Does Social Buffering Continue to be Effective Over the Peripubertal Period When Friends Share the Stressor Experience?: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Does Social Buffering Continue to be Effective Over the Peripubertal Period When Friends Share the Stressor Experience? (NCT04311996), a randomized na study. The primary outcome is Cortisol AUCi.

2. Clinical Question

The study addresses a specific distinction in social buffering: a friend may offer support while remaining outside the stressor, or may share the stressor experience. The four conditions also include an unfamiliar peer and no partner, helping describe how the friend-and-target condition compares with different social contexts. The registry record identifies the outcome timeframe as 2 hours. It does not report enough detail in the available record to establish whether the posted comparisons account for participants’ pubertal stage or test a change in buffering across stages. Accordingly, the posted results describe differences among conditions; they should not be read as a direct demonstration of a developmental pattern unless such an analysis is reported elsewhere in the registry record.

3. Trial Design

The registry record describes a randomized, unmasked, parallel experiment with four conditions. Participants were assigned to the friend-and-target condition, friend support without shared stress, an unfamiliar peer who shares the stressor, or no partner. The listed interventions are categorized as other, and the study phase is recorded as not applicable. Enrollment is listed as 269, while the primary outcome description says 256 participants completed the experiment. These are different registry figures and should not be treated as interchangeable.

For the primary outcome, the record describes 61 participants in the friend-and-target condition, 65 in friend support, 62 in the unfamiliar-peer condition, and 68 in the alone condition as having completed the experiment. The posted outcome values use analyzable numbers that vary by condition and outcome. For cortisol, those numbers are 61, 63, 61, and 66, respectively. The record explicitly notes that available analyzable data were used and that outcome-level numbers could be lower. It does not explain the specific reasons for each difference.

The experiment was recorded as completed. The listed start and primary-completion dates are in the registry record, but those dates do not provide information about how assignment was generated, whether allocation was concealed, or how outcome assessment was conducted. Those details are not reported in the registry record. Masking is listed as none, so the record does not describe a masked study. The original statistical analysis plan, sample-size justification, and any planned interim analyses are not reported in the registry record. The planning illustrations accompanying the posted comparisons are calculations based on the observed effects and are not evidence of the trial’s original design assumptions.

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: 256 participants were completed, with 61 Friend and Target Both, 65 Friend Support, 62 Unfamiliar Peer, and 68 Alone. For each outcomes, all available analyzable data were included in results, sometimes resulting in lower n.

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 Cortisol AUCi, measured over 2 hours.

The primary endpoint is cortisol AUCi, or area under the curve increase from baseline, measured in (ug/dl)*minutes over a timeframe of 2 hours. It summarizes the pattern of cortisol measurements after the stress task relative to a specified starting sample. The registry description says that seven samples were taken during the assessment: one on arrival for training purposes, then samples after a calming video, after speech preparation, after the speech and math performance, and at subsequent intervals. AUCi uses samples 2–7, with sample 2 serving as the intercept.

As a continuous measure, AUCi is reported as a mean with a standard deviation. Summarizing the sequence of measurements in an area-under-the-curve measure can represent the overall increase in cortisol across the assessment rather than focusing on one sample alone. It is therefore suited to describing the physiological response over the stated measurement period. This endpoint concerns cortisol response; it is not itself a measure of subjective stress or a direct measure of whether participants felt supported.

The registry record gives condition means and standard deviations and lists comparisons of each other condition with the friend-and-target condition. It does not report a detailed interpretation of what magnitude of AUCi difference would be clinically or developmentally important. The endpoint can describe observed physiological differences, but by itself it cannot establish why those differences occurred or whether they persist beyond the assessment period.

7. Results Posted in the Registry

• Experimental: Friend Provides Support: mean 0.03 versus 0.00 on control; difference 0.02 (95% CI 0.00 to 0.04), standardized effect d = 0.35; z-test p = 0.048.
• Experimental: Unfamiliar Peer and Target: mean 0.02 versus 0.00 on control; difference 0.01 (95% CI -0.01 to 0.03), standardized effect d = 0.24; z-test p = 0.180.
• Experimental: Alone: mean 0.02 versus 0.00 on control; difference 0.02 (95% CI -0.00 to 0.04), standardized effect d = 0.30; z-test p = 0.097.

