Digital Mental Health Intervention for Nonsuicidal Self-Injury in Young Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Digital Mental Health Intervention for Nonsuicidal Self-Injury in Young Adults (NCT05325944), a randomized na study. The primary outcome is Alexian Brothers Assessment of Self-Injury - Methods Checklist.

2. Clinical Question

The trial asks whether a digital mental health intervention for nonsuicidal self-injury can be evaluated in a randomized feasibility study among young adults who are not currently engaged in mental health treatment. The registry describes three arms: a self-guided intervention, the intervention with low-intensity coaching, and an active control delivering non-interactive psychoeducational content through the same app interface. The primary outcome concerns the number of days of NSSI in the past month, assessed at 16 weeks.

The posted results give a mean for each arm, with lower and upper bounds. They do not provide a two-arm comparison estimate or a statistical test. The results therefore describe the trial arms but do not establish whether observed differences are distinguishable from chance variation. The registry identifies feasibility and preparation for a fully powered randomized-controlled trial as the project's primary goals.

3. Trial Design

NCT05325944 is listed as a completed, randomized, double-masked trial with a factorial model. The registry lists enrollment of 90 and three arms: self-guided digital intervention for NSSI, digital intervention for NSSI with coaching, and active control. Each arm has 30 participants in the posted primary-outcome results. The record gives the study start as 2024-08-02 and primary completion as 2025-08-18.

The study is described as a feasibility trial, not as a definitive test of treatment efficacy. Its stated purpose is to evaluate intervention and trial-procedure feasibility in preparation for a fully powered randomized-controlled trial. The three-arm structure distinguishes a self-guided intervention, the intervention with low-intensity coaching, and psychoeducational content delivered without interactive features. The registry labels the overall model factorial, but it does not report the factor structure or explain how the factorial label relates to the three listed arms.

The self-guided intervention is described as having 8 weekly modules containing psychoeducation and skill-based practice, alongside daily ecological momentary assessments. Content is delivered through an interactive conversational agent, although the registry's arm description ends mid-sentence. The coaching arm's description also ends mid-sentence; the brief summary identifies low-intensity coaching, but further details of its delivery are not reported in the registry record. The active control is described as having 8 weekly psychoeducational modules without interaction features or EMA for personalization.

The record lists a primary outcome at 16 weeks and four secondary outcomes: the Patient Health Questionnaire-9, Generalized Anxiety Disorder-7, Borderline Evaluation of Severity Over Time, and Depression Symptom Inventory - Suicidality Subscale. A full visit schedule, allocation method, masking procedures, adherence definitions, and details of the factorial design 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.

Before assignment: Prior to random assignment, participants who provided affirmative informed consent were required to have completed baseline assessments and confirm their ability to access the intervention via smartphone.

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.

6. Primary Endpoint

The registry lists the primary outcome as Alexian Brothers Assessment of Self-Injury - Methods Checklist, measured over 16 weeks.

The primary endpoint is the Alexian Brothers Assessment of Self-Injury - Methods Checklist, reported as the number of days of NSSI in the past month at 16 weeks. The registry describes the checklist as a 16-item measure assessing the frequency of common NSSI methods in the past month. Higher scores indicate engagement in NSSI on more days. The listed minimum is 0 days and the maximum is 31 days.

This outcome captures a behavioral frequency dimension: how many days in the recent month participants reported NSSI. It is not a count of all episodes, a measure of severity, or a measure of suicidal behavior. The registry separately lists suicidal ideation among the trial conditions and names a suicidality subscale among secondary outcomes, but those are not the primary endpoint.

Repeated measurement of a recent period can be useful in a feasibility study because it offers a defined outcome to collect across the intervention arms. However, the registry record does not state why this particular measure was selected, how it was administered, whether missing checklist items were scored, or whether the measure was assessed at other time points. The posted arm results give means and lower and upper bounds for the endpoint. The record does not identify the type of interval represented by those bounds, so they should not be interpreted as confidence intervals without additional documentation.

