Safety and Efficacy Study of Ramelteon in Subjects With Chronic Insomnia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Study of Ramelteon in Subjects With Chronic Insomnia (NCT00756002), a randomized phase 3 study. The primary outcome is Mean Latency to Persistent Sleep Via Polysomnography (Nights 1-2)..

2. Clinical Question

3. Trial Design

The registry describes a randomized, quadruple-masked, parallel-group, placebo-controlled phase 3 study of ramelteon in adults with chronic insomnia. It reports an enrollment of 259 and lists Takeda as the sponsor. Study status is completed. The registry record gives a start of 2007-08 and primary completion of 2008-03.

The two assigned regimens were ramelteon 4 mg once daily and placebo once daily. The stated purpose was to determine the safety and efficacy of ramelteon. Polysomnography was used for the primary outcome, and the registry also lists subjective sleep-latency assessments from post-sleep questionnaires. This combination allows the reported outcome set to include an instrument-based sleep measure as well as participants’ reported estimates of how long it took them to fall asleep.

The primary analysis population was the Full Analysis Set. It consisted of randomized subjects who received at least one dose of double-blind study medication, and subjects were analyzed according to their randomized assignment. The reported primary arm counts were 128 for placebo and 130 for ramelteon. These analysis counts should not be treated as a reconciliation of the registry’s enrollment figure; the record does not explain the difference.

The registry does not report the original sample-size justification or the trial’s full statistical analysis plan. It identifies ANCOVA as the sponsor’s reported analysis method for the primary comparison, but does not provide the ANCOVA estimate or its confidence limits in the record. The trial’s allocation sequence details and any interim-analysis plan are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Sleep quality was assessed by postsleep questionnaires during a 21-day placebo run-in. Subjects were enrolled in Ramelteon or Placebo once-daily (QD) treatment group. 488 subjects entered placebo Run-in Period. 229 failed randomization criteria for entry into double-blind study medication phase of the study. 259 were randomized into the study.

Placebo Run-in

Double-Blind Treatment Period

Reasons for not completing this period:

Placebo Run-out

Analysis population for the primary outcome, as defined in the registry: The Full Analysis Set (FAS) population consisted of all subjects who were randomized and received at least 1 dose of double-blind study medication. Subjects were analyzed according to the treatment they were randomized to receive. The analysis was performed using the FAS and Last Observation Carried Forward (LOCF) data.

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 Mean Latency to Persistent Sleep Via Polysomnography (Nights 1-2)., measured over Nights 1-2.

The primary endpoint was mean latency to persistent sleep via polysomnography on Nights 1-2, measured in minutes. The registry defines latency as the elapsed time from the beginning of the polysomnography recording to the onset of the first 10 minutes of continuous sleep. The measured time was averaged over the two nights.

This endpoint represents the time until a specified period of continuous sleep begins, rather than a participant’s estimate of sleep onset. Polysomnography provides a recorded assessment of sleep, while the secondary outcomes include subjective latency reported through a post-sleep questionnaire. Those measures address related but not identical aspects of sleep onset, so their results should be interpreted on their own terms.

General statistical context: a continuous endpoint such as latency can be summarized by a group mean, and a difference in means describes the average separation between groups in the endpoint’s units. A lower mean latency represents a shorter average time to the endpoint definition. Whether a particular difference is important to patients depends on clinical context and a meaningful-difference threshold; such a threshold is not reported in the registry record.

The primary endpoint is a defined measurement at an early assessment period. It does not, by itself, establish how results may differ at later assessments or on subjective measures. The registry separately lists later polysomnography assessments and questionnaire outcomes.

7. Results Posted in the Registry

• Ramelteon 4 mg QD: mean 41.70 versus 67.00 on control; difference -25.30 (95% CI -33.21 to -17.39), standardized effect d = -0.78; 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.

Mean Latency to Persistent Sleep Via Polysomnography (Nights 15-16).

• 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 errors.
• Result:
  • Ramelteon 4 mg QD: mean 39.20 versus 65.30 on control; difference -26.10 (95% CI -34.60 to -17.60), standardized effect d = -0.75; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Ramelteon 4 mg QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Mean Latency to Persistent Sleep Via Polysomnography (Nights 29-30).

