An App to Reduce Cannabis Use Among Emerging Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An App to Reduce Cannabis Use Among Emerging Adults (NCT05824754), a na na study. The primary outcome is App Feasibility.

2. Clinical Question

This is a feasibility question about whether participants move from consent to app download, rather than a reported estimate of whether cannabis use changed. The registry describes a broader purpose of piloting prevention methods and learning about appealing and helpful ways to deliver information to young adults who use mobile apps. It does not report a primary cannabis-use outcome result in the material available in the registry record.

The question also has a clear denominator issue. The outcome is described as the percentage of people who consented and downloaded the app, while the record notes that not everyone who consented was included in study flow because not everyone downloaded the app. Without the number who consented, the number who downloaded, and a completed endpoint result, the feasibility proportion cannot be assessed from the registry record.

3. Trial Design

The registry describes a completed pilot study of the Just-In-Time-Adaptive-Intervention (JITAI), titled “An App to Reduce Cannabis Use Among Emerging Adults.” Its model is single group, allocation is listed as NA, and masking is listed as none. Enrollment is reported as 122. The condition recorded is cannabis use. The registry lists the intervention type as other.

The intervention description includes within-participant micro-randomization: twice daily, a participant is assigned to receive a message or no message. A reinforcement learning algorithm makes those assignments based on the participant’s engagement with the intervention. This describes repeated message decisions within the app experience; it does not make the overall study a conventional parallel-group randomized trial. The registry record does not report a separate control arm for comparing the intervention with usual care, another app, or no intervention.

The study status is completed. Its recorded start date is 2024-02-21 and its primary completion date is 2024-05-28. The primary outcome timeframe is 1 month. The official title calls it a feasibility pilot, and the brief summary says the study is intended to help researchers learn about ways to deliver information that are appealing and helpful to young adults who use mobile apps.

Several design details are not reported in the registry record, including more specific eligibility criteria, the target age range, recruitment procedures, the planned duration of individual participation, and how engagement was measured. The registry also does not report a protocol-level description of the reinforcement learning algorithm beyond its dependence on participant engagement. These absences limit how fully the intervention delivery and participant flow can be reconstructed from the registry entry alone.

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: App feasibility outcome is operationalized as % of participants who consented to the study who download the MiWaves app. Not all that consented to the study were included in study flow since not all that consented downloaded the app. Therefore, the number analyzed for app feasibility is greater than the number in the participant flow.

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 App Feasibility, measured over 1 month.

The primary outcome is named “App Feasibility.” It is operationalized as the percentage of participants who consented to the study who downloaded the MiWaves app, with a timeframe of 1 month. In plain terms, it asks whether consented participants proceeded to install or download the app, a practical indicator of uptake at the start of an app-based study.

For a binary endpoint of this kind, each eligible participant would ordinarily contribute to a count of downloads or non-downloads, and the proportion would use the stated consented population as its denominator. That is a general description of how a proportion is interpreted, not a claim about the analysis actually performed in this trial. The registry record does not provide the count who consented, the count who downloaded, the calculated percentage, or an uncertainty interval.

The record notes that not all people who consented were included in study flow because not all consented participants downloaded the app. It then gives an incomplete sentence about the number analyzed for app feasibility. Consequently, the precise endpoint denominator and the handling of participants without a download result are not reported in the registry record. The outcome may inform feasibility of app uptake, but by itself it does not quantify changes in cannabis use or establish the effect of the intervention on substance use.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a binary primary endpoint, a single-group design, and an intervention that uses repeated message or no-message decisions. It does not report an inferential analysis for the primary endpoint. The explanations below are general statistical context for understanding such a study, not a description of methods confirmed to have been used.

How a feasibility proportion is interpreted

In general, a feasibility proportion is a count of participants meeting a defined criterion divided by the population specified for that outcome. Here, the stated criterion is downloading the MiWaves app, and the stated denominator is participants who consented. The result would describe uptake in that defined population. It would not automatically describe all eligible people, all people approached, or all emerging adults who use cannabis. Those broader populations require their own denominators and sampling information.

Denominator definition is part of the endpoint, not a minor bookkeeping choice. If the denominator includes everyone who consented, those who did not download the app need to be represented in a clearly defined way for the proportion to be interpretable. The registry record mentions that not all consented participants were included in study flow because not all downloaded the app, but does not provide a complete participant flow or endpoint count. It is therefore not possible to verify how the reported endpoint population would be assembled.

Uncertainty around a proportion

In general, an estimated proportion has sampling uncertainty: another sample from a similar population could produce a different estimate. A confidence interval is one way to communicate the range of estimates compatible with the observed data and a specified statistical procedure. Its interpretation depends on the method and assumptions used. The registry record does not report a primary endpoint estimate or confidence interval, so neither the observed uptake nor its precision can be evaluated.