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.

Self Report of Stress

• 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:
  • Experimental: Friend Provides Support: mean 1.60 versus 1.65 on control; difference -0.05 (95% CI -0.38 to 0.28), standardized effect d = -0.05; z-test p = 0.767.
  • Experimental: Unfamiliar Peer and Target: mean 1.65 versus 1.65 on control; difference 0.00 (95% CI -0.33 to 0.33), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Experimental: Friend Provides Support 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 Experimental: Unfamiliar Peer and Target and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts self-reported stress as a secondary outcome, measured on a scale over the 2-hour timeframe. Its condition means are close in the posted summaries, and the reported pairwise comparisons against the friend-and-target condition do not show a clear separation. The available comparisons are limited: the registry lists results for friend support and the unfamiliar-peer condition, but not a corresponding posted comparison for the alone condition. The record should not be read as if every possible secondary contrast were reported.

The broad secondary pattern does not mirror a strong, consistent difference across social conditions. That is useful context alongside the primary cortisol findings, but it does not confirm or negate a physiological contrast. Cortisol response and self-reported stress are different outcomes: one summarizes a physiological measure, while the other records participants’ reports on a scale. Differences in what they measure can lead to different patterns without either endpoint being a direct validation of the other.

The secondary results carry less inferential weight because they are labeled unadjusted and belong to a broader set of outcomes. The registry lists 15 outcome measures overall, while only one secondary outcome is described in the posted information. When many outcomes or contrasts are considered, some results can appear notable by chance unless multiplicity is addressed. The registry record does not report a multiplicity plan or a complete analysis plan. Secondary comparisons are therefore best treated as supporting, exploratory context rather than independent confirmation of the primary result.

The record also does not report a prespecified hierarchy explaining how secondary measures were intended to inform the main question. Without that context, the secondary outcome can describe perceived stress during the assessment, but cannot establish whether a physiological pattern is accompanied by a corresponding subjective experience or whether differences reflect a developmental change.

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

• Self Report of Stress

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous, and the registry reports a mean for each condition together with a standard deviation. A mean describes the central level of the measured cortisol response in a condition. A standard deviation describes how spread out individual values are around that mean; it is not a confidence interval and does not directly express uncertainty in the estimated mean.

What does a mean difference represent?

The posted comparisons express the difference between the mean in one condition and the mean in the friend-and-target condition. A positive mean difference in these comparisons corresponds to a higher posted mean in the comparison condition. This is a contrast of group summaries, not a measure of the response for any one participant. The comparison metric is listed as a mean difference, and the registry gives the two condition means used in each contrast.

How should the confidence interval be read?

A confidence interval describes uncertainty around an estimated contrast under the assumptions of the method used to construct it. An interval that includes zero is compatible with no difference as well as with differences in either direction within the interval. An interval that stays on one side of zero indicates that the values in that interval are on that side, but it does not by itself address whether the contrast was planned, whether multiple comparisons were considered, or whether the magnitude matters substantively. The method used to calculate these intervals is not reported in the registry record.

What does an unadjusted p-value tell the reader?

A p-value is a measure of how incompatible the observed data are with a specified null model, under the assumptions of the analysis. The registry labels the posted values unadjusted. They therefore should not be interpreted as accounting for the several condition comparisons or the broader set of recorded outcomes. A p-value does not give the probability that the null hypothesis is true, the probability that a finding will replicate, or the size or importance of an effect. The statistical analysis plan and any multiplicity procedure are not reported in the registry record.

The primary comparison results are posted as pairwise contrasts with the friend-and-target condition as the reference. They do not provide an omnibus test of whether any condition differs from any other, nor a direct estimate of a developmental interaction. The registry reports no primary analysis count in its analysis summary, and a full specification of the statistical model is not reported. It is therefore not possible to determine from the registry record whether the posted contrasts were adjusted for covariates, stratification variables, or other design features.

Why can outcome-specific sample sizes differ?

The primary outcome description states that analyses use all available analyzable data and that the number can be lower for an outcome. This is consistent with the posted cortisol condition counts differing from the completion counts. It does not identify why any participant lacked analyzable data, whether missingness was related to condition or response, or whether any imputation or sensitivity analysis was used. Those details are not reported in the registry record. If missingness is related to observed or unobserved outcomes, an analysis based only on available observations may not represent the full randomized groups in the same way.

