Pilot Study of a Combination Drug Product for Treatment of Short-term Insomnia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pilot Study of a Combination Drug Product for Treatment of Short-term Insomnia (NCT02671760), a randomized phase 2 study. The primary outcome is Total Sleep Time.

2. Clinical Question

The clinical question is narrow: it concerns measured sleep duration in a structured experimental model, not long-term sleep patterns or chronic insomnia. The brief summary describes three one-night stays in a sleep center in New York City. The trial was randomized, quadruple-masked, and used a crossover model, so participants received study conditions in a sequence rather than being assigned to only one parallel group. The registry record does not report the sequence details or the methods used to account for repeated measurements within a participant.

3. Trial Design

The record describes a completed Phase 2 randomized, double-blind, single-dose study using a crossover model. Masking is listed as quadruple. Its official title identifies a 5 hour phase advance model of insomnia in adults with short-term insomnia. The brief summary says participants stayed overnight at a sleep center in New York City on three occasions. The study began in 2016-02, and primary completion was recorded as 2016-06.

The three intervention arms are SM-1, a 2-drug comparator, and placebo. In the crossover design, each participant can contribute measurements under more than one study condition. That feature can make within-person comparisons informative because each person may serve as their own reference. It also means that analysis should account for the relationship among measurements from the same participant. The registry record does not report the treatment sequence, washout arrangements, or the statistical model used for the crossover data.

Enrollment is listed as 39. The primary endpoint is mean total sleep time, measured in minutes over an 8-hour timeframe. Secondary outcomes listed in the record include latency to persistent sleep, awakenings, latency to REM sleep onset, adverse events, and residual sleepiness safety and tolerability. The registry record does not report eligibility criteria in detail, the planned analysis population, or a statistical analysis plan. It lists no reported analyses, so the numerical comparisons discussed in the record should not be read as a full account of the trial’s prespecified analysis.

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 Total Sleep Time, measured over 8 hours.

Total sleep time is the amount of time a participant is recorded as asleep during the assessment period. The primary endpoint is reported as a mean in minutes over an 8-hour timeframe. In this study, it reflects sleep during the experimental night rather than sleep accumulated across ordinary nights or a longer follow-up period.

Mean total sleep time is useful because it summarizes the duration of sleep in a form that can be compared across study conditions. A difference in means describes the observed separation between the average values, while a confidence interval communicates uncertainty around that estimated difference. The mean alone does not show how individual participants varied; the registry also reports standard deviations for each arm.

The measure is directly connected to the study’s stated purpose of evaluating effects on total sleep time. However, it is one aspect of sleep, not a complete measure of sleep quality, daytime function, or longer-term outcomes. The registry also lists other sleep and safety outcomes, but the primary endpoint remains the stated main measure. The record does not report a minimum difference considered clinically important, so the statistical estimate alone cannot establish whether the size of a difference is important to patients.

7. Results Posted in the Registry

• Treatment: mean 382.90 versus 339.20 on control; difference 43.70 (95% CI 13.56 to 73.84), standardized effect d = 0.64; z-test p = 0.004.
• Placebo: mean 256.20 versus 339.20 on control; difference -83.00 (95% CI -122.07 to -43.93), standardized effect d = -0.94; z-test p = <0.001.

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.

Latency to Persistent Sleep

• 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:
  • Treatment: mean 31.20 versus 42.10 on control; difference -10.90 (95% CI -34.06 to 12.26), standardized effect d = -0.21; z-test p = 0.356.
  • Placebo: mean 50.10 versus 42.10 on control; difference 8.00 (95% CI -20.17 to 36.17), standardized effect d = 0.13; z-test p = 0.578.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Treatment and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Awakenings

• 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:
  • Treatment: mean 12.20 versus 11.00 on control; difference 1.20 (95% CI -1.76 to 4.16), standardized effect d = 0.18; z-test p = 0.427.
  • Placebo: mean 9.60 versus 11.00 on control; difference -1.40 (95% CI -4.17 to 1.37), standardized effect d = -0.22; z-test p = 0.322.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Treatment and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Latency to REM Sleep Onset

• 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:
  • Treatment: mean 112.50 versus 71.30 on control; difference 41.20 (95% CI 19.46 to 62.94), standardized effect d = 0.84; z-test p = <0.001.
  • Placebo: mean 61.20 versus 71.30 on control; difference -10.10 (95% CI -28.06 to 7.86), standardized effect d = -0.25; z-test p = 0.270.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Treatment than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Placebo and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the description beyond total sleep time. They include other sleep measures, such as sleep onset and awakenings, as well as safety and tolerability outcomes. The posted continuous secondary results do not all point in a single direction relative to the comparator. Some estimates are on one side of zero and some on the other, and several intervals include zero. This mixed pattern is a reason to treat the secondary findings as exploratory rather than as a single coherent confirmation of the primary result.

