A Digital Intervention to Prevent the Initiation of Opioid Misuse in Adolescents in School-based Health Centers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Digital Intervention to Prevent the Initiation of Opioid Misuse in Adolescents in School-based Health Centers (NCT04941950), a randomized na study. The primary outcome is Percentage Reporting 'Great Perceived Risk of Harm' From Opioid Misuse.

2. Clinical Question

3. Trial Design

The registry describes a randomized, single-masked, parallel-group trial. Participants were assigned to PlaySmart or a control video game. The record identifies the study as completed and reports enrollment of 533. The recorded study start date is 2021-10-21, and primary completion is listed as 2025-03-01. The phase is listed as NA.

The primary outcome was assessed at 3 months. The reported arm results list 231 participants for PlaySmart and 234 for the control game. These are the group sizes associated with the primary outcome results; the registry record does not explain how these figures relate to the total enrollment of 533, including whether any participants were excluded from this outcome analysis or had missing outcome data.

Single masking is recorded, but the identity of the masked group—participants, intervention staff, outcome assessors, or others—is not reported in the registry record. The randomization method, allocation concealment procedures, and any stratification factors are also not reported in the registry record. The record identifies Tyra Boomer as sponsor.

The primary endpoint concerns perceived risk of harm from opioid misuse. The record also lists secondary outcomes on intentions to misuse opioids, self-efficacy for refusing opioids, knowledge about opioid misuse and its risks, and positive and negative attitudes toward opioid misuse. There are 26 outcome measures listed, while no analysis entries are listed. The registry does, however, post a primary comparison summary. Further details about the planned analysis and the handling of secondary outcomes 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.

Overall Study

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: One control participant was withdrawn by the principal investigator due to an incomplete parent/guardian consent form, a protocol deviation that was reported to the Institutional Review Board (IRB), resulting in a sample size of 532.

6. Primary Endpoint

The registry lists the primary outcome as Percentage Reporting 'Great Perceived Risk of Harm' From Opioid Misuse, measured over 3 months.

The primary endpoint is the percentage of participants reporting “great perceived risk of harm” from opioid misuse at 3 months. It was assessed with eight items adapted from the U.S. Monitoring the Future survey: five items address heroin and three address prescription opioid misuse. For each item, participants were asked how much risk people have of harming themselves if they use the drug once or twice, occasionally, or regularly. Responses ranged from “no risk” to “great risk,” with an additional “can’t say, drug unfamiliar” option.

Item responses were scored from 1 to 4, and the eight item scores were summed to produce a total score ranging from 8 to 32. A score of 32 defined “great perceived risk of harm”; scores less than 32 were classified as “no great perceived risk of harm.” Thus, the endpoint represents the share reaching the maximum total score across the items, not an average score or a measure of actual opioid use.

This kind of perception measure is relevant to a prevention study because it captures participants’ stated appraisal of risk. It does not directly establish whether participants initiated misuse, changed their behavior, or experienced harm. The registry outcome is titled as a percentage, while the posted arm values are counts of participants. The endpoint definition and the reported counts should therefore be read together.

7. Results Posted in the Registry

• PlaySmart Game: 28.6% (66/231) versus 22.6% (53/234) on control; risk difference 5.9 percentage points (95% CI -2.0 to 13.8), risk ratio 1.26 (0.92–1.72); unadjusted two-proportion z-test p = 0.143.

8. Statistical Interpretation

9. Statistical Methods Explained

In general, a randomized parallel-group trial assigns participants to separate groups and compares outcomes between those groups. Random assignment is intended to reduce systematic differences at baseline, but it does not by itself resolve questions about adherence, missing outcomes, analysis choices, or the conduct of masking. For this trial, the registry records randomization and a parallel model. Details of the sequence generation, allocation concealment, and masking procedures are not reported in the registry record.

What does the primary comparison estimate?

The main comparison is an absolute risk difference for a binary endpoint: whether a participant did or did not reach the score defined as “great perceived risk of harm.” An absolute difference expresses the contrast in the chance of the endpoint between the intervention and control groups. The posted estimate is 0.059. Its confidence interval includes the no-difference value, with a lower bound of -0.02 and a positive upper bound. This means that the uncertainty range allows both a negative and a positive difference under the interval procedure used for the posted result.

