The Modulatory Effect of Oxytocin on Interoception: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Modulatory Effect of Oxytocin on Interoception (NCT03074409), a randomized na study. The primary outcome is Heartbeat Detection (Interoception) Accuracy.

2. Clinical Question

3. Trial Design

The trial record identifies a randomized, double-masked, parallel-group study. Participants were assigned to an oxytocin group or a placebo group. The registry names the sponsor as the University of Electronic Science and Technology of China. The trial is marked completed, and the phase is recorded as not applicable.

The registry lists 83 enrolled participants. For the primary outcome, the record states that 8 subjects were excluded because of low-quality electrocardiogram data or failure of behavioral data acquisition. The posted outcome groups include 37 participants receiving oxytocin and 38 receiving placebo. The record does not explain how participants were randomized, whether allocation was balanced by design, or who was masked. The designation “double” is recorded, but the specific masked roles are not described.

The study began on 2017-03-02, with primary completion on 2017-09-06. The primary outcome was measured 45 minutes after oxytocin or placebo treatment. The registry does not report the original sample-size justification, intended recruitment target beyond the enrollment figure, treatment dose, or further procedural details. It also does not report whether there were interim analyses or stopping rules. The record lists 2 outcome measures but has no posted secondary outcome results. The study description frames the research around oxytocin, interoception, and emotional valence; however, the posted primary-result comparison concerns heartbeat-detection accuracy, and no analysis of emotional valence is listed.

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

Analysis population for the primary outcome, as defined in the registry: 83 subjects were recruited and 8 subjects were excluded due to low quality of electrocardiogram data (6 subjects) or failure of behavioral data acquisition (2 subjects).

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.

Population: 83 subjects were recruited and 8 subjects were excluded due to low quality of electrocardiogram data (6 subjects) or failure of behavioral data acquisition (2 subjects).

6. Primary Endpoint

The registry lists the primary outcome as Heartbeat Detection (Interoception) Accuracy, measured over 45 minutes after oxytocin/placebo treatment.

The primary endpoint is heartbeat-detection accuracy, described as a ratio of correct heartbeat counting. It is a continuous measure, and the registry reports group means with standard errors. The measure is calculated across blocks of a heartbeat-detection task using the difference between recorded and counted heartbeats relative to recorded heartbeats. A value closer to accurate counting corresponds to less discrepancy between recorded and counted heartbeats under the stated formula.

Heartbeat detection is used in this study as an operational measure of interoception: the task assesses how accurately participants count their heartbeats. The outcome is recorded 45 minutes after treatment. It therefore captures task performance at the specified post-treatment assessment, rather than a broader measure of interoception or a clinical outcome. The registry does not report additional assessment times or a validated threshold for interpreting the ratio.

The endpoint description states that the calculation averages across the blocks completed for the task. Electrocardiogram quality and behavioral-data acquisition affected inclusion in the posted primary-outcome population: the record reports 6 exclusions for low-quality electrocardiogram data and 2 for acquisition failure. The registry does not provide block-level observations, individual scores, or further details on handling incomplete blocks. Consequently, the reported arm means describe the posted endpoint summaries, but the record does not allow reconstruction of participant-level accuracy or evaluation of variation across blocks.

7. Results Posted in the Registry

• Oxytocin: mean 0.64 versus 0.73 on control; difference -0.10 (95% CI -0.19 to -0.00), standardized effect d = -0.46; z-test p = 0.045.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry presents a continuous outcome as a mean for each arm, accompanied by a standard error. It also presents a mean difference, a confidence interval, an unadjusted p-value, and a standardized effect. These statistics answer related but distinct questions. The means describe the average observed accuracy in each group. The mean difference describes the estimated contrast on the endpoint’s original scale. The confidence interval conveys uncertainty around that contrast, while the p-value describes how compatible the observed data are with a specified null comparison under the assumptions of the test.

What does the mean difference describe?

The reported contrast is the oxytocin-group mean minus the placebo-group mean: -0.098. It is expressed in the heartbeat-detection accuracy measure, rather than in standardized units. The negative direction indicates that the oxytocin group’s posted mean was lower. This is a comparison of group summaries, not a description of every participant’s response. Individual results can vary around their group mean, and the registry does not provide participant-level data to show that distribution.