The secondary endpoints relate to the primary question by describing other features of the same short-term sleep assessment. They do not measure the same construct as total sleep time: latency measures concern timing, while awakenings concern sleep interruptions. A difference in one measure need not imply a corresponding difference in another. The secondary outcomes may help characterize the experimental night, but they do not substitute for the primary endpoint or extend the study to long-term functioning.

The registry record lists multiple outcomes and does not report a multiplicity adjustment or a statistical analysis plan. When many outcomes or contrasts are examined, the chance of obtaining a small unadjusted p-value somewhere among them can increase. That general concern makes isolated secondary p-values less persuasive than results from a prespecified analysis with an appropriate approach to multiplicity. The record lists no reported analyses, so the status of each secondary measure as prespecified or exploratory is not reported in the registry record.

Accordingly, the secondary results are best read as descriptive context around the primary sleep-duration finding. They do not independently establish a broad effect across all aspects of sleep, and they should not be used to resolve uncertainty about the crossover analysis or the clinical importance of the primary estimate.

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

• Latency to Persistent Sleep
• Awakenings
• Latency to REM Sleep Onset
• Adverse Events
• Safety and Tolerability in Terms of Residual Sleepiness
• Safety and Tolerability in Terms of Residual Sleepiness

9. Statistical Interpretation

10. Statistical Methods Explained

The posted endpoint summaries are means and standard deviations. A mean is the arithmetic average of the recorded values, and a standard deviation describes how spread out participant measurements are around that average. These summaries give a compact description of each arm, but they do not show the full distribution or the pattern of individual responses.

What does a mean difference represent?

A mean difference subtracts one arm’s mean from another’s. For the primary comparisons, the comparator is the reference. The SM-1 contrast is SM-1 minus comparator; the placebo contrast is placebo minus comparator. The sign therefore depends on which arm is subtracted from which. A positive estimate in the first contrast means the SM-1 mean is higher than the comparator mean; a negative estimate in the second means the placebo mean is lower than the comparator mean.

The registry lists the arm means and standard deviations but no reported analyses. The displayed comparison estimates are unadjusted recalculations from those posted summary statistics. They do not include adjustment for other variables, and the record does not explain whether the original sponsor analysis accounted for the crossover structure. In a crossover study, repeated measurements from one person are generally related rather than independent. An analysis that ignores that relationship can represent uncertainty differently from an analysis that models it. The within-person data needed to assess that issue are not reported in the registry record.

What does a confidence interval add?

A confidence interval gives a range of values compatible with the estimate under the calculation’s assumptions. It helps distinguish a precise estimate from one with substantial uncertainty. A narrow interval indicates less uncertainty than a wide interval, while an interval that includes zero is compatible with no difference as well as with differences in either direction. The interval is not a range containing every participant’s outcome, and it is not a probability statement about the particular treatment effect.

How should the p-value be read?

A p-value measures how unusual the observed result, or a result more extreme, would be if the null hypothesis and the calculation’s assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The registry comparisons are unadjusted. Since the record lists multiple outcomes and comparisons but no multiplicity procedure, the reported p-values should not be treated as adjusted for all opportunities to observe a small p-value.

Why does the crossover structure matter?

In a crossover design, a participant may receive multiple conditions, making paired comparisons possible. The analysis can use the fact that each person serves as their own reference, but it needs information about the pairing, treatment order, and any period or carryover effects. Those details are not reported in the registry record. The recalculated summaries compare arm-level means without access to participant-level pairing, so they cannot reproduce a crossover model or quantify the benefit of within-person comparison.

The registry lists a sample-size illustration based on the observed effect. Such planning calculations estimate how many participants per arm would be needed for a future two-arm comparison under stated assumptions. They are not an analysis of the current trial’s probability of detecting an effect. Because the original design assumptions are not reported in the registry record, the planning illustration should be kept separate from the study’s actual design rationale and analysis.

11. Confidence Intervals and Precision

For the primary comparison of SM-1 with the comparator, the mean difference is 43.7 minutes, with a 95% confidence interval from 13.56 to 73.84. This interval gives a range of values compatible with the unadjusted recalculation and its assumptions. It is entirely above zero, but its width also shows that the exact size of the difference is uncertain.

For placebo compared with the comparator, the mean difference is -83.0 minutes, with a 95% confidence interval from -122.1 to -43.93. This interval is entirely below zero, indicating that the posted mean was lower for placebo in this contrast. Its range likewise allows for a variety of possible magnitudes rather than identifying a single exact effect.

