Study of the Effects of HIRREM-SOP for Insomnia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of the Effects of HIRREM-SOP for Insomnia (NCT03607994), a randomized na study. The primary outcome is Change in ISI Score From V1 to V3.

2. Clinical Question

The registry identifies the study as a randomized controlled pilot trial. The primary outcome is a continuous score, and the posted comparison describes a difference between the arms’ mean changes. The central statistical issue is not simply whether each group’s score moved, but how different the changes were between groups and how precisely that between-group difference was estimated. The record does not report a separate clinical threshold for an important change in ISI score, so the statistical estimate cannot by itself establish how meaningful a difference would be to participants.

3. Trial Design

This was a randomized, triple-masked, parallel-group pilot trial registered as NCT03607994. The record lists the study as completed, with enrollment of 22. The start date was 2018-11-19 and the primary completion date was 2020-03-05. The phase is listed as not applicable. Wake Forest University Health Sciences is named as sponsor.

Participants were assigned to HIRREM-SOP (BCC) or nonspecific acoustic stimulation (NCC). Both arms continued current care. The distinction described in the registry is whether the acoustic stimulation was linked to brainwave activity: it was in HIRREM-SOP and was not in NCC. Since both arms received acoustic stimulation and continued current care, the comparison evaluates the additional features of the HIRREM-SOP approach against an active acoustic-stimulation comparator.

The record reports 10 participants in the HIRREM-SOP arm and 12 in the NCC arm for the primary outcome. Those arm counts describe the outcome record, but the registry excerpt does not explain participant flow, withdrawals, or whether every randomized participant contributed data at each visit. It also does not detail how the triple masking was implemented, who was masked, or whether masking success was evaluated. These details are not reported in the registry record.

The primary endpoint was change in ISI score from V1 to V3. V1 was baseline; V2 was 0-14 days after intervention completion; and V3 was 4-6 weeks following V2. The registry also lists a secondary outcome concerning the number of participants who believed they were receiving HIRREM-SOP. It lists 2 outcome measures and 0 analyses. The method used to estimate the primary comparison, any adjustment for baseline values, and the trial’s original sample-size justification are not reported in the registry record. As a pilot study, these results are best read as a small-sample estimate with substantial uncertainty, not as a definitive assessment of the intervention.

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 Change in ISI Score From V1 to V3, measured over Baseline, V2 (0-14 days after intervention completion), V3 (4-6 weeks following V2).

The primary endpoint is change in Insomnia Severity Index score from baseline (V1) to V3. The ISI is a 7-question measure. Each response ranges from 0-4, yielding a total score from 0-28. Higher scores indicate greater insomnia severity. Change is useful because it expresses how a participant’s measured score differs over the study period, rather than describing only the score at one visit. A negative change therefore indicates a decrease in the score; in this measure, that corresponds to a decrease in the measured severity of insomnia symptoms.

The registry specifies that the endpoint is summarized as a mean, with standard error as the measure of dispersion. The posted mean change was -7.27 in the HIRREM-SOP arm and -4.69 in the NCC arm. Both means are decreases from baseline, but the primary between-arm question concerns the contrast between those changes. The registry gives a mean difference of 2.58 for NCC relative to HIRREM-SOP. The direction follows from the stated arm means: a positive difference under that ordering indicates a larger decrease in the HIRREM-SOP arm.

A repeated symptom measure can capture change over time, but its interpretation depends on when measurements are taken and how missing visits are handled. The timing is described in the registry, while the missing-data approach and any threshold for a clinically important difference are not reported in the registry record. The endpoint can therefore describe the observed score changes, but the record alone does not establish how those changes should be judged in clinical terms.

7. Results Posted in the Registry

• Nonspecific Acoustic Stimulation (NCC): mean -4.69 versus -7.27 on control; difference 2.58 (95% CI -1.51 to 6.67), standardized effect d = 0.53; z-test p = 0.217.