A proportion also needs context. A result can be useful for planning a later study, but what level of uptake would count as feasible depends on criteria established for the project. Such a threshold is not reported in the registry record. No feasibility benchmark or decision rule should be inferred from the endpoint title alone.

Single-group design and causal comparison

In general, a single-group study can describe outcomes among participants who received or were offered an intervention, but without a separate comparison group it does not provide the usual randomized contrast between intervention and control. Changes over time, if measured, can have multiple explanations, including factors unrelated to the intervention. The registry record does not provide a comparative outcome estimate for cannabis use.

Micro-randomization is a different level of assignment. The study description says that a participant is assigned to receive a message or no message twice daily, with decisions informed by engagement. Such repeated decisions may support questions about message delivery at decision points when the design, outcome, and analysis are specified. They do not by themselves establish an overall intervention-versus-control effect. The registry record does not report the decision-level analysis, the outcome attached to those decisions, or an estimate of a message effect.

Repeated decisions and dependence

In general, repeated observations from the same participant are correlated: observations from one person may be more alike than observations from different people. An analysis that treats every decision as independent can understate uncertainty if it does not account for that dependence. Appropriate methods depend on the study’s specific estimand, timing, randomization process, outcome definition, and data structure. Those analytical details are not reported in the registry record, so no particular repeated-measures method can be attributed to this pilot.

Engagement may also affect when messages are assigned, according to the description of the reinforcement learning algorithm. In general, when assignment probabilities or decision rules depend on participant information, interpretation requires knowing how those probabilities were handled in the analysis. The registry record does not report the algorithm’s detailed rules, assignment probabilities, or an analysis plan for accounting for them.

Over-dispersion and outcome structure

In general, over-dispersion arises when outcome variability exceeds what a chosen statistical model expects. For counts of events or repeated binary responses, unaccounted extra variability can make standard errors too small and confidence intervals too narrow. Whether over-dispersion is relevant depends on the actual outcome and model. The primary outcome is app download, and no fitted model or model diagnostics are reported. It would be inappropriate to assert that over-dispersion occurred or that a particular adjustment was used.

Planning calculations versus observed evidence

The registry record does not report a trial-specific sample-size justification. Planning calculations are not results: they describe assumptions under which a later study might be designed, rather than the precision or probability of success observed in this pilot. Any planning illustration based on an observed effect would require that effect and a defined comparison, neither of which is reported as a primary outcome analysis in the registry record. The stated illustration assumptions include a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect expressed as a fraction of an observed effect. Those assumptions do not describe the reported single-group study design, and no numerical planning result can be evaluated from the registry information.

The registry lists 0 analyses. It does not identify an estimation method, statistical model, confidence-interval procedure, or hypothesis test for the primary endpoint. Accordingly, statistical concepts in this explanation clarify what such methods would mean in general; they should not be mistaken for the analysis actually used in the pilot.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("JITAI")
n      <- c(137)
events <- c(122)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  JITAI  137  122  89.1  82.6  93.7  82.7  93.3  82.7  93.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 2 outcome measures and 0 analyses. It does not describe a multiplicity adjustment, a hierarchy among outcomes, or a plan for controlling the chance of false-positive findings across multiple statistical tests. Because no analyses are listed, it is not possible to determine whether any such procedure was conducted. In general, when several outcomes or repeated hypotheses are tested, multiplicity can affect how confirmatory claims should be interpreted; the appropriate approach depends on the prespecified questions and analysis plan.

Missing-data handling is also not reported in the registry record. The primary outcome description indicates that not all who consented were included in study flow because not all downloaded the app. The description does not provide a complete flow count, explain how people without a download were classified in the analysis, or state whether any other endpoint data were missing. The endpoint’s stated denominator is people who consented, but the record does not give the denominator count or a completed result.

The analysis population is not reported in the registry record. In particular, it does not specify whether the primary outcome was assessed in all consented participants, only those who downloaded the app, or another defined group. It also does not report whether the recorded enrollment of 122 corresponds to the primary endpoint population. These distinctions are important because a feasibility estimate can change with the population included in its denominator.

The registry record does not report an interim-analysis plan, a missing-data sensitivity analysis, or a statistical analysis plan. Their absence from the entry does not establish whether such plans existed outside the registry; it means they are not reported in the registry record.

12. 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 0 serious adverse events among 122 participants at risk in the JITAI group. It also reports 0 other adverse events among 122 participants at risk. These figures are counts recorded for the group; they are not comparative estimates against a separate control group, which is not described in this single-group study.

A zero count means that no events in the corresponding category are recorded in these registry figures. It does not, by itself, show that the risk is absent, nor does it establish a comparative safety conclusion. In general, the amount of information in a zero-event observation depends on the number and follow-up of participants, the event definitions, and how events were collected. The registry record gives the count at risk but does not report event definitions, collection procedures, or further safety-analysis details.