Confidence intervals do not describe the range of individual sleep times. The standard deviations reported for the arms describe participant-level variation, whereas the intervals describe uncertainty in the estimated mean differences. The two forms of spread answer different questions.

Precision depends on the data and on the analysis model. Here, the reported comparisons are unadjusted recalculations from summary statistics, while the study used a crossover design. The registry record does not report participant-level paired data or the sponsor’s statistical model. Therefore, these intervals should be interpreted as precision estimates for the recalculations, not as a substitute for an unreported crossover analysis. The record also does not report a clinically important threshold, so the intervals cannot by themselves determine whether the range of plausible differences would be meaningful in practice.

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 illustrations for a future two-arm comparison, not estimates of the number needed to repeat this exact crossover study. They are reported as participants per arm, not as total events. The calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a design with an 80% chance of rejecting the null hypothesis if the assumed effect and calculation assumptions hold. The 90% power column gives the corresponding estimate for a 90% chance. These are design targets, not guarantees that a future study will detect an effect. The values are planning illustrations based on an observed estimate and should not be mistaken for a design selected prospectively for this trial.

Rows that plan for a smaller fraction of the observed effect require more participants per arm. That pattern reflects the general statistical principle that smaller differences are harder to distinguish from random variation. A planning calculation is sensitive to the assumed effect: if the true effect differs from the observed estimate, the actual operating characteristics can differ from the illustration.

This is not observed power. The calculations do not tell whether the completed trial had enough power after the fact, and they do not validate the observed results. They also use equal allocation assumptions for two-arm planning, whereas this registry record describes a crossover study. The original analysis plan and the link between these illustrations and any future design are not reported in the registry record.

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 = 43.700; pooled standard deviation σ = 67.901.
2. n = 2 × (1.96 + 0.842)² × 67.901² / 43.700² = 37.9, rounded up to 38 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 <- 43.7                      # observed difference in means
sigma <- 67.9012790748451                     # 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(2671760)
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 = 38 (80% power), 51 (90% power)
 75% of observed effect: n per arm = 68 (80% power), 91 (90% power)
 50% of observed effect: n per arm = 152 (80% power), 203 (90% power)
power.t.test: 38.9 
Simulated power at n = 38 per arm: 0.792 (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. Multiplicity, Missing Data and Analysis Populations

The registry record contains one primary endpoint and multiple secondary outcomes, with comparisons involving the comparator and placebo. It does not report a multiplicity adjustment, a hierarchy for testing endpoints, or a plan for controlling the overall type I error rate. In general, testing multiple outcomes or contrasts creates more opportunities to obtain a small p-value by chance. Unadjusted p-values therefore do not have the same confirmatory interpretation as results from a prespecified testing strategy that addresses multiplicity.

The record lists no reported analyses. It does not specify the model used for the crossover design, the handling of treatment sequence or period effects, the analysis population, or how missing endpoint measurements were handled. It also does not report whether any participants were excluded from particular analyses or how incomplete stays affected the summaries. These details are important because the arm-level means alone cannot establish how all randomized participants were included in the analysis.

Missing-data methods matter when outcomes are unavailable for some participants. Complete-case analysis, imputation, and model-based approaches rely on different assumptions, and those assumptions can affect estimates. No missing-data approach is reported in the registry record, so the effect of missing measurements cannot be assessed from the posted summaries.

The registry lists enrollment and arm-level outcome summaries, but the exact analysis population and the relationship between enrollment and each outcome summary are not described. This leaves uncertainty about whether the estimates represent all randomized participants or a subset with available measurements. The absence of these details does not show that data were missing or that an analysis was inappropriate; it means that the registry record does not report enough information to evaluate those possibilities.

15. 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 non-serious adverse event terms

The registry reports zero serious adverse events in each of the three arms, with 39 at risk in each arm. It also reports zero deaths in each arm. These are the serious-event counts recorded for the study; they do not establish that serious events cannot occur or that the conditions have identical safety profiles.

For other adverse events, the record lists 2 affected participants in the SM-1 arm, 3 in the comparator arm, and 3 in the placebo arm, with 39 at risk in each. The registry does not provide event terms in the posted summary, so the types, timing, severity, and relationship to study conditions are not described. The record also lists safety and tolerability in terms of residual sleepiness among the secondary outcomes, but no further interpretation should be inferred from the serious-event counts.

Safety counts from a small, short-term study have limited ability to characterize uncommon events or risks with longer exposure. The brief summary describes one-night stays, and the study is identified as single-dose. These results therefore describe events recorded under the study conditions, not a complete safety profile for other settings or durations.