• 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 errors.
• Result:
  • Ramelteon 4 mg QD: mean 37.20 versus 60.50 on control; difference -23.30 (95% CI -31.39 to -15.21), standardized effect d = -0.70; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Ramelteon 4 mg QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Subjective Sleep Latency, Per Post-sleep Questionnaire (Nights 1-2).

• 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 errors.
• Result:
  • Ramelteon 4 mg QD: mean 61.50 versus 73.50 on control; difference -12.00 (95% CI -19.82 to -4.18), standardized effect d = -0.38; z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Ramelteon 4 mg QD than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary results include later polysomnographic latency assessments and subjective sleep latency measured by post-sleep questionnaire. The available secondary outcomes span early and later assessment periods, and the displayed results generally follow the direction of the primary finding: the reported ramelteon means are lower than the placebo means. This provides a consistent pattern across the posted objective and subjective measures, while leaving the primary endpoint as the main inferential anchor.

Secondary measures are related to the primary question but are not interchangeable with it. Polysomnography records the defined sleep-onset measure, whereas questionnaire outcomes rely on participant reports. Agreement in direction can add context, but it does not mean that every measure estimates the same effect or that the secondary results independently confirm the primary result.

The registry lists 45 outcome measures and 45 analyses, and 44 secondary outcomes are posted. The multiplicity approach is not reported in the registry record. General statistical context: examining many outcomes creates more opportunities for nominally notable results by chance when no adjustment is used. For that reason, unadjusted secondary p-values generally carry less evidential weight than a clearly identified primary comparison, particularly when the analysis plan and outcome hierarchy are unavailable.

The secondary pattern should therefore be read as supportive descriptive context rather than as a separate guarantee of efficacy or a resolution of limitations in the primary analysis. The registry does not report a clinical-importance threshold for these outcomes, and the posted endpoint list does not establish effects beyond the assessment periods measured.

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

• Mean Latency to Persistent Sleep Via Polysomnography (Nights 15-16).
• Mean Latency to Persistent Sleep Via Polysomnography (Nights 29-30).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Nights 1-2).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Nights 15-16).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Nights 29-30).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Week 2).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Week 4).
• Subjective Sleep Latency, Per Post-sleep Questionnaire (Week 5).

9. Statistical Interpretation

10. Statistical Methods Explained

The sponsor’s reported primary method was ANCOVA, with a reported p-value of <0.001. ANCOVA is a regression-based method that compares groups while accounting for specified covariates. General statistical context: adjusting for baseline or other prespecified variables can change the estimated comparison and its precision, depending on the model and the relationship between those variables and the outcome. The covariates used in this trial’s ANCOVA are not reported in the registry record, and the reported analysis does not include its treatment estimate or confidence limits.

The posted arm summaries are least-squares means, with standard errors. A least-squares mean is a model-based group mean rather than necessarily a simple arithmetic mean of all observed values. A standard error describes the uncertainty in an estimated mean; it is not the spread of individual participants’ measurements. The primary summaries give standard errors of 2.88 minutes for placebo and 2.83 minutes for ramelteon.

The recalculated comparison is an unadjusted difference between the posted means, with ramelteon minus placebo as the direction of subtraction. For the primary endpoint, that estimate is -25.3 minutes. Its reported confidence interval is -33.21 to -17.39, and the recalculated p-value is <0.001. The approximate standard deviation used for standardized-effect calculations is inferred from the reported standard errors and group sizes, rather than directly reported as a participant-level standard deviation. Such a derivation is useful for a simple summary-based comparison, but does not reproduce the sponsor’s adjusted ANCOVA.

General statistical context: a confidence interval describes a range of values compatible with the data and the specified method, subject to its assumptions. It expresses uncertainty about the estimated group contrast; it does not describe the range of individual responses. A p-value describes how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and the analysis assumptions. It is not the probability that the null hypothesis is true, and it does not measure clinical importance.

