This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Controlled Trial of Brexpiprazole For The Treatment of Co-occurring Schizophrenia and Substance Use Disorder (NCT03526354), a randomized phase 4 study. The primary outcome is Timeline Follow Back (TLFB).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03526354 |
| Phase | Phase 4 |
| Condition | Schizophrenia, Schizoaffective Disorder, Substance Use Disorders |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 50 |
| Arms in the primary outcome | 2 |
| Lead sponsor | University of Massachusetts, Worcester |
| Results posted | Yes |
2. Clinical Question
Population
The trial concerns people with co-occurring schizophrenia and a current substance use disorder. The registry lists schizophrenia, schizoaffective disorder, and substance use disorders as conditions. The brief summary describes alcohol, cocaine, heroin, or cannabis use and a four-site feasibility investigation.
Intervention
The experimental group was assigned to switch to brexpiprazole. The arm description specifies brexpiprazole 4mg daily for 12 weeks.
Comparator
The treatment-as-usual group was to stay on its current antipsychotic medication for 12 weeks. The registry describes random assignment to the experimental or control group.
Primary question
The primary outcome was change in weekly substance-use days between Week 0, baseline, and Week 12, end of study. The TLFB outcome counts days of use for a participant's primary substance. The registry says a decrease represents an improved outcome. The posted data give an estimate for each arm, but a two-arm comparison is not estimable in the registry record.
3. Trial Design
This is a randomized, parallel-group, Phase 4 trial, marked completed in the registry. The masking field says none, so the record describes an unmasked study. Its official title identifies a multi-center controlled trial of brexpiprazole for co-occurring schizophrenia and substance use disorder. The brief summary describes a 12-week feasibility investigation.
The experimental arm received brexpiprazole 4mg daily for 12 weeks. Participants in the treatment-as-usual arm were to remain on their current antipsychotic medication for 12 weeks. The record describes random assignment between these groups. It names the University of Massachusetts, Worcester, as sponsor and lead site, and lists Massachusetts General Hospital and the University of North Carolina at Chapel Hill as subsites. The summary also describes the study as involving four sites, while its site description names the lead site and two subsites.
Several enrollment figures appear in the record and are not reconciled there. The enrollment field lists 50. The brief summary says thirty-nine participants will be randomly assigned, and also says investigators expect to enroll 80 subjects across 3 sites. Posted primary-outcome arm counts are 21 and 18. These are distinct statements in the record and should not be treated as interchangeable estimates of enrollment.
The recorded start date is 2018-06-19, and the primary completion date is 2024-02-29. The completion status does not by itself establish that every planned analysis or outcome was reported. A protocol description beyond the registered arm assignments and outcomes is 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
| Stage | Experimental | Treatment as Usual |
|---|---|---|
| STARTED | 21 | 18 |
| COMPLETED | 15 | 14 |
| NOT COMPLETED | 6 | 4 |
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.
| Characteristic | Unit | Experimental | Treatment as Usual | Total |
|---|---|---|---|---|
| Participants analysed | 21 | 18 | 39 | |
| Age, Continuous | years | 36.1 (SD 12.0) | 39.8 (SD 11.2) | 37.8 (SD 11.6) |
| Sex: Female, Male: Female | Participants | 5 | 4 | 9 |
| Sex: Female, Male: Male | Participants | 16 | 14 | 30 |
| Race/Ethnicity, Customized: Caucasian | Participants | 8 | 9 | 17 |
| Race/Ethnicity, Customized: African American | Participants | 9 | 10 | 19 |
| Race/Ethnicity, Customized: Other | Participants | 4 | 0 | 4 |
| Region of Enrollment: United States | participants | 21 | 18 | 39 |
6. Primary Endpoint
The registry lists the primary outcome as Timeline Follow Back (TLFB), measured over 12 weeks.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | score on a scale |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the change from Week 0 baseline to Week 12 in the number of days per week that participants used a substance. It is measured using the Timeline Follow Back scale, which asks about use of alcohol, marijuana, cocaine, and/or heroin during the past week. The scale runs from 0 to 7, where 0 means no substances that week and 7 means substance use every day.
The record says that the outcome is based on a participant's primary substance when more than one substance is used; it defines this as the substance used most frequently. Although the TLFB was assessed weekly, the posted endpoint is the change between baseline and the end of the study. A decrease in days of use is designated as an improved outcome in the registry description.
