale all affect how confidently a result can be interpreted. The registry does not report those details for the analyses. Its planning illustrations are explicitly based on observed effects and assumptions, and should not be confused with the trial’s design justification or observed power.

For clinical researchers and biostatisticians, the central statistical lesson is one of alignment: the claim, endpoint, measurement timing, analysis population, and inferential method need to correspond. If the scientific question concerns a post-treatment outcome, a pre-treatment measurement cannot answer it. If several outcomes are analyzed, the inferential role of each should be clear. The reported figures are useful descriptions, but the record’s limits define how far the conclusions can extend.

16. 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.

A central limitation is that the posted primary outcome is measured before treatment. This prevents the primary result from directly evaluating the stated objective of reducing depression severity. The registry’s brief summary also names depression incidence and relapse, but no corresponding estimates are reported in the results represented. The relationship between the stated objectives and the posted outcome set is therefore not fully documented.

The registry record gives little detail about the analysis methods. It does not report the trial’s statistical analysis plan, sample-size justification, interim analyses, covariate adjustment, analysis populations, or missing-data handling. It also does not report multiplicity control or an account of whether secondary outcomes were prespecified as confirmatory or exploratory. These omissions restrict assessment of bias, robustness, and the intended interpretation of the p-values.

The outcome-specific participant counts differ, and the safety denominators do not match the counts used for the score summaries. The record does not explain these differences. They should not be treated as proof of attrition or as evidence of a particular analysis population. Without reasons for missing or unavailable outcome values, it is not possible to judge the potential impact of missingness.

The trial is recorded as randomized and unmasked, but the registry information does not provide further allocation procedures, adherence data, or details of outcome assessment. The available results are unadjusted summaries, and the two secondary measures concern insomnia rather than the primary depression measure. The safety record reports zero events in its listed categories but does not provide event ascertainment details or exposure time. Each limitation narrows the conclusions that can be drawn from this registry record alone.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Sleep to Prevent Evolving Affecting Disorders — NCT02988375.
• Registry record last updated 2024-01-12; status last verified 2024-01.
• Results first posted 2024-01-12.
• 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.

20. Record Summary

NCT02988375 is a completed, randomized, parallel-group, unmasked trial sponsored by Henry Ford Health System, with recorded enrollment of 1385. It compared online access to the Sleepio dCBTI program with weekly sleep-hygiene email messages. The posted primary outcome is pre-treatment QIDS depression severity; the arm means are both 10.8, with an unadjusted mean difference of 0.0 and p-value of 1.0. Two secondary insomnia outcomes are reported at post-treatment and one-year follow-up. The record does not report post-treatment depression results, analysis populations, missing-data methods, multiplicity procedures, or the trial’s own sample-size justification. Safety summaries list zero serious adverse events, other adverse events, and deaths in each safety group.