The arm means are 0.637 for oxytocin and 0.735 for placebo. Their reported standard errors are 0.035 and 0.034, respectively. A standard error represents the sampling uncertainty of an estimated mean under the statistical model; it is not the same as the standard deviation of participant observations. The comparison separately reports a pooled standard deviation of 0.211, which summarizes variability for the standardized contrast. The registry does not provide additional detail on how this pooled value was estimated.

What is the role of the confidence interval?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. It is not a range containing a stated proportion of individual participants, and it does not assign a probability to the true effect being within that particular interval. The reported interval for the mean difference runs from -0.194 to -0.002. Its direction is consistent with the negative point estimate, while its upper endpoint is close to zero. The precision and implications of that interval are discussed separately in the interpretation of uncertainty.

How should the p-value be understood?

The registry reports an unadjusted p-value of 0.045. A p-value is calculated under a null hypothesis and the assumptions of the analysis; it is not the probability that the null hypothesis is true, and it does not measure the size or importance of a difference. “Unadjusted” means the posted value is not identified as adjusted for multiple comparisons. The registry does not report a multiplicity procedure, an analysis plan, or details sufficient to determine whether this comparison was selected in advance. The p-value therefore needs to be read as the reported unadjusted result, not as proof of a general treatment effect.

The planning assumptions elsewhere in the record use a two-sided significance level of 0.05, but this is an assumption for a planning illustration and is not a substitute for missing details of the trial’s own statistical plan. The original design assumptions and any planned testing hierarchy are not reported in the registry record.

What does the standardized effect add?

The record reports d as -0.464. A standardized effect expresses a difference relative to variability, allowing a contrast to be considered without relying only on the original measurement scale. It does not provide a clinical interpretation by itself. Its value depends on the definition of the variability measure and the analysis conventions used. Although a pooled standard deviation is also posted, the registry does not give a fuller account of the standardized-effect calculation. The standardized statistic should therefore complement, not replace, the original-scale mean difference and its interval.

What can be inferred from the design and posted summaries?

Randomization and a parallel-group design describe how the trial was structured, while the double-masked designation indicates that masking was recorded. These design features do not resolve missing details about how randomization was implemented, who was masked, or how exclusions and incomplete data were handled. The registry describes 83 recruited subjects and 8 exclusions associated with data quality or acquisition. It does not report a detailed analysis-population definition, imputation method, sensitivity analysis, or participant flow beyond those statements.

The reported primary-outcome sample sizes are 37 and 38. The adverse-event record lists different at-risk counts, 41 and 42, and should be read as a separate set of safety denominators. The registry gives no participant-level data or model diagnostics, so the reported summaries cannot be independently checked for distributional assumptions, outliers, or the influence of individual observations. The result is best understood as the posted comparison and its stated uncertainty, with unreported analytic details left unresolved.

10. Confidence Intervals and Precision

The estimated mean difference is -0.098, and its 95% confidence interval extends from -0.194 to -0.002. The interval describes uncertainty in the estimated group contrast under the analysis method used. It is not an interval for individual heartbeat-detection scores. Since both endpoints are negative, the reported interval is on the same side of zero as the point estimate; the upper endpoint, however, is close to zero. The interval therefore permits contrasts ranging from more negative values to a small negative value, rather than identifying a single precise magnitude.

The interval is useful alongside the unadjusted p-value of 0.045 because it gives information about the direction and range of estimates compatible with the calculation, while the p-value alone does not show that range. Neither statistic establishes the practical importance of the difference. The registry does not report a threshold for a meaningful change in this accuracy ratio, so the clinical or behavioral importance of values within the interval cannot be determined from the record.

The group means are accompanied by standard errors of 0.035 for oxytocin and 0.034 for placebo. These values describe precision of the estimated means, not the variability of individual scores. The interval for the difference is the relevant posted uncertainty measure for the comparison. The registry does not give participant-level observations, alternative model estimates, or sensitivity analyses with which to assess how robust the interval is to analysis choices, exclusions, or distributional features.

The confidence interval should also be interpreted with the unadjusted nature of the reported p-value in mind. The registry does not report a multiplicity adjustment or a testing hierarchy. The interval is attached to the posted comparison, but the record does not establish how it would change under any broader inferential plan.

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 gives the number of participants needed per arm for each stated fraction of the observed effect. These are participants per arm, not total event counts. The assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The registry does not report the original trial’s sample-size justification or design assumptions, so these figures should not be read as the trial’s planned enrollment target.