The endpoint reflects reported NSSI days within the specified past-month window. It does not by itself describe the full experience of NSSI, intervention engagement, or whether trial procedures were feasible. Those broader goals would require other measures and analysis details, which are not fully reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the trial as randomized, double-masked, and factorial. Those labels describe aspects of the design, but the record does not provide enough detail to reconstruct the allocation procedure, masking procedures, or the factors and contrasts in the factorial structure. It is therefore possible to describe the listed design features, but not to infer how randomization or masking was implemented or how a factorial analysis would have been specified.

What an arm mean describes

General statistical explanation: A mean is an average across the participants contributing data to an outcome summary. It gives a compact description of the group's results, but it does not show the distribution of individual values. For a count-like measure such as days with NSSI, participants can have quite different values while sharing the same mean. The registry reports means for the three arms and lists 30 participants per arm for these results. It does not state how many participants contributed an observed endpoint value or how missing observations were handled.

The lower and upper bounds listed with each mean convey a range around that summary, but the registry does not name the interval type or explain its construction. A confidence interval, for example, is a statistical interval intended to express uncertainty in an estimated quantity under specified assumptions. That general definition does not establish that the bounds in this record are confidence intervals. Their interpretation requires documentation that is not reported in the registry record.

Arm summaries and comparisons

General statistical explanation: A comparison analysis estimates a contrast between groups, such as a difference in their mean outcomes, and describes uncertainty in that contrast. A test or interval for the contrast addresses a different question from separate summaries within each arm. Separate arm means do not supply the uncertainty of a between-arm contrast because the comparison depends on the data and analysis across both groups. The record lists no analyses and no two-arm comparison estimates for the primary endpoint. The posted means should therefore remain descriptive rather than being treated as estimates of an intervention effect.

Random allocation can support causal comparisons when the trial is implemented and analyzed according to an appropriate plan. It does not eliminate the need to report the planned estimand, analysis population, outcome handling, and comparison method. Those details are not reported in the registry record. The factorial label also does not alone identify which main effects or interactions were to be estimated.

Why count outcomes need care

General statistical explanation: Counts can have a skewed distribution, and their variability can increase as their average increases. When variability exceeds what a chosen model expects, this is called over-dispersion. Ignoring over-dispersion can make uncertainty estimates too small and tests too optimistic. Other features, such as many zero values or repeated observations from the same person, can also matter for count data. This is general methodological context, not a statement about the distribution in this trial. The registry does not report the observed outcome distribution or a model used to analyze it.

The endpoint is recorded as days during a past-month window, with a listed minimum of 0 days and maximum of 31 days. This bounded outcome is not an unrestricted count. Appropriate analysis choices depend on the distribution, the research question, and the planned estimand. The record does not report whether the primary analysis treated the outcome as a count, used a model for a bounded outcome, or used another approach.

Power and planning

General statistical explanation: Statistical power is a planning concept describing the chance that a specified analysis will detect a prespecified effect under stated assumptions. A calculation based on an effect observed in a small feasibility study is an illustration for future planning, not observed power and not proof that the effect is real. Planning assumptions need an effect size, variability, allocation, significance level, and analysis method. The registry record does not report the trial's own sample-size justification or a power calculation. It describes the study as preparatory for a fully powered trial, which is distinct from claiming that this feasibility trial was powered to establish a treatment effect.

The record includes general assumptions for planning illustrations based on observed effects: a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the target effect defined as a fraction of the observed effect. These are described as planning assumptions, not the trial's own design. No resulting planning values are included in the registry entries used for this interpretation. Such illustrations should not be confused with observed outcomes or with a prespecified analysis of this three-arm study.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and four secondary outcomes. The secondary measures are the Patient Health Questionnaire-9, Generalized Anxiety Disorder-7, Borderline Evaluation of Severity Over Time, and Depression Symptom Inventory - Suicidality Subscale. Although the record lists six outcome measures in total, it does not identify the other measure or explain how that count relates to the named outcomes.