The registry also reports a risk ratio of 1.261 and an odds ratio of 1.366. In general, a risk ratio compares the probabilities of an outcome between groups. An odds ratio compares outcome odds instead, so it is a different scale and should not be read as an absolute percentage-point difference. These relative estimates provide additional descriptions of the same binary outcome, not independent trial outcomes. The reported relative confidence intervals include the value corresponding to no group contrast.

How should the p-value be understood?

The unadjusted p-value is 0.143. In general, a p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that null hypothesis and the test’s assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The registry labels this value as unadjusted. The record does not report a prespecified analysis plan or say whether this was the sole primary test in the original analysis plan.

A p-value should be considered alongside the estimated contrast and its confidence interval. The estimate indicates the direction and scale of the posted contrast; the confidence interval indicates the range of values compatible with the data under the procedure. Neither alone establishes a behavioral or clinical effect. The outcome is a self-reported perception endpoint, and the posted result concerns that endpoint at 3 months.

Why can the interval be wide or cross no difference?

In general, a confidence interval describes uncertainty in an estimate under a statistical model and procedure. Its width reflects factors such as the amount of information available and the variability of the outcome. For a binary outcome, the precision depends in part on how many participants contribute usable outcome data and on how common the outcome is. Here, the arm counts associated with the outcome are 231 and 234. The registry does not describe missingness or the reasons why the primary outcome counts differ from total enrollment.

An interval that includes the no-difference value does not prove that the groups are identical. It indicates that the data and procedure do not distinguish a range of plausible contrasts from no difference at the interval’s stated level. The exact interval level and method are not reported in the registry record, so those features cannot be independently assessed from the entry.

What does over-dispersion mean in general?

Over-dispersion is a general concern when observed variability is greater than a model assumes. It can make standard errors too small and confidence intervals too narrow if the model does not account for the extra variability. The primary endpoint is a binary participant-level outcome, not a count of repeated events, and the registry does not report a dispersion parameter or an over-dispersion adjustment. Such an adjustment should not be inferred from the posted summary.

The posted comparison gives a useful compact description, but a full statistical assessment would also require the analysis population, treatment of incomplete responses, model specification, and prespecified testing strategy. Those details are not reported in the registry record. The registry lists zero analysis entries while also posting a comparison summary, so the comparison can be described, but the underlying analytic choices cannot be reconstructed from the record.

10. Confidence Intervals and Precision

The absolute contrast is reported as a risk difference of 0.059. Its confidence interval has a lower bound of -0.02 and an upper bound above zero. Because the interval spans zero, the posted uncertainty range includes no absolute difference as well as positive and negative differences. The record does not state the confidence level or the method used to construct this interval, so its precise statistical calibration is not reported in the registry record.

In general, a confidence interval is a range generated by a statistical procedure that reflects uncertainty in an estimate. It is not a guarantee that the true value lies in the interval, nor is it a range of individual participant outcomes. The interval’s breadth is important: it indicates that the posted data do not pin down the size or direction of the absolute contrast with high precision. The exact positive upper endpoint is not restated because the displayed precision limit is three decimal places.

The registry additionally reports a risk ratio of 1.261 and an odds ratio of 1.366. These point estimates describe relative contrasts, while their reported intervals include the respective no-contrast value. This reinforces that the point estimates should not be treated as precise evidence of a particular relative effect. The unadjusted p-value is 0.143, which is also consistent with uncertainty about the contrast under the test used.

Precision should be considered in light of the discrepancy between total enrollment of 533 and the primary-outcome group counts of 231 and 234. The reason for the difference and the amount or pattern of missing outcome data are not reported in the registry record. Without those details, it is not possible to assess how missingness may have affected precision or the estimated contrast.

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 sample-size figures are planning illustrations based on the observed effect, not the original trial design calculation. Because this is a binary participant-level outcome, the figures are participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

For planning around 100% of the observed effect, the illustration gives 852 participants per arm for 80% power and 1,140 participants per arm for 90% power. For 75% of the observed effect, it gives 1,486 participants per arm for 80% power and 1,989 participants per arm for 90% power. For 50% of the observed effect, it gives 3,277 participants per arm for 80% power and 4,387 participants per arm for 90% power.