8. Statistical Interpretation

9. Statistical Methods Explained

A continuous endpoint is measured on a numeric scale where differences between values carry meaning. ISI score is treated this way in the registry. The primary outcome is a mean change, so each participant’s score change contributes to an arm-level average. A mean offers a compact description of the group’s average change, but it does not show how each participant responded or whether responses varied in direction and size.

What does the mean difference represent?

The comparison is reported as a mean difference of 2.58, with NCC as the comparison arm relative to HIRREM-SOP. This ordering is consistent with the posted mean changes: -7.27 in HIRREM-SOP and -4.69 in NCC. Because more negative change indicates a larger decline in ISI score, a positive NCC-minus-HIRREM-SOP contrast corresponds to a larger average decline in HIRREM-SOP. The contrast is a between-arm estimate, not an estimate of change within either arm alone.

Random assignment is intended to make the arms comparable on average at the point of assignment, including characteristics that may not be measured. In a small pilot, however, randomization does not guarantee that every characteristic is balanced in the realized groups. The registry does not report baseline characteristics or an adjusted model. It is therefore not possible from the record to assess baseline balance or determine whether the comparison adjusted for baseline ISI score or other variables.

How should standard errors be understood?

The registry reports standard errors alongside the arm means. A standard error describes the statistical uncertainty in an estimate of a mean, rather than the spread of individual participant scores. General explanation: a smaller standard error indicates greater precision in estimating the mean, while a larger standard error indicates more sampling uncertainty. It should not be read as a range containing most participants’ scores. The arm-level standard errors are 1.55 for HIRREM-SOP and 1.4 for NCC. The record does not provide arm-level confidence intervals for these means.

The comparative confidence interval is more directly useful for judging uncertainty in the primary between-arm estimate. It runs from -1.514 to 6.674 around the mean difference of 2.58. General explanation: a confidence interval describes a range of values compatible with the data under the statistical method used, and narrower intervals indicate more precision. This interval includes zero, the value representing no difference. Its width reflects substantial uncertainty about the underlying contrast. The registry does not identify the method used to construct this interval, so a more specific account of the inferential procedure is not possible.

Why does a small pilot matter for precision?

With relatively few participants contributing to the outcome, a group mean and the difference between means can shift appreciably with the observations in the sample. This general feature of small studies is consistent with the broad interval in the posted comparison. A point estimate is still informative as an estimate from this trial, but it should not be mistaken for a fixed or highly precise value of the effect in a wider population. The record does not report a formal sample-size justification for the trial or state a target precision.

General explanation: the standard error of a mean depends on both the variability in observations and the number of observations. Greater variability or fewer observations tends to increase uncertainty. For counts or other outcomes whose variability exceeds what a simple model assumes, over-dispersion can make standard errors too small if it is ignored. The primary outcome is a continuous score, and the registry does not report an over-dispersion assessment or a model for the score. That concept is not a claim about an observed issue in this trial.

What does the p-value add?

The unadjusted p-value is 0.217. General explanation: a p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the model assumptions were true. It is not the probability that the null hypothesis is true, nor does it measure the probability that the result occurred by chance. It does not give the size or clinical importance of a difference. The estimate and confidence interval are therefore essential to interpretation alongside the p-value.

The word “unadjusted” indicates that the reported p-value is not described as adjusted for multiple comparisons. The registry lists 2 outcome measures but 0 analyses, and it does not describe a multiplicity procedure or an analysis plan. No further claim about which outcomes or tests were prespecified can be made from the registry record. The sponsor’s own analysis takes precedence when a discrepancy is flagged, but the record does not flag a discrepancy for these results.

Finally, the analysis population and missing-data method matter for a change endpoint requiring measurements at multiple visits. The registry supplies visit timing but does not state how incomplete follow-up was handled, whether participants were analyzed according to randomized assignment, or whether a model used all available visits. Those choices can affect both the estimate and its uncertainty. In the absence of that detail, the posted summary statistics can be described, but the analysis cannot be fully reproduced or audited from the registry record.