When a study evaluates several outcomes or multiple contrasts, the chance of at least one apparently unusual result can increase if each is tested separately without a plan for multiplicity. General statistical explanation: Prespecified outcome hierarchies, adjusted testing procedures, or a clear emphasis on estimation can help readers understand how multiple results are interpreted. The registry record does not report a multiplicity strategy, an outcome hierarchy beyond the primary label, or a statistical analysis plan. It also reports no secondary-outcome analysis estimates.

Missing-data handling is not reported in the registry record. The record does not say how incomplete questionnaires, missed assessments, or participants who did not contribute a 16-week measurement were addressed. The arm counts shown with the primary results do not establish how many endpoint observations were actually available, whether all randomized participants were included, or what assumptions were made about missing outcomes.

The analysis population is also not reported in the registry record. It is not specified whether analyses were intended to follow randomized assignment, include only participants who started the intervention, or use another population definition. Interim analyses, stopping rules, and adjustment for any interim testing are not reported either. These omissions limit assessment of how the posted descriptive summaries relate to the planned analysis and how multiple outcomes or missing observations might affect conclusions.

11. 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 lists 30 participants at risk for serious adverse events in each of the three arms and records zero serious adverse events in each arm. It also records zero other adverse events in each arm, with 30 participants at risk for those events, and zero deaths in each arm. No serious adverse-event terms are listed.

These are event counts as reported in the registry. A zero count means no event of that category was recorded in the posted results for the listed group; it does not demonstrate that an event is impossible or that there was no risk. The registry record does not describe the adverse-event collection period, definitions, monitoring procedures, or how event reports were reviewed. It also does not provide detail on non-serious events beyond the aggregate count.

The safety figures are descriptive and do not establish a comparative safety effect. With no events recorded in any arm, the posted figures do not quantify differences in event risk or provide an uncertainty estimate for such differences. The trial is described as a feasibility study, and the registry does not report a separate safety analysis. These counts should therefore be interpreted as the recorded events during the period represented in the registry, not as evidence that the intervention or control is risk-free.

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

The primary-outcome results show arm-level means for days of NSSI in the past month at 16 weeks: 4.3 in the self-guided arm, 3.1 in the coaching arm, and 4.0 in the active-control arm. Each arm has 30 participants listed. These values indicate the reported average outcome within each group; they do not, on their own, establish a difference attributable to the intervention.

No comparison estimate, test, or p-value is reported for the primary endpoint. The type of the lower and upper bounds around the means is also not identified. As a result, the precision of the arm means and the uncertainty around differences among arms cannot be fully assessed from the registry record. The findings should not be read as proof of benefit, lack of benefit, or equivalence between arms.

The trial's stated purpose is feasibility: to evaluate the intervention and trial procedures in preparation for a fully powered randomized-controlled trial. The posted endpoint summaries do not answer every feasibility question. Recruitment, retention, adherence, acceptability, completion, and prespecified feasibility thresholds are not reported in the registry record. Nor are analysis results for the four named secondary outcomes.

The record does describe the intervention and active control at a broad level, but details for the self-guided and coaching arm descriptions are incomplete. It does not explain the factorial structure, analysis population, missing-data approach, or primary comparison method. Those gaps make it difficult to connect the observed arm summaries to a specific planned statistical question.

Accordingly, the result supports a limited descriptive reading: the registry reports endpoint means and bounds for each arm, alongside zero recorded serious adverse events. It does not report the comparative analyses or broader feasibility evidence needed for stronger conclusions. A future definitive trial would need a clear analysis plan and complete reporting of its outcomes and feasibility measures to address those questions.

