The Identity Project: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Identity Project (NCT03772522), a non-randomized na study. The primary outcome is Mean Change From Baseline in Hopelessness as Measured by the Beck Hopelessness Scale.

2. Clinical Question

The study was framed as an impact and process evaluation of transition-related support for young people formerly experiencing homelessness. Its description emphasizes whether and how an identity capital intervention delivered outside the social service sector might relate to participants’ life trajectories, poverty, and social inclusion. Those aims are not themselves findings: the registry record does not report numerical results that establish whether the intervention changed hopelessness or other outcomes.

3. Trial Design

The Identity Project was registered as a completed, parallel-group, non-randomized study sponsored by Unity Health Toronto. The registry record gives a study start date of 2018-04-01 and a primary completion date of 2019-08-06. It lists 27 participants enrolled and identifies dk Leadership as the behavioral intervention.

Participants were allocated to an immediate intervention group or a delayed intervention group. The immediate group was assessed immediately before and after the six-week intervention. The delayed group first had a six-week period without intervention, with assessments before and after that period; it then received the same intervention. The registry description says changes during the delay were compared with changes in the group that had received the intervention.

Because allocation was non-randomized, differences between groups may reflect pre-existing differences as well as any differences associated with intervention timing. The record explicitly cautions that the arms are not a like-for-like control comparison. Masking is listed as none. The registry record does not report how allocation decisions were made, whether baseline characteristics were balanced, or whether any adjustment was planned to address group differences.

The primary outcome was mean change in Beck Hopelessness Scale score. Assessments were described at six weeks before baseline for the delayed group only, at baseline, immediately after intervention, and at 3, 6, and 9 months after intervention. The record also lists secondary outcomes on vocational participation or employment, academic participation or educational attainment, and themes from participant focus groups. The detailed statistical plan and the numerical results for these 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.

Before assignment: Although out recruitment goal was 26 youth, we exceeded that goal by recruiting 27 youth.

Pre-intervention

Reasons for not completing this period:

Intervention

Reasons for not completing this period:

Post-intervention Follow-up

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 1 participant in the delayed intervention group missed the 6-week delay period and joined immediately pre-intervention.Thus, the number of participants at 6-weeks pre-baseline is 10 rather than 11. 1 participant was lost to follow-up at the 9-month follow-up. Thus the number of participants in the immediate intervention group was 7 rather than 8.

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 Mean Change From Baseline in Hopelessness as Measured by the Beck Hopelessness Scale, measured over Assessed at 6 weeks pre-baseline (for the delayed intervention group only), at baseline (immediately pre-intervention), immediately post-intervention, 3 months post-intervention, 6 months post-intervention, and 9 months post-intervention.

The primary endpoint is mean change from baseline in hopelessness, measured using the Beck Hopelessness Scale. The registry describes this as a 20-item measure of motivation, expectations, and feelings about the future, with a total score range from 0 to 20. It gives the scale’s internal consistency as α = .93 and describes score bands for levels of hopelessness. Those descriptions identify the intended construct and scoring; they do not show participants’ observed scores or changes.

A mean change summarizes the average direction and size of score change in a group over a specified period. The registry states that baseline scores would be compared with post-intervention scores and that changes over time would describe the general trajectory of hopelessness. Repeated assessments can help distinguish an immediate pattern from later follow-up patterns, but interpretation depends on the reported estimates, their uncertainty, and how missing observations are handled.

The endpoint is relevant to the stated evaluation because hopelessness concerns expectations and feelings about the future, while the intervention description emphasizes identity capital and career direction. That conceptual connection does not establish that a change in the scale was caused by the intervention. No numerical mean changes, group contrasts, confidence intervals, or hypothesis-test results for this endpoint are reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Mean change is a summary for a continuous outcome. It represents the average difference between a participant’s score at a later assessment and the score used as that participant’s baseline, aggregated across the participants included in the analysis. The registry identifies the primary outcome as a mean and its dispersion as a standard deviation. A standard deviation describes variation among individual scores; it is not the same as uncertainty in the estimated mean. The registry record does not provide the observed means or standard deviations for the primary outcome.

A confidence interval is a statistical range used to express the precision of an estimate under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval indicates less precision. An interval that includes values consistent with no change may leave substantial uncertainty about whether a change occurred. The registry record does not report a confidence interval for the primary outcome, so the precision of any estimated change cannot be assessed from this record.

A hypothesis test evaluates how compatible the observed data are with a specified null hypothesis under a statistical model. A p-value is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. Interpretation should consider the estimated effect and its uncertainty, not a test result alone. No primary-outcome p-value or test result is reported in the registry record.

The study uses a parallel-group structure: participants were allocated to an immediate or delayed intervention group, and the groups were observed over the same general study period. The comparison described for the delay period concerns changes in outcomes between groups. In general, comparing changes can account for baseline outcome levels in a simple way, but it does not remove confounding caused by non-random group allocation. More advanced adjustment can address measured differences, though it cannot automatically remove differences that were not measured. The registry record does not report a specific adjusted model or covariates.