The secondary outcome is self-reported stress, also reported as a continuous mean and standard deviation. The same broad ideas about means, variability, contrasts, confidence intervals, and unadjusted p-values apply. However, the posted secondary comparisons are not a substitute for the primary cortisol endpoint, because they measure reported stress on a scale rather than cortisol physiology. The record lists one secondary outcome and states that there are 15 outcome measures overall, but does not provide all of those outcome results in the information summarized by the posted comparisons.

What do the planning illustrations mean?

The sample-size illustrations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They show participants per arm for continuous-outcome comparisons, using stated fractions of the observed effect as planning targets. The calculations are illustrations based on observed results, not the original trial design plan. Smaller target effects require more participants because they are harder to distinguish from the variability in the outcome. These calculations do not establish how the study was designed or analyzed.

Over-dispersion is a general concern when observed variability is greater than a model assumes: standard errors can be too small if the model fails to account for that extra variability, making intervals too narrow and tests too optimistic. The registry record does not identify an over-dispersion problem or report a model for it in this experiment. It would therefore be inappropriate to claim that over-dispersion was present or that a particular correction was used.

In general, estimates should be read together with their uncertainty, the analysis population, and the number of comparisons being considered. In this experiment, the record provides useful descriptive summaries and unadjusted pairwise comparisons, but it does not report enough methodological detail to reconstruct all analytic decisions or verify whether the posted contrasts correspond to a prespecified analysis.

11. Confidence Intervals and Precision

The confidence intervals show that precision varies across the posted primary comparisons. The friend-support contrast has an interval whose lower endpoint is close to zero and whose upper endpoint is above zero. This indicates that the estimated contrast is uncertain near the no-difference value, even though its unadjusted p-value is 0.048. The closeness of that result to the conventional threshold makes it especially important not to treat the p-value as a standalone conclusion.

The intervals for the unfamiliar-peer and alone comparisons extend across zero. They allow both a difference in the direction of a higher mean in the comparison condition and values consistent with little or no difference. Such intervals do not demonstrate equivalence: equivalence requires a prespecified range of differences considered unimportant and an analysis designed to assess that range. No such margin is reported in the registry record.

Confidence intervals communicate uncertainty in the estimated mean differences; they do not capture every source of uncertainty. They cannot resolve questions about missing outcomes, the choice of analysis, unreported multiplicity procedures, or whether an effect is important for the study’s developmental question. The registry record does not report the interval-construction method, so its precise assumptions cannot be assessed from the posted information.

The condition standard deviations are similar in the posted primary summaries, but similarity of standard deviations does not make the estimated contrasts precise by itself. Precision also depends on the analyzable sample size and the interval method. The outcome counts are 61, 63, 61, and 66 across the four conditions. The record notes that available analyzable observations were used. Thus, the intervals should be read as uncertainty for the posted outcome-level comparisons, not as a guarantee that every randomized participant is represented in the same way.

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 participants and not total events. The registry’s 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 to a stated fraction of the observed effect. For time-to-event outcomes, the convention would instead be total events, but the comparisons are continuous outcomes, so the figures are per arm.

The column for 80% power gives the number of participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the target effect under the stated assumptions, if that effect and those assumptions applied. The 90% power column uses a higher chance of detection and therefore generally requires a larger planned sample. These are planning probabilities under a model, not guarantees about what a future study will find.

Rows planning for 75% or 50% of the observed effect target a smaller effect than a row using 100% of the observed effect. Smaller effects are harder to distinguish from outcome variability, so they require more participants. For example, for the friend-support contrast, planning for 75% of the observed effect gives 222 participants per arm for 80% power and 297 for 90% power; planning for 50% gives 499 and 668 participants per arm, respectively. These are illustrations rather than recommendations, and the original trial’s design assumptions are not reported in the registry record.

The figures are not observed power. They do not measure the probability that the completed experiment detected an effect, and they should not be used to reinterpret the posted p-values. They show how sample-size requirements would change under the listed planning assumptions and target effects.