10. Confidence Intervals and Precision

The comparative estimate is a mean difference of 2.58 points, with a confidence interval from -1.514 to 6.674. This interval is wide relative to the point estimate and crosses zero. Accordingly, the result is compatible with no average difference between arms, as well as with differences in either direction. It also accommodates a range of values where the average decrease is larger in HIRREM-SOP, given the comparison’s stated ordering. The interval is more informative about uncertainty than the point estimate alone.

General explanation: a confidence interval is an uncertainty range for an estimated quantity, based on a statistical method and its assumptions. A confidence interval does not mean that there is a specified probability that the fixed true effect lies inside this particular interval. It communicates how much uncertainty the analysis leaves about plausible values for the effect. A narrower interval would provide greater precision; the present interval leaves substantial uncertainty about the comparative change.

The p-value is 0.217 and is reported as unadjusted. It is consistent with the interval crossing zero, but it should not replace interpretation of the estimate and interval. The confidence interval’s endpoints include a negative difference, zero, and a positive difference. Thus the posted data do not rule out no difference or a difference in the opposite direction from the point estimate. Nor does the interval establish that all values within it are equally likely or clinically equivalent.

The registry does not give a threshold for a clinically important difference on the ISI. Statistical precision and clinical importance are separate questions: the interval describes uncertainty in the comparative estimate, while a clinical interpretation would also require a justified threshold and relevant context. The registry also does not state the method used to construct the comparison interval, so interpretation should stay with the reported range rather than assume details of a particular model.

11. 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 illustration uses participants per arm, not total event counts. Total-event planning applies to time-to-event outcomes, which are not the primary outcome in this trial. The stated 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.

At 100% of the observed effect, the planning illustration gives 57 participants per arm for 80% power and 75 participants per arm for 90% power. At 75% of the observed effect, it gives 100 participants per arm for 80% power and 134 participants per arm for 90% power. At 50% of the observed effect, it gives 225 participants per arm for 80% power and 300 participants per arm for 90% power.

In plain language, the 80% column gives the per-arm sample size associated with an 80% chance of detecting the specified effect under the stated assumptions if that effect is real. The 90% column gives the corresponding per-arm sample size for a higher chance of detection. These are prospective planning illustrations, not observed power for the completed trial. Smaller effects need more participants to detect under the same assumptions, as reflected by the increasing per-arm figures when the assumed effect fraction is reduced.

The illustration uses the observed effect as its starting point. That estimate comes from a small pilot and has a confidence interval from -1.514 to 6.674, so it is uncertain. Planning from a single observed estimate can be misleading if it differs from the effect a future study should target. The figures do not establish the appropriate sample size for another trial; that would require a justified target effect and a prespecified design. No table values should be interpreted as the number who actually participated in this study.

12. 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 = 2.580; pooled standard deviation σ = 4.873.
2. n = 2 × (1.96 + 0.842)² × 4.873² / 2.580² = 56.0, rounded up to 57 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 <- 2.58                      # observed difference in means
sigma <- 4.87311502018986                     # 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(3607994)
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 = 57 (80% power), 75 (90% power)
 75% of observed effect: n per arm = 100 (80% power), 134 (90% power)
 50% of observed effect: n per arm = 225 (80% power), 300 (90% power)
power.t.test: 57 
Simulated power at n = 57 per arm: 0.799 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary endpoint as change in ISI score from V1 to V3. It also lists a secondary outcome concerning the number of participants who believe they are receiving HIRREM-SOP. There are 2 outcome measures listed and 0 analyses listed. The record does not identify a prespecified hierarchy of outcomes, state whether the listed secondary outcome was analyzed, or describe any adjustment for multiple testing.