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

The posted results describe mean sleep duration during an 8-hour assessment in adults with short-term insomnia participating in a randomized crossover study. The SM-1 arm had a higher mean total sleep time than the comparator arm, while the placebo arm had a lower mean than the comparator arm. The unadjusted recalculated confidence intervals for both primary contrasts do not include zero. Those findings describe the observed summaries and their uncertainty under the recalculation assumptions.

They do not establish that the observed differences are clinically important. The registry record does not report a threshold for a meaningful change in total sleep time. Nor do the results establish effects on every aspect of sleep, daytime function, quality of life, or longer-term outcomes. The experimental model and brief single-dose setting limit how far the findings can be extended beyond the assessed nights.

The numerical comparisons are not a substitute for the trial’s own statistical analysis. The record lists no reported analyses and does not describe how the crossover structure, treatment order, period effects, or within-participant correlation were handled. The estimates are unadjusted recalculations from posted summary statistics. Their confidence intervals and p-values should therefore be interpreted in that context.

Safety summaries show no serious adverse events or deaths recorded in the arms, alongside a small number of participants with other adverse events. These observations are limited by the study’s size and duration, and the event descriptions are not reported in the registry record.

Overall, the registry supports a cautious statistical reading of short-term outcome summaries, not a broad conclusion about treatment in routine care. It does not report enough information to establish the analysis plan, long-term effects, or comparative clinical importance.

17. Why This Trial Matters Statistically

This trial is statistically interesting because it combines a crossover design with a controlled, short-term sleep model and a continuous primary endpoint. Crossover studies can make efficient use of participant-level comparisons, but those comparisons depend on repeated measurements and design details. Without reported sequence information, paired data, or a stated model, arm-level summaries alone cannot show how the design was incorporated into analysis.

The registry also illustrates why effect estimates and uncertainty should be read together. The primary mean difference is accompanied by a confidence interval, which makes visible the range of values compatible with the unadjusted calculation. A p-value alone would not convey that range or the potential clinical relevance of the estimate. The registry does not report a clinically important threshold, so statistical evidence and clinical importance remain separate questions.

Multiple secondary outcomes create another interpretive challenge. Their results can help describe related sleep measures, but without a reported multiplicity plan, isolated unadjusted findings deserve less weight. The absence of reported analyses and missing-data details also limits assessment of whether the posted summaries reflect the intended crossover analysis.

Finally, the planning illustrations show how sample-size requirements depend on the assumed effect size and desired power. They can inform future planning, but cannot be used as observed power or as proof that the completed study was adequately designed. The original design assumptions are not reported in the registry record.

18. 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 most important trial-specific limitation for interpreting the numerical comparisons is that the study used a crossover model, while the registry record does not report the treatment sequence, washout arrangements, or statistical model. The posted comparisons are unadjusted recalculations from arm-level summary statistics and do not use the participant-level pairing needed to reconstruct a crossover analysis. This leaves uncertainty about whether the calculations reflect the structure of the trial.

The registry record lists no reported analyses and does not describe the analysis population, missing-data handling, or any adjustment for multiplicity. It also does not report a clinically important threshold for total sleep time. Those omissions limit evaluation of the analysis and of the practical meaning of the estimated differences.

The study assessed single-dose effects in a 5 hour phase advance model, with the primary endpoint measured over 8 hours. The brief summary describes three one-night stays. This controlled, short-term setting does not establish what results would be in routine sleep conditions, with repeated use, or among people with chronic insomnia. Enrollment was 39, and the duration and size of the study limit the amount of safety information it can provide.

The comparator is described as a 2-drug combination, but its ingredients are not reported in the registry record. The record also does not provide detailed event terms for the non-serious adverse events. These limitations concern what can be evaluated from the registry record; they do not show that unreported design or analysis steps were absent from the conduct of the study.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Pilot Study of a Combination Drug Product for Treatment of Short-term Insomnia — NCT02671760.
• Registry record last updated 2018-03-06; status last verified 2018-01.
• Results first posted 2018-03-06.
• 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.

22. Record Summary

NCT02671760 is a completed Phase 2 randomized, quadruple-masked crossover study in adults with short-term insomnia. It compared single-dose SM-1, a 2-drug comparator, and placebo in a 5 hour phase advance model. The primary endpoint was mean total sleep time in minutes over an 8-hour assessment.

The posted means were 382.9 minutes for SM-1, 339.2 minutes for the comparator, and 256.2 minutes for placebo. Unadjusted recalculated comparisons report a mean difference of 43.7 minutes for SM-1 versus comparator and -83.0 minutes for placebo versus comparator. The registry record lists no reported analyses and does not describe the crossover analysis or missing-data methods.

The record reports no serious adverse events or deaths in any arm. These results describe a short-term experimental study and do not establish long-term or routine-setting outcomes.