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 = 0.021; pooled standard deviation σ = 0.060.
2. n = 2 × (1.96 + 0.842)² × 0.060² / 0.021² = 124.6, rounded up to 125 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 <- 0.0213                      # observed difference in means
sigma <- 0.06                     # 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(4311996)
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 = 125 (80% power), 167 (90% power)
 75% of observed effect: n per arm = 222 (80% power), 297 (90% power)
 50% of observed effect: n per arm = 499 (80% power), 668 (90% power)
power.t.test: 125.5 
Simulated power at n = 125 per arm: 0.795 (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

Multiplicity arises when several outcomes or comparisons are evaluated. Each additional test creates another opportunity for an apparently notable result to occur by chance, even when no underlying difference exists. The primary results include comparisons of three conditions against the friend-and-target condition, and the registry identifies 15 outcome measures overall. The record does not report whether the study used a multiplicity adjustment, a prespecified hierarchy of outcomes, or another method to control the interpretation of multiple tests. The posted p-values are explicitly unadjusted.

This is relevant to the friend-support comparison, whose unadjusted p-value is 0.048. That value is close to a commonly used threshold, but the registry record does not establish that the comparison was the sole planned test or that the threshold accounts for other contrasts. The other primary comparisons have unadjusted p-values of 0.18 and 0.097. Those results should likewise be interpreted with their confidence intervals and the broader comparison context, rather than as isolated pass-or-fail decisions.

Missing-data handling is also incompletely described. The primary outcome description says that available analyzable data were used and that the number may be lower for an outcome. The record gives outcome-specific counts but does not explain why observations were unavailable, how missingness was assessed, whether imputation was used, or whether sensitivity analyses were conducted. The analysis population definition beyond available analyzable data is not reported in the registry record.

The registry record does not report the original statistical analysis plan, model details, covariate adjustment, interim analyses, or sample-size justification. These omissions limit assessment of whether the posted comparisons were prespecified and whether the analytic choices align with the study’s primary question. They also limit evaluation of how the randomized design was incorporated into the reported estimates. These are limitations of the available record, not evidence that a particular procedure was or was not performed.

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.

The adverse-event record lists serious adverse events and other adverse events as zero in each of the four reported groups. The numbers at risk shown for serious and other adverse events are 63 for friend and target both, 72 for friend provides support, 65 for unfamiliar peer, and 69 for alone. Deaths are listed as zero in every group. These figures describe what was recorded in the registry’s adverse-event fields for the groups and categories shown.

Zero recorded events does not establish that no adverse events occurred outside the record’s definitions, collection period, or reporting process. The registry record does not provide event terms, a detailed safety-monitoring description, or an explanation of how adverse events were collected. No serious adverse-event terms are listed. The safety figures should therefore be read as registry reporting, not as a comprehensive independent assessment of safety or as a comparison demonstrating that one condition is safer than another.

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

The posted results describe differences in mean cortisol AUCi among four randomized conditions. The friend-and-target condition has the lowest reported mean, and the comparison with friend support without shared stress has an unadjusted p-value of 0.048. The unfamiliar-peer and alone comparisons have unadjusted p-values of 0.18 and 0.097, with confidence intervals that include zero. Taken together, the data show a suggestive contrast for friend support relative to shared stress, while the other contrasts remain uncertain.

These results do not establish that friends’ support is effective or ineffective, and they do not show that sharing the stressor causes a particular response. The registry record does not report enough analysis detail to assess all assumptions, missing-data choices, or multiplicity handling. Nor does it report an analysis directly testing whether social buffering changes over puberty, despite the developmental focus in the study title. The posted condition comparisons alone cannot answer that developmental question.

The primary endpoint measures cortisol physiology during the assessment. It does not directly measure emotional support, perceived safety, longer-term stress regulation, or later outcomes. The secondary self-report measure addresses a different aspect of the experience, and its posted comparisons do not supply a direct substitute for a developmental analysis of cortisol.

Confidence intervals matter because estimates from a study are uncertain. The intervals for two primary contrasts include zero, and the interval for the friend-support contrast approaches zero at its lower end. The absence of a clear separation in a comparison is not proof that conditions are the same. Conversely, a p-value near a conventional threshold does not by itself establish a robust or important effect. Interpretation should account for the scale and uncertainty of the estimates, the number of comparisons, and the limited analysis details in the registry record.