For planning at 100% of the observed effect, the illustration gives 73 participants per arm for 80% power and 98 participants per arm for 90% power. At 75% of the observed effect, it gives 130 participants per arm for 80% power and 174 participants per arm for 90% power. At 50% of the observed effect, it gives 292 participants per arm for 80% power and 391 participants per arm for 90% power. The pattern illustrates that smaller effects require more participants to achieve the same target power under these assumptions.

In plain language, the 80% power column describes the planning sample size associated with an 80% chance of detecting the specified effect if the assumptions and effect are correct; the 90% column gives the corresponding planning size for a higher target chance. These are prospective planning calculations based on the observed effect, not a measure of the trial’s observed power. The illustration does not establish what effect the trial was originally designed to detect.

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 = -0.098; pooled standard deviation σ = 0.211.
2. n = 2 × (1.96 + 0.842)² × 0.211² / -0.098² = 72.9, rounded up to 73 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 <- -0.098                      # observed difference in means
sigma <- 0.211227203519929                     # 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(3074409)
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 = 73 (80% power), 98 (90% power)
 75% of observed effect: n per arm = 130 (80% power), 174 (90% power)
 50% of observed effect: n per arm = 292 (80% power), 391 (90% power)
power.t.test: 73.9 
Simulated power at n = 73 per arm: 0.797 (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 2 outcome measures, but it posts no secondary outcome result. The primary outcome is heartbeat-detection accuracy, and one comparison is reported between oxytocin and placebo. The study description also refers to whether effects vary with emotional valence, but the record does not post an analysis of that question. It does not report a complete list of planned comparisons, a testing hierarchy, or a multiplicity adjustment. The reported p-value is explicitly unadjusted, so the record does not show whether or how the possibility of multiple analyses was addressed.

Multiplicity matters when several outcomes, contrasts, subgroups, or analytic choices are examined. Without a stated plan, it is difficult to know whether the posted comparison represents the sole primary test or one among several considered. This is a general statistical concern; it does not establish that additional testing occurred in this study. The analysis plan and interim-analysis details are not reported in the registry record.

The primary-outcome population description states that 83 subjects were recruited and that 8 were excluded because of low-quality electrocardiogram data or failure of behavioral data acquisition. The posted endpoint summaries include 37 participants in the oxytocin group and 38 in the placebo group. The registry does not give a complete participant-flow account, specify the analysis population in formal terms, or explain whether the analysis followed randomized assignment regardless of data availability.

Missing-data methods are not reported in the registry record. There is no description of imputation, assumptions about missingness, sensitivity analyses, or how incomplete task data were handled. The exclusion reasons indicate data-quality and acquisition issues, but do not establish whether any other records were missing or how such records were treated. The registry also does not report an intention-to-treat or per-protocol analysis label. These omissions limit assessment of how exclusions and missing observations may relate to the estimated comparison.

The safety record uses at-risk counts of 41 for oxytocin and 42 for placebo, distinct from the primary-outcome counts. The record does not reconcile those denominators with the endpoint population. No explanation for the difference in populations or reporting conventions is included, so the safety and efficacy summaries should be read using their own stated denominators.

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 0% reporting threshold.

The registry reports serious adverse events for 0 of 41 participants at risk in the oxytocin group and 0 of 42 participants at risk in the placebo group. It also reports other adverse events for 0 of 41 at risk in the oxytocin group and 0 of 42 at risk in the placebo group. Deaths are recorded as 0 in each group. The record lists no serious adverse-event terms.

These entries mean that no affected participants are recorded in the posted serious and other adverse-event categories for the stated at-risk groups. They do not demonstrate that an event is impossible or establish the full safety profile. The registry does not describe event-collection procedures, follow-up duration, severity grading, or whether every type of event was actively solicited. The safety denominators also differ from the primary-outcome group sizes, and the registry does not explain that difference. The counts should be understood as registry entries for the specified categories and at-risk groups, not as a complete account of all possible adverse experiences.

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

The result describes a difference in mean heartbeat-detection accuracy between the two randomized groups at the assessment 45 minutes after treatment. The oxytocin group mean was 0.637, and the placebo group mean was 0.735. The reported mean difference was -0.098, with a 95% confidence interval from -0.194 to -0.002. Thus, for the participants represented in the posted endpoint summaries, the oxytocin group had a lower average on this task measure than the placebo group.

This is a result about one operational measure of interoception, not a comprehensive measure of internal bodily awareness. The registry does not provide a threshold for a meaningful change in the ratio, so it cannot establish whether the observed contrast matters in practical terms. The reported endpoint also does not directly measure a health outcome. No medical advice or treatment conclusion follows from this statistical comparison.