The standardized effect is a way to express a mean contrast relative to an estimate of outcome variability. It can help compare effects measured on different scales, but it depends on the variability estimate and assumptions behind it. The standardized effect shown for the primary comparison is -0.780. Its negative sign follows the ramelteon-minus-placebo direction; it is not a separate clinical outcome.

General statistical context: over-dispersion is especially relevant when outcomes are counts or binary events and the observed variability exceeds what a simple probability model expects. The primary endpoint is continuous, so that count-model issue is not the central concern. The more relevant distinctions are the sponsor’s adjusted model, the unadjusted summary-based comparison, and the assumption used to infer variability for recalculation. The registry does not provide enough model detail to assess all ANCOVA assumptions or independently reproduce its analysis.

The analysis population is described as the Full Analysis Set: randomized participants who received at least one dose of double-blind medication, analyzed according to assigned treatment. This is not a statement that every randomized participant contributed an observed primary outcome. The registry’s missing-data description is truncated after “Last Observation Carried,” and the complete method, including its implementation, is not reported in the registry record.

The posted results include many outcomes and analyses. General statistical context: when investigators test many outcomes, the chance of at least one apparently unusual result can increase if each is judged using an unadjusted threshold. An adjustment can address this concern, but the registry record does not report a multiplicity procedure. This makes the primary endpoint and its prespecified status important to interpretation, while secondary findings warrant more caution.

ANCOVA, confidence intervals, p-values and standardized effects convey different information. ANCOVA describes the sponsor’s modeling approach; the confidence interval conveys uncertainty around an estimated contrast; the p-value addresses compatibility with a null comparison under a model; and the standardized effect rescales a contrast relative to variability. None alone establishes clinical importance, and the summary-based recalculation should not be mistaken for the sponsor’s complete analysis.

11. Confidence Intervals and Precision

The recalculated primary mean difference is -25.3 minutes, with a 95% confidence interval from -33.21 to -17.39. With the contrast defined as ramelteon minus placebo, the interval corresponds to a lower mean latency in the ramelteon group throughout its reported range. Its limits are separated from zero, which is consistent with the sponsor’s reported ANCOVA p-value of <0.001. The confidence interval is based on the summary-statistic recalculation, not on a reported ANCOVA estimate.

General statistical context: a confidence interval conveys uncertainty about a group-level estimate under the method’s assumptions. A narrower interval indicates greater precision for that estimate; a wider interval indicates more uncertainty. The interval is not a range expected to contain most individual participants’ values, and it does not account for unreported sources of bias or model misspecification.

The arm means are accompanied by standard errors of 2.88 minutes for placebo and 2.83 minutes for ramelteon. These describe the estimated means’ precision, not the variability of individual sleep latencies. The recalculated interval reflects the estimated contrast and uncertainty derived from the posted summaries. The registry does not report an ANCOVA confidence interval, so the precision of the sponsor-adjusted estimate cannot be assessed from the reported inferential results alone.

Precision is distinct from importance. Even an estimate that is relatively precise does not establish whether the difference is meaningful to patients. A clinical interpretation threshold is not reported in the registry record. The interval also pertains to the primary measure and assessment period only; it does not guarantee similar precision or effects for other outcomes or later assessments.

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 illustrative calculations based on the observed effect, rather than a record of the trial’s original design calculation. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is set at the stated fraction of the observed effect.

The 80% power column means the illustrated design would have an 80% probability of detecting the specified effect under the planning assumptions, if that effect and the model assumptions were correct. The 90% power column expresses the corresponding planning target at 90% probability. These are prospective design quantities, not evidence that the completed study had a particular probability of finding its result. The figures are not observed power.

Each row changes the assumed effect relative to the observed effect. Planning for a smaller effect requires more participants per arm to reach the same target power, because a smaller group difference is harder to distinguish from variation in the outcome. The displayed scenarios therefore show how sensitive a planning estimate can be to the effect size selected.

The registry’s own original design assumptions, including its sample-size justification, are not reported in the registry record. These illustrations should not be used to infer what the sponsor planned or whether the original study was adequately designed. They are calculations tied to the observed summary effect and the stated general assumptions, not a post hoc measure of the trial’s success.