Tracking days of use gives a continuous measure of frequency over a week and allows change to be summarized on the scale. The registry reports least-squares mean changes and 95% confidence intervals for the two arms. It does not report a two-arm comparison for this endpoint, so the posted arm estimates alone do not quantify an estimated treatment-group difference.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome estimates are changes in weekly substance-use days, not endpoint values at Week 12 alone. The experimental arm's least-squares mean change is -1.765, with a 95% confidence interval from -3.117 to -0.413. The treatment-as-usual arm's estimate is -0.532, with a 95% confidence interval from -1.91 to 0.846. Under the registry's outcome definition, a decrease in days of use indicates an improved outcome. Both point estimates are decreases, while the interval for treatment as usual includes zero.
These are separate within-arm estimates. A treatment effect question requires a direct comparison between arms, and the registry record says values are missing or not estimable for a two-arm comparison of this continuous endpoint. No between-arm estimate or corresponding confidence interval is posted. The arm estimates should therefore not be substituted for a direct randomized comparison, and their relative magnitudes do not establish that one assignment produced a different change from the other.
The experimental-arm interval does not include zero, while the treatment-as-usual interval does. That describes the uncertainty around each posted arm estimate under its reported interval. It does not resolve the missing between-arm comparison. The registry lists zero posted analyses. It does not report a p-value for the primary comparison, a model specification, or a sponsor's primary comparative estimate. Without those comparative results, the amount and precision of any difference between assignments cannot be characterized from the registry record.
The posted arm counts are 21 for the experimental group and 18 for treatment as usual. The brief summary also refers to thirty-nine randomly assigned participants, whereas the enrollment field lists 50 and another statement says investigators expect to enroll 80. These counts and expectations are reported differently in the record; the registry does not explain their relationship. The two outcome counts describe the arms contributing to the posted primary estimates, not necessarily all enrollment or all trial participants.
Statistical precision is conveyed by the confidence intervals: narrower intervals generally indicate more precision, while wider intervals reflect greater uncertainty. That is a general interpretation of interval estimates. The reported intervals are compatible with a range of underlying arm-level changes, and the control-arm interval extends across no change. They should not be read as a range of individual participant responses or as a probability that a particular result is true. The missing direct comparison is the central limitation for interpreting the primary question.
The registry lists four secondary outcomes, including substance spending, symptom ratings, clinical severity, and quality of life. Their outcome estimates are not included in the record details summarized for interpretation. Their presence as registered outcomes does not supply evidence about their results. Overall, the posted primary figures describe arm-level change with uncertainty, but they do not provide an estimable randomized contrast.
9. Statistical Methods Explained
The primary outcome is continuous: it records the number of substance-use days per week on a scale from 0 to 7 and reports change between baseline and Week 12. The registry's statistical summary uses least-squares means. In general, a least-squares mean is a model-based group mean, often adjusted for variables included in an analysis model. The model and any adjustment variables used for these estimates are not reported in the registry record, so no more specific interpretation of adjustment is possible.
What do the confidence intervals describe?
The registry reports 95% confidence intervals around each arm's least-squares mean change. In general, a confidence interval expresses the uncertainty of an estimate under the statistical procedure used to construct it. It is not a range that contains most individual participants' outcomes, and it does not itself state the probability that a hypothesis is true. The posted intervals are -3.117 to -0.413 for the experimental arm and -1.91 to 0.846 for treatment as usual. The former does not include zero; the latter does. These intervals describe arm-level changes, not the difference between the arms.
A confidence interval for a direct between-arm contrast would answer a different question: how precisely the randomized groups differ in average change. That contrast is not estimable in the registry record. It is therefore not appropriate to infer a comparative effect by treating the two arm-specific intervals as though they were an interval for a group difference. A direct comparison can have uncertainty that is not represented by either individual interval.
Why is the direct comparison important?
Random assignment is intended to support a comparison between the assigned groups, subject to the study conduct and analysis. A group-specific change can also reflect factors common to participants in that group, changes over time, or the composition of those with observed data. The direct randomized contrast is the statistic needed to assess the difference in changes between assignments. The record explicitly says that values are missing or not estimable for a two-arm comparison of this continuous endpoint, and it lists zero analyses. No reason for the missing comparison is reported in the registry record.
The endpoint description says that the primary substance was used when a participant used more than one substance, and that this was the substance used most frequently. A primary-substance outcome is a defined summary rather than a combined count across all substances. The registry does not give details about how ties, changing primary substances, or incomplete weekly reports were handled. Those procedures are not reported in the registry record.
What do planning sample sizes mean?
In general, sample-size calculations estimate how many participants would be needed under specified assumptions to detect a chosen effect with a chosen probability of detection. They are planning illustrations, not evidence that a trial had that probability of detecting an effect after it was completed. The sample-size assumptions described for this record use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a stated fraction of the observed effect. Any such counts are participants per arm for a specified planning scenario, not observed trial outcomes or observed power. The registry's own design justification and the numeric planning scenarios are not reported in the registry record details.