The record describes measurements at multiple times. Repeated observations from the same participant are correlated because they come from that participant; they should not generally be treated as independent observations. General methods for longitudinal data account for within-person correlation and can use the pattern of available measurements. The registry record does not specify whether such a model was used, how time was represented, or which time-point contrast was primary for analysis.

Standard deviation is a common way to describe spread when a continuous outcome is summarized by a mean. If outcomes are strongly skewed or have unusual distributions, a mean and standard deviation may not fully describe their shape. Over-dispersion, in general, means that observed variability is greater than a model expects; ignoring it can make uncertainty estimates too optimistic. Those are general statistical considerations, not claims about the distribution of scores in this study. The registry record does not report the observed score distribution or diagnostic checks.

Non-randomized allocation changes what a between-group contrast can support. Randomization, when used, aims to balance measured and unmeasured factors across groups on average. Here, allocation was non-randomized, and the record states that the arms are not a like-for-like control comparison. A difference in mean change therefore should not automatically be interpreted as the causal effect of the intervention. Baseline characteristics, allocation procedures, confounding control, and sensitivity analyses would help readers judge the comparison, but these details are not reported in the registry record.

The registry record lists 9 outcome measures and 25 analyses. When many outcomes, time points, or statistical tests are examined, the chance of at least one apparently notable result can increase. Multiplicity methods can control or describe this issue, but the appropriate approach depends on the analysis goals and the prespecified hierarchy of outcomes. The record does not report whether outcomes were prioritized, whether tests were adjusted, or how the 25 analyses relate to the 9 outcome measures.

The sample-size assumptions describe planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. These illustrations are planning calculations, not observed power and not evidence that the study had a particular probability of detecting an effect. The record does not report the trial’s own sample-size justification or planning assumptions. Numerical planning outputs are not reported in the registry record.

The record describes an assessment schedule and reports some participant-flow details, but it does not provide the primary analysis population, a complete accounting of observations at each assessment, or the statistical method used for missing data. General approaches include analyzing available observations, imputing missing values under stated assumptions, or conducting sensitivity analyses. These approaches answer different questions and rely on different assumptions. The methods used in this study are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 9 outcome measures and 25 analyses. It does not report a multiplicity adjustment, a hierarchy among outcomes, or which analyses were designated as primary or exploratory beyond identifying hopelessness as the primary outcome. When multiple outcomes, time points, or comparisons are examined, individual test results can be harder to interpret in isolation. A prespecified analysis plan and clear reporting of all planned contrasts help readers distinguish confirmatory analyses from exploratory ones. Those details are not reported in the registry record.

Some missing-observation details are recorded for the primary outcome. One participant in the delayed intervention group missed the six-week delay period and joined immediately pre-intervention; the number of participants at six weeks before baseline is therefore reported as 10 rather than 11. One participant was lost to follow-up at the 9-month follow-up. The record does not provide a complete count of analyzed participants at every assessment or explain how these missing observations were handled.

Missing-data methods matter because the estimate can be affected if participants with missing outcomes differ from those with observed outcomes. General approaches rely on assumptions about why data are missing; those assumptions should be described and, where appropriate, examined through sensitivity analyses. The registry record does not report the missing-data assumptions, imputation method, or sensitivity analyses.

The record also does not report which participants were included in each primary analysis or whether the analysis followed an intention-to-treat principle. Without a defined analysis population and a clear account of missing outcomes, readers cannot determine precisely which observations contributed to the primary estimate. No numerical primary-outcome estimate is reported in the registry record.

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 adverse-event figures list the immediate intervention group with 8 participants at risk for serious adverse events and 0 affected, and 8 at risk for other adverse events and 0 affected. For the delayed intervention group, the record lists 11 participants at risk for serious adverse events and 0 affected, and 11 at risk for other adverse events and 0 affected. Deaths are listed as 0 in both groups, and no top serious-event terms are listed.

These figures describe what is recorded for the participants counted at risk; they do not show that every possible event was captured or that the intervention has been established as safe. The registry record does not provide further detail on adverse-event definitions, collection procedures, observation time, or whether events were assessed as related to the intervention. The listed figures should therefore be read as the adverse-event counts recorded, not as a complete safety evaluation or a comparison capable of establishing a causal safety difference.

The groups were non-randomized, and the registry record cautions that they are not a like-for-like control comparison. The adverse-event counts alone do not establish that one group had greater or lesser risk. No medical conclusion should be drawn from the absence of recorded affected participants in these figures.

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

The registry record describes an evaluation of a six-week identity capital intervention for young people transitioning out of homelessness. It identifies hopelessness, measured with the Beck Hopelessness Scale, as the primary outcome and describes assessments extending through 9 months. It also lists vocational, academic, and focus-group outcomes. These details show what the study set out to assess; they do not establish what changed.