The most defensible reading is that the posted summaries offer an initial description of cortisol response across the four conditions, with one unadjusted comparison near a conventional threshold and other comparisons inconclusive. They do not settle the wider question of how social buffering operates across puberty.

17. Why This Trial Matters Statistically

This experiment is statistically interesting because its conditions distinguish several social contexts rather than reducing the question to support versus no support. A friend may be present and share the stressor, provide support without sharing it, or be replaced by an unfamiliar peer who shares the stressor; the alone condition provides a no-partner context. Those contrasts make the role of shared experience a central part of the question.

The primary measure is a physiological summary of multiple cortisol samples, while the secondary measure captures self-reported stress. Reading both kinds of outcomes helps distinguish what the experiment measured from what it did not: a physiological response over an assessment period is not identical to a participant’s subjective report. Their patterns need not match for either measure to be informative.

The example also illustrates why statistical interpretation cannot rest on a single p-value. The friend-support comparison is near a conventional threshold, but it is one of several comparisons, and the registry record does not report multiplicity handling or a complete analysis plan. Confidence intervals and outcome-level sample sizes provide important context, while missing-data details and the absence of a reported developmental interaction limit broader conclusions.

Finally, the planning illustrations show how substantially the required sample size can change when the target effect is smaller than the observed effect. They are useful for understanding planning sensitivity, but they are not the experiment’s original design rationale and do not establish observed power. Separating the posted findings, their precision, and the information not reported in the registry record is essential for a balanced statistical reading.

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 registry record does not report a direct analysis of change across puberty or an interaction between developmental stage and condition. The primary comparisons therefore address condition differences as posted, not whether social buffering changes over the peripubertal period. The absence of that result is particularly important given the developmental question stated in the study title.

Some details of the randomized design and analysis are also not reported. The record lists randomization and no masking, but does not explain the assignment procedure, allocation concealment, outcome assessment procedures, or the statistical model. It does not report the original analysis plan, interim analyses, or the study’s own sample-size justification. The planning calculations are based on observed effects and cannot fill those gaps.

Outcome-specific counts are lower than the completion counts in some conditions, and the registry says available analyzable data were used. Reasons for unavailable data, missing-data assumptions, and sensitivity analyses are not reported. The record also does not state whether the primary comparisons were adjusted for multiple testing. These omissions make it difficult to assess potential bias or to determine how strongly the posted p-values should be weighted.

The posted primary comparisons use the friend-and-target condition as their reference. The registry does not provide all pairwise contrasts among conditions, and the secondary outcome information does not include a posted comparison for every condition. The record lists 15 outcome measures overall, but the available results describe only a subset. Selective availability of outcome details can limit a complete view of the study’s findings.

These are limitations of what the registry record reports. They do not establish that specific methods were absent or that the results are invalid. They do mean that interpretation should stay close to the posted summaries and avoid claims about developmental moderation, equivalence, or a definitive causal mechanism.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Does Social Buffering Continue to be Effective Over the Peripubertal Period When Friends Share the Stressor Experience? — NCT04311996.
• Registry record last updated 2025-02-24; status last verified 2025-02.
• Results first posted 2025-02-24.
• 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

The registry describes a completed randomized, unmasked, parallel experiment with four conditions: a friend and target sharing the stressor, friend support without shared stress, an unfamiliar peer sharing the stressor, and no partner. Enrollment is listed as 269. The primary endpoint is cortisol AUCi, a continuous measure of cortisol response over a 2-hour timeframe. The primary outcome description states that 256 participants completed the experiment and that outcome analyses used available analyzable data.

The friend-and-target condition has the lowest posted primary mean. Relative comparisons show an unadjusted p-value of 0.048 for friend support, 0.18 for the unfamiliar-peer condition, and 0.097 for alone. The latter two confidence intervals include zero. The registry also posts self-reported stress as a secondary outcome, but the available comparisons are limited. The record does not report a direct analysis of change across puberty, multiplicity handling, the full statistical plan, or missing-data methods. The results are descriptive evidence about the posted contrasts, with important uncertainty and reporting limitations.