The record does not state how the observed effect used for planning was defined, or provide a trial-specific sample-size justification in the information available. Because the direct primary comparison is not estimable, the registry entries also do not establish a comparative effect from which to evaluate the planning scenarios. A planning illustration should not be interpreted as confirming that the completed trial was adequately powered or as explaining why the recorded enrollment differs from the figures in the summary.
What cannot be recovered from the posted methods?
The registry record does not report the statistical model used for the least-squares means, assumptions checked for that model, handling of missing observations, analysis population, or a prespecified method for the primary contrast. It also does not report whether covariates, site, or repeated weekly measurements were included in the primary analysis. The fact that TLFB assessments occurred weekly does not establish that the weekly observations were all included in the model; the endpoint description identifies the baseline-to-Week-12 change as the reported outcome.
General statistical principles can explain what a method would do, but they cannot fill in undocumented choices. For example, repeated-measures models can use multiple observations while accounting for their correlation within participants, but the registry record does not say that such a model was used. Likewise, assumptions about the distribution of continuous outcomes or methods for handling incomplete data cannot be inferred from the least-squares mean label alone. These missing details matter when judging how the estimates were obtained and how broadly they apply.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists four secondary outcomes: amount of money spent on substances, Positive and Negative Symptom Scale, Clinical Global Impression - Severity, and Heinrichs Carpenter Quality of Life Scale. It reports six outcome measures in total, but provides no posted analyses in the registry record. Results for the secondary outcomes and details of their analysis are not reported in the registry record.
When a study tests multiple outcomes or multiple hypotheses, the chance of at least one apparently unusual result can increase. General methods for addressing this include prespecifying a primary outcome, adjusting significance thresholds, or treating secondary analyses as exploratory. The registry information does not report a multiplicity plan, hierarchy among outcomes, or whether any adjustment was used. No conclusion about control of multiplicity can be drawn from the list of outcomes alone.
The record gives arm-level primary outcome estimates but no estimable two-arm comparison. It does not report the analysis population or whether participants were analyzed according to their random assignment. Missing-data methods, reasons for missing observations, and the amount of missing primary outcome data are not reported in the registry record. The arm counts for the posted outcome are 21 and 18, but these alone do not show how many randomized participants had incomplete assessments or how any incomplete records were handled.
The weekly assessment schedule and the baseline-to-Week-12 endpoint are stated, but the record does not describe how intermediate assessments contributed to analysis. It also does not say whether analyses were planned at interim time points. Those details are not reported in the registry record and should not be assumed from the completed status or the number of registered outcomes.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Experimental | 0 | 21 | 0.0% |
| Treatment as Usual | 0 | 18 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Experimental | 0 of 21 |
| Treatment as Usual | 0 of 18 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Experimental | 10 of 21 |
| Treatment as Usual | 7 of 18 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Experimental | Treatment as Usual |
|---|---|---|---|
| Cold-like symptoms | General disorders | 2 of 21 | 3 of 18 |
| Nausea | General disorders | 3 of 21 | 2 of 18 |
| Fatigure/Tiredness | General disorders | 3 of 21 | 0 of 18 |
| Dizziness | General disorders | 2 of 21 | 1 of 18 |
| Diarrhea | Gastrointestinal disorders | 2 of 21 | 0 of 18 |
| Trouble Sleeping | General disorders | 2 of 21 | 0 of 18 |
| Restlessness/Irritability | General disorders | 1 of 21 | 1 of 18 |
The registry reports serious adverse events separately from other adverse events. For the experimental group, 0 serious adverse events were reported among 21 at risk; for treatment as usual, 0 were reported among 18 at risk. The record also lists 0 deaths in each group. These are the counts posted for the recorded safety categories and trial groups.
For other adverse events, the record reports 10 among 21 at risk in the experimental group and 7 among 18 at risk in treatment as usual. The registry does not provide the event terms in the registry record for this interpretation, so the nature, severity, duration, or clinical importance of those other events cannot be described. The zero serious-event counts do not establish that no events occurred outside the registry's recorded categories, nor do they establish equivalence in safety between the groups.
These are descriptive counts, not a comparative safety analysis. No statistical comparison or confidence interval for adverse events is reported in the registry record. The numbers at risk are the posted group counts, and the record does not explain whether they represent everyone randomized or a safety population defined in another way. A reader should therefore interpret the figures as the registry's event counts, not as a complete account of each participant's experience.
12. What the Result Does, and Does Not, Mean
The registry reports a randomized comparison of switching to brexpiprazole versus remaining on current antipsychotic treatment, with a primary measure based on change in weekly substance-use days. The experimental arm's posted mean change is -1.765, and the treatment-as-usual arm's is -0.532. The record defines a decrease as improvement on this outcome. Those estimates provide a descriptive account of the change reported for each group.