In plain language, the 80% power column describes the planned number of participants per group under the stated assumptions for a design with an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column represents a higher target chance of detecting that effect, also under those assumptions. Smaller effects need more participants to detect, which is why the planning numbers rise as the assumed effect becomes a smaller fraction of the observed effect. This is not observed power; it is a planning illustration rather than a measure of what the completed trial achieved.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.226; treatment proportion p₁ = 0.286; average p̄ = 0.256.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6173 = 1.2098.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6159 = 0.5183.
4. n = (1.2098 + 0.5183)² / (0.0592)² = 851.6, rounded up to 852 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
p0 <- 0.226495726495726                          # control proportion
p1 <- 0.285714285714286                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4941950)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 852 (80% power), 1,140 (90% power)
 75% of observed effect: n per arm = 1,486 (80% power), 1,989 (90% power)
 50% of observed effect: n per arm = 3,277 (80% power), 4,387 (90% power)
power.prop.test: 851.6 
Simulated power at n = 852 per arm: 0.794 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 26 outcome measures and identifies five secondary outcomes: intentions to misuse opioids, self-efficacy for refusing opioids, knowledge about opioid misuse and its risks, and positive and negative attitudes toward opioid misuse. It also lists zero analysis entries, although a primary comparison summary is posted. The record does not report which outcomes were designated in advance as primary or secondary beyond the listed primary endpoint and secondary outcomes, nor does it describe a hierarchy or testing strategy for the outcomes.

In general, when many outcomes or multiple statistical tests are examined, the chance of at least one apparently unusual result can increase. Statistical approaches can address multiplicity, for example by controlling a family-wise error rate or using a prespecified sequence of tests. Whether any such method was planned or used is not reported in the registry record. The posted primary p-value is explicitly unadjusted, and no multiplicity-adjusted result is reported.

Information about missing data is also limited. The registry records total enrollment of 533 and primary endpoint group sizes of 231 and 234 but does not explain the relationship between them. It does not report the number of participants with missing responses, reasons for missingness, imputation methods, or sensitivity analyses. The analysis population—such as whether results include all randomized participants or only participants with available outcome data—is not reported in the registry record.

These omissions matter because missing outcomes and multiple testing can affect both the interpretation of uncertainty and the credibility of a result. Without the planned analysis, outcome hierarchy, missing-data procedures, and participant flow for the primary measure, the posted comparison can be described but cannot be fully evaluated against the trial’s original statistical plan.

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.

Most frequent non-serious adverse event terms

The registry safety figures list 269 participants at risk in the PlaySmart group and 263 in the control group. It reports zero serious adverse events in each group and zero deaths in each group. The “other” adverse-event counts are 2 for PlaySmart and 0 for the control game. These figures are counts, not rates or percentages.

In general, adverse-event counts should be interpreted alongside the number of participants monitored, the event definitions, and the observation period. The record does not provide detailed event descriptions for the two other events, nor does it report the period or procedures used to collect safety information. The listed serious-event and death counts are therefore a limited summary of what was recorded, not a complete account of all possible harms or of the comparative safety profile.

The safety at-risk figures differ from the primary outcome group sizes. The registry does not explain the relationship between those denominators. No conclusion about causation should be drawn from the counts alone: the summary does not establish whether an event was related to the game intervention. It records the observed adverse-event counts, without providing details sufficient for broader safety interpretation.

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

The posted result concerns whether participants reported the maximum perceived-risk score at 3 months. It does not directly measure opioid misuse, initiation of misuse, or subsequent health outcomes. The endpoint is based on answers to eight questions about perceived risk associated with heroin and prescription opioid misuse, and it classifies only participants with a total score of 32 as reporting “great perceived risk of harm.”

In the reported groups, 66 PlaySmart participants and 53 control participants met that endpoint. The estimated absolute contrast is 0.059, with a confidence interval that includes zero, and the unadjusted p-value is 0.143. The result therefore has uncertainty that includes no absolute difference and contrasts in either direction. The relative estimates are also uncertain, with their reported intervals including the no-contrast value. These statistics describe the posted data; they do not establish that PlaySmart changed perceived risk or prevented opioid misuse.