13. Why This Trial Matters Statistically

Feasibility trials can have an important statistical role before a larger randomized study. They can help evaluate whether an intervention and its trial procedures can be delivered and studied as intended. The registry explicitly describes this study as preparatory to a fully powered randomized-controlled trial. That purpose is distinct from establishing a definitive effect on the primary outcome.

This trial also illustrates why descriptive results and comparative analyses should be kept separate. The registry reports a mean for each arm, but does not report a contrast among arms, a test, or an analysis method. Arm summaries can describe the observed groups, while a comparative analysis addresses whether the data support a difference under a specified statistical model. Having the former without the latter limits interpretation even when the study is randomized.

The three arms raise a further design-reporting issue. The registry labels the study factorial, but the factor structure and planned contrasts are not described. Clear reporting of those elements would help readers understand whether the intended questions concern the intervention package, coaching, interaction features, or another component. Without that information, the contribution of individual elements cannot be separated from the descriptive results.

The record also highlights the importance of reporting feasibility outcomes and the statistical handling of missing data. The stated aim concerns intervention and trial-procedure feasibility, yet relevant measures and thresholds are not reported. A future study's interpretation would be stronger if it specified its analysis population, outcome definitions, missing-data procedures, multiplicity approach, and contrast estimates in advance and reported them clearly.

The posted data therefore matter less as a final comparative answer than as an illustration of the information needed to move from a feasibility study toward a definitive trial. The registry provides a defined primary measure and arm summaries, while leaving several elements of statistical interpretation unresolved.

14. 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 primary limitation is that the registry reports arm summaries but no analysis estimates for comparisons between arms. The means and lower and upper bounds are available, but the interval type is not identified. Without a contrast estimate and its uncertainty, the comparative evidence cannot be evaluated from the posted results.

There is limited detail on the endpoint's analysis. The record does not specify the distribution of observed scores, the statistical model, whether adjustment was planned, or how the primary outcome was handled if values were missing. The number of participants contributing data to each mean is not separately reported. The analysis population is not identified.

The registry's factorial designation is not accompanied by a description of the factors or planned factorial contrasts. The arm descriptions for the digital interventions are incomplete, including text that ends mid-sentence. The coaching arm is identified as including low-intensity coaching in the brief summary, but its delivery details are not reported. These gaps limit understanding of what distinguishes the intervention components and how their effects might have been evaluated.

The study is described as a feasibility trial, but feasibility measures, success thresholds, and results for intervention engagement or trial procedures are not reported in the registry record. The statistical plan, sample-size justification, multiplicity strategy, interim analyses, and missing-data procedures are also not reported. The record lists secondary outcomes but no analysis results for them.

Finally, the safety information is limited to aggregate event counts. Definitions, monitoring period, and assessment procedures are not specified. These limitations do not negate the descriptive results; they define what can and cannot be concluded from the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Digital Mental Health Intervention for Nonsuicidal Self-Injury in Young Adults — NCT05325944.
• Registry record last updated 2026-10-29; status last verified 2025-04.
• Results first posted 2026-10-29.
• 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.

18. Record Summary

NCT05325944, titled Digital Mental Health Intervention for Nonsuicidal Self-Injury in Young Adults, is listed as completed. It is a randomized, double-masked study with a factorial model and enrollment of 90. The three arms are a self-guided digital intervention, the intervention with coaching, and an active control. The project is described as a feasibility trial for young adults ages 18-25 with NSSI who are not currently engaged in mental health treatment.

The primary outcome is the Alexian Brothers Assessment of Self-Injury - Methods Checklist, measuring days of NSSI in the past month at 16 weeks. The posted arm-level means are 4.3, 3.1, and 4.0, respectively, with lower and upper bounds reported for each. The registry does not report a comparison analysis for these results. Four secondary measures are named, but their analyses are not reported. Serious adverse events, other adverse events, and deaths are recorded as zero in each arm. Details of the statistical plan, missing-data handling, analysis populations, and factorial contrasts are not reported in the registry record.