They do not establish the difference between the assigned groups. The registry record identifies the continuous two-arm comparison as missing or not estimable, and no analysis is posted. The group-specific intervals are not a substitute for a contrast and cannot answer whether the randomized groups differed. As a result, the primary comparative question remains unresolved in the registry record.
The posted outcome is limited to days of use of a participant's primary substance, rather than a combined measure of every substance used. It is a change from baseline to Week 12, even though assessments were made weekly. The endpoint does not, by itself, describe other dimensions of substance use or the secondary outcomes. The record lists outcomes concerning spending, symptoms, clinical severity, and quality of life, but their results are not reported in the record details.
The results also do not provide a complete account of the trial's conduct or analysis. The statistical model, missing-data procedures, analysis population, and multiplicity approach are not reported. Enrollment information is presented with differing figures, and the relationship among them is not explained. These issues limit what can be inferred about precision, the handling of all enrolled participants, and the scope of the posted estimates.
In general, a completed trial status means the registry marks the study as completed; it does not imply that every planned outcome or analysis is available. The appropriate reading is therefore narrow: the record provides arm-level primary outcome estimates and safety counts, but not an estimable primary group comparison. It does not support a conclusion about the comparative effect of the assignments.
13. Why This Trial Matters Statistically
This trial is statistically instructive because it illustrates the distinction between reporting change within each randomized group and estimating a difference between groups. The primary endpoint has a point estimate and confidence interval for each arm, but the direct two-arm comparison is not estimable in the registry record. For a randomized trial, that comparative estimate is central to the question of how outcomes differed between assigned groups. Arm-level numbers alone do not complete that analysis.
The record also shows why outcome definitions matter. The primary measure is weekly days of use for the most frequently used substance, expressed as change from baseline to Week 12. A decrease is designated as improved. Readers need to keep that specific outcome distinct from other possible measures of substance use and from the registered secondary measures of spending, symptoms, severity, and quality of life.
Uncertainty needs to be matched to the statistic being discussed. The 95% confidence intervals describe the arm-specific mean changes. They do not quantify uncertainty in the difference between the groups. When a comparative estimate is absent, it is not possible to describe its magnitude or precision from the separate intervals alone. This remains true even though randomization is recorded as the allocation method.
Finally, differing enrollment statements and absent details about analysis populations, missing observations, and statistical methods affect interpretability. The registry lists a completed trial and primary outcome estimates, but not a posted primary comparative analysis. That distinction is useful beyond this study: a record can contain outcome values without containing the analysis needed to answer its central comparative question.
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 record does not provide an estimable two-arm comparison for the continuous primary endpoint. The posted means and intervals are arm-specific, so they cannot establish the magnitude or precision of a difference between randomized assignments. The reason the comparison is missing or not estimable is not reported in the registry record.
There are unresolved differences among the enrollment statements: enrollment is listed as 50, the summary says thirty-nine participants will be randomly assigned, and another statement says investigators expect to enroll 80. The summary describes four sites but names a lead site and two subsites, while its enrollment expectation refers to 3 sites. The record does not reconcile these figures or descriptions. The posted arm counts for the primary outcome are 21 and 18.
The record does not report the model details, analysis population, missing-data handling, multiplicity plan, or numeric interim-analysis information. The results of the listed secondary outcomes are not reported in the registry record. Further limitations specific to participant characteristics, adherence, substance-use patterns, and outcome completeness cannot be assessed from the registry entries. No additional details about these issues should be inferred from the brief trial description.
The safety information is also limited to the posted counts. Serious adverse event counts and deaths are zero in both groups, while counts for other adverse events are given without event terms in the available record details. These counts do not provide a comparative safety estimate, and they do not describe the nature of the other events.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Controlled Trial of Brexpiprazole For The Treatment of Co-occurring Schizophrenia and Substance Use Disorder — NCT03526354.
- Registry record last updated 2025-05-23; status last verified 2025-05.
- Results first posted 2025-05-23.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The record describes a completed, randomized, unmasked, parallel-group Phase 4 trial of brexpiprazole versus treatment as usual for co-occurring schizophrenia and substance use disorder. The experimental arm was assigned brexpiprazole 4mg daily for 12 weeks; the comparator remained on current antipsychotic medication for 12 weeks.
The primary outcome was change from baseline to Week 12 in weekly days of use of a participant's primary substance, measured using the Timeline Follow Back scale. Arm-level estimates and 95% confidence intervals are posted, but a two-arm comparison is not estimable in the registry record. Serious adverse events and deaths are listed as zero in both groups; other adverse event counts are 10 and 7. Enrollment and site statements differ within the record and are not reconciled. Details of the statistical model, missing-data handling, and secondary outcome results are not reported in the registry record.