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 = -25.300; pooled standard deviation σ = 32.424.
2. n = 2 × (1.96 + 0.842)² × 32.424² / -25.300² = 25.8, rounded up to 26 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 <- -25.3                      # observed difference in means
sigma <- 32.4243723074364                     # 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(756002)
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 = 26 (80% power), 35 (90% power)
 75% of observed effect: n per arm = 46 (80% power), 62 (90% power)
 50% of observed effect: n per arm = 104 (80% power), 139 (90% power)
power.t.test: 26.8 
Simulated power at n = 26 per arm: 0.789 (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 sponsor’s reported primary analysis used ANCOVA and reported p < 0.001 for the comparison between placebo and ramelteon. The registry does not report the ANCOVA treatment estimate or confidence limits. The sponsor’s p-value supports a statistically significant difference under that analysis, but it does not give the size or precision of the adjusted contrast on its own.

The recalculated result uses the posted least-squares means and standard errors to form an unadjusted mean difference. It gives an estimate of -25.3 minutes and a 95% confidence interval from -33.21 to -17.39. The recalculated p-value is <0.001. The direction and broad statistical conclusion are consistent with the sponsor’s reported result.

The recalculation is not a replication of ANCOVA. It does not apply the sponsor’s model or any covariate adjustment, and the registry does not give the model details needed to reproduce that analysis. Its confidence interval is therefore the interval for the summary-based unadjusted comparison, not a substitute for a missing ANCOVA interval. The registry record reports no discrepancy between the sponsor’s result and the recalculation. The sponsor analysis remains the trial’s own reported analysis; the recalculation is an additional, simpler description based on posted summaries.

15. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary endpoint and a large set of additional outcomes: 45 outcome measures and 45 analyses, with 44 secondary outcomes posted. It reports ANCOVA analyses for the primary and the displayed secondary comparisons. The registry record does not report whether the analyses were organized into a prespecified hierarchy, whether any multiplicity adjustment was used, or how interpretation of the secondary outcomes was controlled for multiple testing.

General statistical context: testing many outcomes can increase the chance that one or more results appear statistically notable even when there is no real group difference. Multiplicity procedures are methods for controlling that risk, but they can also affect the interpretation and precision of individual comparisons. Without a reported procedure or analysis hierarchy, secondary p-values should be treated as less conclusive than the primary result. This does not establish that an adjustment was or was not used; the procedure is not reported in the registry record.

The registry describes the primary analysis population as the Full Analysis Set, including randomized participants who received at least one dose of double-blind study medication and were analyzed according to randomized assignment. It also mentions “Last Observation Carried” in its description of the analysis, but the wording ends there. The complete missing-data method, including what was carried forward and how missing measurements were handled, is not reported in the registry record.

Details of the original sample-size justification, the complete statistical analysis plan, and interim analyses are also not reported in the registry record. These omissions limit assessment of whether outcome selection, model specification, missing-data handling and multiplicity methods were determined before results were examined. They do not, by themselves, show that the methods were inappropriate; they define what cannot be established from the registry information.

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 0.4% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports no serious adverse events among the listed participants at risk: 0 of 129 in the placebo group and 0 of 130 in the ramelteon group. It reports other adverse events for 34 of 129 participants at risk with placebo and 35 of 130 at risk with ramelteon. These are the posted counts and denominators; the registry record does not provide additional detail about the events in this summary.

These figures should be read as descriptive safety counts, not as proof that the treatments have identical safety profiles. The record does not provide event terms, severity or causality details for the other adverse events in the information. Death figures are not reported in the registry record. The safety denominators also do not match every efficacy analysis count, and the registry does not explain that difference. No conclusion about rare or longer-term events can be drawn from these counts alone.

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

The posted primary result describes a lower average latency to persistent sleep in the ramelteon group than in the placebo group during Nights 1-2. The recalculated difference is -25.3 minutes, with a 95% confidence interval from -33.21 to -17.39, and the sponsor’s ANCOVA comparison reports p < 0.001. These statistics describe the observed group-level comparison and its uncertainty under the respective analyses.

They do not show that every participant experienced a shorter latency, nor do they describe the range of individual responses. A group mean can coexist with substantial variation among participants. The standard errors reported for the means describe the precision of those estimated means, not individual variability.