The study description asks whether effects vary with emotional valence, but no such result is posted. There are no secondary outcome results in the registry record. Consequently, the primary comparison does not establish whether emotional valence modifies the observed group contrast or whether other measures show a similar pattern.

The unadjusted p-value is 0.045, but the registry does not report a multiplicity strategy or enough information about the analysis plan to interpret the result within a broader set of potential tests. The confidence interval provides a range for the estimated contrast under the calculation used, yet does not remove uncertainty about the study’s analytic choices, exclusions, or generalizability. The result should therefore be read as the posted estimate for this task and time point, with the stated uncertainty and the unreported details kept in view.

16. Why This Trial Matters Statistically

This trial illustrates the distinction between a research question and the endpoint that is actually reported. The study description concerns oxytocin, interoception, and possible variation by emotional valence. The posted primary result is narrower: a comparison of heartbeat-detection accuracy between oxytocin and placebo at 45 minutes after treatment. Statistical interpretation depends on keeping the measured outcome and the broader scientific question separate.

It also shows why an estimated difference should be read together with its uncertainty. The mean difference of -0.098 is accompanied by a confidence interval that extends from -0.194 to -0.002. The interval conveys a range of estimates compatible with the calculation, while the unadjusted p-value does not describe the size of the contrast or its practical importance. The registry provides no threshold for meaningful change on the endpoint’s ratio scale.

Another statistical issue is the distinction between planning and observed results. The sample-size illustration uses fractions of the observed effect under stated planning assumptions, but the trial’s original design assumptions are not reported. Those planning figures cannot establish the design rationale or the probability that this completed study detected an effect.

Finally, the record identifies exclusions due to measurement quality and data acquisition, but does not detail missing-data handling, multiplicity, or analysis populations. These are not merely technical footnotes: they shape how confidently readers can connect a reported estimate to the randomized comparison and to the broader research question. The posted statistics support a focused reading of one endpoint while leaving important analytic context unreported.

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 registry record gives limited information about the trial procedures. It identifies a randomized, double-masked, parallel design but does not explain the randomization method, allocation concealment, or which roles were masked. It describes intranasal treatment but does not report the oxytocin dose or further administration details. These omissions restrict assessment of design implementation.

The endpoint population is affected by 8 exclusions among the 83 recruited subjects, attributed to electrocardiogram quality or failure of behavioral-data acquisition. The registry does not provide a full participant flow, individual reasons by arm, or methods for handling missing observations. The posted endpoint counts are 37 for oxytocin and 38 for placebo, while the safety record lists 41 and 42 at risk; the record does not explain the distinct denominators.

The registry does not report the original sample-size justification, statistical analysis plan, interim analyses, missing-data approach, or multiplicity procedure. Although it posts an unadjusted p-value, it does not state how the primary comparison was situated among any other planned analyses. It lists no secondary results, including no posted result addressing variation by emotional valence, despite that topic appearing in the study description.

The primary result concerns one task-based measure at one stated assessment time. The record does not provide participant-level data, details about task-block completion, or a threshold for interpreting the size of a difference. It also does not describe longer-term assessment or outcomes beyond the reported task measure. These limitations do not change the posted values, but they narrow the conclusions that can be drawn from them.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Modulatory Effect of Oxytocin on Interoception — NCT03074409.
• Registry record last updated 2019-10-18; status last verified 2019-09.
• Results first posted 2019-10-18.
• 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

NCT03074409 is a completed, randomized, double-masked, parallel-group study titled “The Modulatory Effect of Oxytocin on Interoception,” sponsored by the University of Electronic Science and Technology of China. The registry records 83 enrolled subjects and a primary endpoint of heartbeat-detection accuracy measured 45 minutes after intranasal oxytocin or placebo.

The posted primary-outcome means are 0.637 for oxytocin and 0.735 for placebo. The reported mean difference is -0.098, with a 95% confidence interval from -0.194 to -0.002 and an unadjusted p-value of 0.045. The record notes 8 exclusions related to electrocardiogram data quality or behavioral-data acquisition and provides no secondary outcome results.

Serious and other adverse-event counts are recorded as 0 in each group, with 41 and 42 participants at risk, respectively; deaths are recorded as 0 in both groups. Important details about the original design assumptions, analysis plan, missing-data handling, and multiplicity are not reported in the registry record.