The JITAI group label refers to the intervention group recorded in the adverse event information. These safety counts should be kept distinct from the app-feasibility endpoint: they do not indicate how many participants downloaded the app. No additional serious adverse event terms are listed in the record.

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

This completed pilot establishes that the registry recorded an enrollment of 122 and evaluated a mobile-app intervention designed for emerging adults in relation to cannabis use. The primary endpoint is defined as the percentage of consented participants who downloaded MiWaves within a timeframe of 1 month. That definition makes app uptake the central reported feasibility question.

The registry record does not report the number of consented participants, the number who downloaded the app, or the resulting percentage. It therefore does not allow readers to judge how often participants proceeded from consent to download or how precise an estimate of that rate might be. Enrollment alone is not a substitute for the endpoint denominator or result.

The record also does not report an analysis showing whether cannabis use changed. The study’s title and stated purpose concern reducing cannabis use and piloting prevention methods, but a stated purpose is not an outcome result. The message/no-message micro-randomization description gives information about intervention delivery, not an estimate of the effect of messages on cannabis use.

These limitations do not show that the intervention succeeded or failed. They describe what can and cannot be concluded from the registry information: study structure, the stated feasibility outcome, enrollment, and recorded adverse event counts are available; the primary outcome estimate, its statistical uncertainty, and a comparative estimate of effects are not reported in the registry record.

In general, feasibility evidence can help inform whether an intervention can be delivered and evaluated in a larger study. Its value depends on clearly defined outcomes, denominators, participant flow, and estimates with appropriate uncertainty. The registry entry does not report enough endpoint detail to make that assessment of app uptake. Nor does it report a sample-size justification or statistical analysis plan that would clarify the intended precision or inferential goals.

14. Why This Trial Matters Statistically

Statistically, this pilot raises a useful distinction between feasibility and efficacy questions. App download is a concrete implementation outcome: it concerns whether consented participants took an initial step toward using the intervention. A cannabis-use outcome would address a different question about behavior. The registry names app feasibility as the primary outcome, but does not report its value or a cannabis-use analysis.

The case also illustrates why participant flow and denominators matter. A percentage cannot be interpreted without knowing both its numerator and denominator. Since the record says that some people who consented did not download the app, the population used for the app-feasibility endpoint is especially important. The missing count and result prevent an assessment of uptake, even though enrollment is recorded.

The intervention delivery has an additional statistical feature: message decisions are repeated within participants and informed by engagement. General analysis of this kind of design must distinguish participant-level comparisons from decision-level questions and account for dependence among repeated observations. The registry description makes the adaptive delivery feature visible, while the analyses needed to interpret it are not reported.

Finally, the record demonstrates the limits of treating a trial’s stated aim as evidence of its outcome. A study intended to reduce cannabis use is not, on that basis alone, evidence that use changed. Transparent estimates, uncertainty measures, and a clear description of the analysis population are needed for readers to assess what the observed data support. The registry entry records a completed feasibility pilot, but does not report those outcome analyses.

15. 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 most consequential limitation for interpretation is that the registry record does not report the primary app-feasibility result. It names the outcome and its timeframe but gives no download count, consent count, percentage, or confidence interval. The endpoint description also ends with an incomplete statement about the number analyzed, leaving the analysis denominator unresolved.

The study is recorded as single group, with allocation listed as NA and masking as none. There is no separate control group described for a trial-level comparison. The within-participant message or no-message assignments are described as micro-randomized, but the registry does not report an analysis of those decisions or an effect estimate for messages.

Further limitations are that the registry record does not report detailed eligibility criteria, participant demographics, recruitment information, individual follow-up, outcome measurement procedures, or the analysis population. It lists 2 outcome measures but 0 analyses, and does not report statistical methods, a sample-size justification, a multiplicity approach, or missing-data handling. These omissions restrict assessment of both the endpoint and the broader study questions.

The record reports zero serious and zero other adverse events in the JITAI group among 122 at risk, but does not give event definitions or further ascertainment details. Those counts do not provide a comparative safety estimate. These limitations describe the information available in the registry record and do not establish what may have been reported elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: An App to Reduce Cannabis Use Among Emerging Adults — NCT05824754.
• Registry record last updated 2026-10-27; status last verified 2025-06.
• Results first posted 2025-06-12.
• 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.

19. Record Summary

The registry record describes a completed single-group feasibility pilot of a just-in-time adaptive mobile intervention for emerging adults in relation to cannabis use. Enrollment is 122. The intervention description says participants were assigned to receive a message or no message twice daily, with decisions based on engagement through a reinforcement learning algorithm.

The primary outcome, “App Feasibility,” is defined as the percentage of consented participants who downloaded the MiWaves app, with a timeframe of 1 month. The download count, consent denominator, and endpoint result are not reported in the registry record. The registry lists 2 outcome measures and 0 analyses.

The JITAI group has 0 reported serious adverse events and 0 other adverse events, with 122 at risk in each count. A separate control group and a comparative outcome estimate are not reported.