Random assignment is an important design feature for comparing groups, but interpretation also depends on how assignment and masking were implemented, which participants contributed outcome data, and how missing responses and multiple outcomes were handled. Those details are not reported in the registry record. In particular, the registry does not explain why the primary outcome group sizes are smaller than total enrollment.

The results should be read as a comparison on one self-reported measure at one stated follow-up time. They do not show whether any difference persisted, whether participants changed actual behavior, or whether effects differed across participant subgroups. Those matters are not reported in the registry record. No medical or behavioral recommendation follows from these summary statistics alone.

16. Why This Trial Matters Statistically

This trial raises statistical questions that are common in prevention research. The primary outcome is a self-reported binary classification derived from a multi-item score. A clear definition is essential: here, the endpoint is limited to participants who reach the maximum total score across eight items. That definition makes the analysis interpretable, while also making clear that the result is about a specific perception threshold rather than opioid use itself.

The randomized parallel-group design supports a direct comparison of assigned groups, but the registry’s posted outcome counts do not account for the full enrollment. Understanding participant flow and missing responses is important for judging whether the analyzed groups remain comparable after randomization. The registry record does not explain this difference or identify the analysis population.

The posted absolute estimate and relative estimates are accompanied by uncertainty that includes no contrast. This highlights why statistical interpretation should consider effect estimates and intervals, not just counts or a p-value. A non-significant test does not establish equivalence, while a point estimate alone does not establish a reliable difference. The size of the uncertainty and its implications depend on the analytic method, which is not fully documented in the record.

The trial also lists many outcomes. Without a reported multiplicity strategy, it is not possible to determine how the full set of outcomes was intended to inform conclusions. These design and reporting considerations are not unique to digital interventions: they are central to assessing randomized studies that use behavioral measures, multiple endpoints, and follow-up data that may be incomplete.

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.

Several trial-specific limitations affect interpretation of the posted results. First, the primary endpoint is a self-reported measure of perceived risk. It is not a direct measure of opioid use, initiation, or harm. The threshold is also stringent by definition: “great perceived risk” requires a total score of 32 across all eight items. The result therefore addresses that specific classification and should not be generalized to every aspect of risk perception.

Second, the registry reports enrollment of 533, while the primary outcome groups contain 231 and 234 participants. The record does not explain the relationship between these numbers, describe participant flow, or report missing outcome data. This prevents assessment of how many randomized participants contributed data and whether missingness might have affected the comparison.

Third, masking is recorded as single, but the masked party is not identified. The randomization process, allocation concealment, intervention exposure, adherence, and the content or matching of the two games are not described in the registry record. These details would help readers assess implementation and possible sources of bias.

Finally, the record lists 26 outcome measures and five secondary outcomes but does not report a statistical analysis plan, multiplicity procedures, analysis population, or missing-data methods. No analysis entries are listed, despite a posted primary comparison. The confidence-interval method and level are also not stated. These gaps limit independent evaluation of the statistical procedures and the scope of conclusions that can be drawn from the registry summary.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Digital Intervention to Prevent the Initiation of Opioid Misuse in Adolescents in School-based Health Centers — NCT04941950.
• Registry record last updated 2026-07-01; status last verified 2026-05.
• Results first posted 2026-04-24.
• 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.

21. Record Summary

NCT04941950 is a completed randomized, single-masked, parallel-group trial of PlaySmart versus a control video game. The registry records enrollment of 533 and a primary outcome at 3 months: the percentage reporting great perceived risk of harm from opioid misuse. The posted primary endpoint counts are 66 of 231 for PlaySmart and 53 of 234 for the control game. The comparison reports a risk difference of 0.059 and an unadjusted p-value of 0.143; the confidence interval includes no difference.

The endpoint is based on eight self-reported items about heroin and prescription opioid misuse, with the maximum total score defining great perceived risk. The registry also lists secondary measures on intentions, refusal self-efficacy, knowledge, and attitudes. It reports zero serious adverse events and zero deaths in both groups, with other adverse-event counts of 2 and 0. Important details about the original sample-size plan, masking roles, missing data, analysis population, and statistical methods are not reported in the registry record.