The result does not establish clinical importance. The registry record does not report a threshold for a meaningful difference in sleep latency, so statistical significance should not be treated as a substitute for such a threshold. Nor does the primary endpoint address every aspect of chronic insomnia: it measures a specific polysomnographic definition of sleep onset at a specific assessment period.

The registry’s later objective and subjective outcomes provide additional observations, but they remain secondary outcomes with a multiplicity procedure not reported in the record. The trial information also does not establish how the result would apply to people outside the described adult chronic-insomnia population, to different regimens, or to periods beyond those assessed.

General statistical context: a randomized comparison can help estimate differences between assigned groups, but interpretation still depends on analysis choices, missing-data handling, measurement quality and the completeness of reporting. The available record does not resolve every one of those issues. The result is best understood as evidence about the stated primary measure under the reported trial comparison, not as a universal claim about treatment response or safety.

18. Why This Trial Matters Statistically

This trial illustrates why an endpoint’s definition matters. “Latency to persistent sleep” is not an informal judgment of sleep quality: the registry specifies a polysomnographic starting point and a sustained-sleep criterion, then averages measurements across the stated nights. That precision makes the outcome interpretable while also limiting the claim to what the measure captures.

It also illustrates the distinction between the sponsor’s analysis and a summary-based recalculation. The sponsor reports ANCOVA and a p-value, while the posted arm summaries support an unadjusted estimate and interval. Agreement in direction is informative, but the analyses are not identical. Keeping those methods distinct prevents a recalculation from being mistaken for the trial’s adjusted result.

Finally, the record places the primary result among many secondary outcomes and supplies limited detail on multiplicity and missing data. Those reporting gaps matter because statistical evidence depends not only on estimates and p-values, but also on how outcomes were selected, modeled and handled. The posted results support a clear description of the primary comparison, while careful interpretation requires acknowledging what the registry does not report.

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.

The registry record does not provide the covariates or full model specification for the sponsor’s ANCOVA. It reports the primary p-value but not the model-based treatment estimate or its confidence interval. Consequently, the adjusted effect size and precision cannot be evaluated from the reported ANCOVA result alone. The summary-based recalculation is unadjusted and uses an approximate variability estimate, so it cannot fill that reporting gap.

The record’s description of missing-data handling ends after “Last Observation Carried.” The complete procedure is not reported, and the impact of missing outcomes cannot be assessed from the available description. The registry also does not report the original sample-size justification, any interim-analysis plan, or a multiplicity procedure. With many listed outcomes, the absence of a reported multiplicity approach limits how strongly secondary p-values can be interpreted.

The registry does not report a clinical-importance threshold for the primary latency measure or the subjective secondary measures. The primary outcome is restricted to the stated polysomnographic measure and Nights 1-2; later and subjective outcomes are distinct assessments. The record’s summary also does not provide event-level detail for the other adverse events, and death figures are not reported in the registry record.

The enrollment figure and analysis counts are not fully reconciled in the registry information. The efficacy summaries report group counts of 128 and 130, while the safety summary lists participants at risk of 129 and 130 for the groups. The record does not explain the difference, so the figures should be kept attached to their respective analyses rather than combined or interpreted as a single denominator.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Safety and Efficacy Study of Ramelteon in Subjects With Chronic Insomnia — NCT00756002.
• Registry record last updated 2010-06-02; status last verified 2010-05.
• Results first posted 2009-11-05.
• 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 describes a completed, randomized, quadruple-masked, parallel-group phase 3 trial of ramelteon 4 mg once daily versus placebo once daily in adults with chronic insomnia. Enrollment was 259. The primary endpoint was mean polysomnographic latency to persistent sleep on Nights 1-2.

The posted primary means were 67.0 minutes for placebo and 41.7 minutes for ramelteon. The recalculated unadjusted mean difference was -25.3 minutes, with a 95% confidence interval from -33.21 to -17.39. The sponsor reported an ANCOVA comparison with p < 0.001. The record also reports later polysomnographic and subjective sleep-latency outcomes, but multiplicity and complete missing-data methods are not reported in the registry record.