Multiplicity arises when several outcomes, time points, comparisons, or analyses are considered. General explanation: evaluating multiple questions can increase the chance that at least one result appears noteworthy by chance, so a prespecified analysis plan may use an adjustment or a structured hierarchy. The registry’s unadjusted p-value does not indicate whether any such procedure was applied elsewhere. No multiplicity adjustment is reported in the registry record, and its absence from this record should not be treated as proof that none was used in the full study.

The registry identifies the primary comparison and reports arm means and a mean difference, but does not describe the statistical model, test, or covariates. It does not report whether the analysis was intention-to-treat, per-protocol, or based on another analysis population. It also does not state whether all randomized participants were included in the primary analysis. The outcome arm counts are 10 and 12; the record does not explain their relation to participant flow or missing visits.

These details are important because the primary endpoint depends on repeated measurements. The record gives the scheduled timing of V1, V2, and V3, but does not explain how missing V3 scores, incomplete visits, or withdrawals were handled. It does not state whether any imputation was used, whether complete cases were analyzed, or whether a model incorporated available observations from different visits. Those choices may influence the estimate and its uncertainty, and cannot be reconstructed from the posted summary values.

The registry also does not report a formal statistical analysis plan, interim analyses, subgroup analyses, or sensitivity analyses. Those elements are not reported in the registry record. Consequently, the available results support description of the posted outcome and comparison, but not a complete assessment of analysis choices, multiplicity control, or robustness to missing data.

14. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 10 participants at risk in the HIRREM-SOP arm and 12 in the NCC arm. There were 0 serious adverse events in each arm. Deaths were also 0 in each arm. The record lists no serious adverse event terms.

For other, non-serious adverse events, the reported counts are 1 among 10 participants at risk in HIRREM-SOP and 0 among 12 participants at risk in NCC. These are counts as recorded, not evidence that every possible adverse event was captured or that the groups had identical safety experience. The record does not describe event definitions, severity grading, collection procedures, duration of safety follow-up, or whether one participant could contribute more than one event.

Zero recorded serious events in this small trial means no serious events were reported in the registry for the listed participants at risk. It does not demonstrate that serious events cannot occur or establish comparative safety. Likewise, the single other event in HIRREM-SOP and none in NCC are descriptive registry counts; they do not by themselves establish a causal relationship to an intervention. The types of other adverse events are not reported in the registry record.

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

The posted outcome data show mean decreases in ISI score in both arms: -7.27 for HIRREM-SOP and -4.69 for NCC. The reported mean difference is 2.58, with NCC relative to HIRREM-SOP, so the point estimate favors a larger average decrease in the HIRREM-SOP arm. The confidence interval, -1.514 to 6.674, is broad and includes zero. The unadjusted p-value is 0.217. These statistics do not establish a precise difference between the arms.

The result does not show that the groups were equivalent. A confidence interval spanning zero can be compatible with no difference, but it can also include differences in either direction. Equivalence or non-inferiority would require a prespecified margin and an analysis designed for that question; neither is reported in the registry record. Similarly, the p-value is not the probability that the intervention is ineffective, nor is it a measure of clinical importance.

The within-arm decreases should not be confused with evidence of a between-arm effect. Participants in both arms received acoustic stimulation and continued current care. Without a no-stimulation or no-intervention comparison, the results do not isolate change attributable to receiving either intervention rather than other influences over time. The randomized comparison addresses the difference between the two approaches as described, subject to the limitations of the posted analysis details.

The registry does not report a clinically important ISI threshold, baseline participant characteristics, analysis population, missing-data methods, or full statistical plan. It is therefore not possible to determine from this record alone how the observed difference relates to a threshold meaningful to participants, or how sensitive it is to analysis choices. The small pilot enrollment and the interval’s width reinforce that the estimate should be treated as uncertain.

These results describe the registry record for this trial. They do not establish that HIRREM-SOP works or does not work, and they do not support a definitive conclusion about comparative effectiveness. They identify a point estimate and uncertainty that could inform the design of further research, while leaving substantial uncertainty about the true between-arm difference.