No numerical primary-outcome results are reported in the registry record. Without observed mean changes, a between-group estimate, and its uncertainty, the direction and size of any difference cannot be judged. The record therefore does not support a conclusion that hopelessness improved, worsened, or remained unchanged, nor does it establish an effect on vocational participation, education, or qualitative themes.

The non-randomized design is another important boundary. The immediate and delayed groups may have differed in ways that influenced their outcomes. A comparison between them would need to be interpreted in light of those possible differences and the analysis methods used. The registry record does not report enough detail to evaluate adjustment for confounding or the robustness of any result.

Even when an estimate is available, statistical uncertainty and practical meaning are distinct. A confidence interval describes precision under model assumptions; it does not by itself establish whether a change matters to participants. The registry record gives score ranges and interpretive bands for the scale, but it does not provide observed changes or a threshold for a meaningful change in this study.

The record’s completion status and stated evaluation aims do not substitute for reported outcome findings. The appropriate conclusion from this record is limited to the study design, measurement plan, and listed adverse-event counts. Whether the intervention was associated with changes in the measured outcomes is not reported in the registry record.

13. Why This Trial Matters Statistically

This trial raises a statistical question that is common in evaluations of social and behavioral interventions: how to interpret changes when allocation is not randomized and participants are assessed repeatedly. A comparison of immediate and delayed intervention timing can be informative, but the groups may differ before the comparison begins. Recognizing that design feature is essential to avoid treating a group contrast as though it were automatically causal.

The use of a continuous measure of hopelessness also makes estimation and precision important. Mean changes can communicate the average pattern, while standard deviations describe participant-level variation. Confidence intervals would help readers understand the uncertainty around group changes or contrasts. The registry record identifies these measurement concepts but does not report the numerical estimates needed to judge them.

The assessment schedule includes an initial delay period and follow-up after intervention. Repeated outcome measurements can describe trajectories rather than only a single before-and-after contrast, but they require clear reporting of which time points and contrasts were analyzed. Missing observations can affect those trajectories, especially at later follow-up. The registry record notes a missed delay-period assessment and a participant lost at the 9-month follow-up, but does not report the complete analysis approach.

The study also includes quantitative and qualitative outcomes. This breadth can help address questions about both measured change and participant perspectives, while increasing the importance of a transparent analysis plan when many outcomes and analyses are considered. The record lists 9 outcome measures and 25 analyses but does not explain how multiplicity was handled.

These statistical issues matter because a useful evaluation requires more than an intervention description: readers need estimates, uncertainty, analysis populations, and enough methods detail to judge the comparison. The registry record documents an important evaluation question, but its reported information does not permit conclusions about outcome effects.

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 central limitation for interpreting results is that numerical outcome estimates are not reported in the registry record. The primary endpoint is described, but observed mean changes, group contrasts, confidence intervals, and test results are absent. Consequently, the magnitude, direction, and precision of any outcome pattern cannot be evaluated from this record.

Allocation was non-randomized, and the registry record explicitly warns that the study arms are not a like-for-like control comparison. The record does not describe the allocation process, baseline group characteristics, or any covariate adjustment. It is therefore not possible to determine how much the groups differed at baseline or how potential confounding was addressed.

The registry describes multiple assessments and reports limited missingness information: one participant in the delayed group missed the six-week delay period and joined immediately pre-intervention, and one participant was lost to follow-up at 9 months. The full number analyzed at each time point, the primary analysis population, and the method used to handle missing data are not reported in the registry record.

The record lists 9 outcome measures and 25 analyses, but does not report a full statistical analysis plan or any multiplicity procedure. It also does not provide the trial’s own sample-size justification, the basis for choosing enrollment, or details of interim analyses. Those matters are not reported in the registry record.

For adverse events, the listed counts show zero affected participants in the recorded groups, but event definitions, surveillance methods, relatedness assessment, and observation details are not reported. The figures cannot alone establish a comprehensive safety profile. Together, these limitations restrict interpretation to the stated design and recorded counts rather than conclusions about efficacy or safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: The Identity Project — NCT03772522.
• Registry record last updated 2023-06-22; status last verified 2023-06.
• Results first posted 2020-05-12.
• 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.

18. Record Summary

The Identity Project is recorded as a completed, non-randomized, parallel-group evaluation sponsored by Unity Health Toronto. The registry lists 27 participants enrolled, no masking, and a six-week behavioral intervention delivered through dk Leadership for young people transitioning out of homelessness.

The primary outcome was mean change in Beck Hopelessness Scale score, with assessments described from the delayed group’s pre-baseline period through 9 months after intervention. Secondary outcomes addressed vocational and academic participation and themes from focus groups. The registry lists 9 outcome measures and 25 analyses.

Adverse-event figures list zero serious events, zero other events, and zero deaths in the recorded groups. The record does not report numerical primary-outcome results, the full statistical analysis plan, multiplicity handling, or the method used for missing data. Its design description and recorded counts do not establish outcome effects or a comprehensive safety conclusion.