16. Why This Trial Matters Statistically

This trial is statistically useful as an example of why an estimated effect should be read together with its precision. The point estimate of 2.58 is easy to focus on, but the confidence interval from -1.514 to 6.674 shows that the estimate is uncertain and compatible with no difference. A single number cannot convey this range of plausible values. The interval prevents the mean difference from being mistaken for a precise result.

The two arm means also illustrate the distinction between change within groups and contrast between groups. Both mean changes are negative, but the question in a randomized comparison is whether the amount of change differs between the arms. The reported contrast, rather than the two separate means alone, addresses that question. This distinction applies broadly to trials where an outcome is measured before and after intervention.

The active comparator matters to interpretation. NCC included nonspecific acoustic stimulation and continued current care, while HIRREM-SOP included acoustic stimulation linked to brainwave activity and continued current care. The observed contrast therefore concerns these differing approaches as described in the registry. It should not be read as a comparison with no treatment or with no acoustic stimulation.

Finally, a pilot result can be informative while remaining inconclusive. The planning illustration makes clear that proposed sample sizes depend on the effect selected for detection and the desired power, and that smaller target effects require more participants under the stated assumptions. Yet the observed effect itself is uncertain, and the trial’s original sample-size justification is not reported. This is a reminder that planning calculations are conditional illustrations, while the confidence interval describes uncertainty in the completed trial’s estimate.

17. 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 enrolled total was 22, and the primary outcome counts were 10 for HIRREM-SOP and 12 for NCC. This is a small evidence base for estimating a between-arm difference, and the confidence interval is correspondingly broad. The trial is identified as a pilot, but the registry does not report the pilot’s aims beyond determining effects, or whether its purpose was feasibility, preliminary estimation, or another objective.

The registry does not give participant demographics, baseline ISI values, or the distribution of baseline severity among enrolled participants. It therefore does not allow assessment of how representative the sample was of adults with insomnia or whether the arms began with comparable symptom scores. Eligibility is described in terms of documented sleep trouble within subthreshold, moderate, or severe clinical insomnia ranges, but detailed recruitment and screening information are not reported.

Although masking is listed as triple, the record does not identify who was masked, how assignment was concealed, or whether participants or investigators correctly guessed assignments. The secondary outcome about participant beliefs indicates that treatment belief was of interest, but the registry does not provide its result. The masking label alone does not resolve the possibility of unmasking or expectation effects.

The statistical and follow-up details are also limited. The analysis model, analysis population, missing-data handling, baseline adjustment, and multiplicity plan are not reported in the registry record. The posted endpoint requires measurements over time, but participant flow and completion at each visit are not described. No clinically important ISI difference is stated. These limitations prevent a full assessment of analysis robustness, representativeness, and clinical interpretation.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Study of the Effects of HIRREM-SOP for Insomnia — NCT03607994.
• Registry record last updated 2022-12-01; status last verified 2022-05.
• Results first posted 2021-06-22.
• 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.

21. Record Summary

NCT03607994 is a completed, randomized, triple-masked, parallel-group pilot trial sponsored by Wake Forest University Health Sciences. The registry lists enrollment of 22 and compares HIRREM-SOP, with acoustic stimulation linked to brainwave activity, against nonspecific acoustic stimulation not linked to brainwave activity. Both groups continued current care.

The primary endpoint was change in ISI score from baseline to V3. Mean changes were -7.27 for HIRREM-SOP and -4.69 for NCC. The reported mean difference was 2.58, with a confidence interval from -1.514 to 6.674 and an unadjusted p-value of 0.217. The interval includes zero and leaves substantial uncertainty about the between-arm difference.

The registry reports 0 serious adverse events and 0 deaths in both arms, with 1 other adverse event in HIRREM-SOP and 0 in NCC. Important analysis details, including missing-data handling and the trial’s original sample-size justification, are not reported in the registry record.