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 Family Psychoeducation for Adults With Psychotic Disorders in Tanzania (NCT04013932), a randomized na study. The primary outcome is Proportion of Patient Participants Who Experience Illness Relapse.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT04013932 |
| Phase | NA |
| Condition | Schizophrenia Spectrum and Other Psychotic Disorders, Mental Disorders, Severe |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 66 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Duke University |
| Results posted | Yes |
2. Clinical Question
Population
Adults with psychotic disorders and their relatives were the intended participants. The registry lists schizophrenia spectrum and other psychotic disorders, as well as severe mental disorders, among the conditions studied. The trial enrolled 66 participants. The primary outcome concerns patient participants, not caregivers.
Intervention
The intervention was KUPAA, a culturally tailored family psychoeducation model, delivered alongside standard of care. The stated aim was to pilot a model that could accommodate traditional and biomedical ideas about mental illness and include relatives as co-facilitators. The intervention description includes joining sessions with a provider, a group educational workshop, and subsequent weekly family psychoeducation.
Comparator
The comparison group received standard of care. The randomized, parallel design assigned participants to the KUPAA intervention plus standard of care or to standard of care alone. Masking was listed as none.
Primary question
The primary outcome was the proportion of patient participants experiencing illness relapse from immediate post-intervention to endline, approximately 6-7 months after the intervention. Relapse could be a hospitalization or a non-hospitalized new illness episode with symptom recurrence after one month of controlled symptoms. The posted proportions were 0.27 in the KUPAA group and 0.45 in the control group, each with a 95% confidence interval. The registry record does not provide a usable inferential comparison between those groups.
3. Trial Design
This was a randomized, parallel-group pilot clinical trial, listed as completed. The registry identifies Duke University as the sponsor. The study title places the work in Tanzania, and its stated purpose was to pilot a culturally tailored family psychoeducation model for adults with psychotic disorders and their relatives. The record lists phase as not applicable and masking as none.
The intervention arm received KUPAA plus standard of care; the control arm received standard of care. The intervention description says patients would first participate in 1-2 joining sessions with a provider, then a 1-day group educational workshop, followed by weekly family psychoeducation. The groups were described as composed of approximately 6 patients joined by their 6 matched caregivers. The registry record does not report the number or duration of weekly sessions, details of treatment adherence, or whether every participant received each planned component.
Enrollment was 66. The outcome summaries list 33 participants in the KUPAA arm and 31 in the control arm. The primary-outcome description notes that data were not collected on two Control participants. Those details identify missing primary-outcome data, but the registry record does not state why those data were unavailable or how they were handled analytically. Safety summaries list 33 at risk in each arm; those counts should not be assumed to be the same analysis population as the primary outcome.
The registry lists a study start date of 2019-08-06, a primary completion date of 2020-10-27, and a completed status. Randomization is reported, but the method used to generate or conceal the allocation is not reported in the registry record. Likewise, recruitment procedures, site-level details, protocol deviations, and the statistical analysis plan are not reported in the registry record. The design therefore establishes a randomized comparison, while leaving important implementation and analysis details unavailable.
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 | KUPAA Intervention + Standard of Care | Control - Standard of Care |
|---|---|---|
| STARTED | 33 | 33 |
| COMPLETED | 33 | 28 |
| NOT COMPLETED | 0 | 5 |
Analysis population for the primary outcome, as defined in the registry: Data not collected on two Control participants.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
Population: The primary analysis population is the patients and that is what is reported in the tables (not caregivers). Patients could participate in the intervention arm even if a caregiver stopped attending group sessions. The intervention is focused on patient outcomes (disability, quality of life, and relapse).
| Characteristic | Unit | KUPAA Intervention + Standard of Care | Control - Standard of Care | Total |
|---|---|---|---|---|
| Participants analysed | 33 | 33 | 66 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 33 | 33 | 66 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age, Continuous | years | 33.64 (SD 8.52) | 32.45 (SD 7.94) | 33.05 (SD 8.19) |
| Sex: Female, Male: Female | Participants | 13 | 9 | 22 |
| Sex: Female, Male: Male | Participants | 20 | 24 | 44 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 33 | 33 | 66 |
| Race (NIH/OMB): White | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Region of Enrollment: Tanzania | Participants | 33 | 33 | 66 |
| 4 or more years since illness onset | Participants | 21 | 22 | 43 |
6. Primary Endpoint
The registry lists the primary outcome as Proportion of Patient Participants Who Experience Illness Relapse, measured over Relapse is measured from immediate post-intervention to endline (~6-7 months post-intervention).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | proportion of participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the proportion of patient participants who experienced illness relapse. The registry defines relapse as hospitalization or a non-hospitalized new illness episode involving recurrence of symptoms after one month of controlled symptoms. The stated measurement window runs from immediate post-intervention to endline, approximately 6-7 months after the intervention.
This endpoint is intended to capture a clinically consequential worsening of illness over follow-up, while allowing an event to qualify even if it did not lead to hospitalization. Including non-hospitalized relapse means the outcome is broader than hospital admission alone. That breadth can be useful when hospital use is only one way an illness episode may be recognized, though the record does not report how events were assessed, adjudicated, or verified.
A proportion describes the share of participants in a group who experienced the event during the defined window. The registry reports 0.27 for KUPAA plus standard of care and 0.45 for standard of care, with 95% confidence intervals. These are group-level estimates, not a complete account of the timing or duration of events. The endpoint does not by itself describe symptom severity, functioning, caregiver experience, or other outcomes. Although the record lists three outcome measures, it does not identify additional posted secondary results in the information available for this analysis.
7. Results Posted in the Registry
- Control - Standard of Care: 0.0% (0/31) versus 0.0% (0/33) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.
Risk difference: Control - Standard of Care
95% CI: 0.0 to 0.0 percentage points
Recalculated from the posted summary statistics
8. Statistical Interpretation
The posted primary-outcome estimate was 0.27 in the KUPAA intervention group, with a 95% confidence interval from 0.13 to 0.46. In the control group, the estimate was 0.45, with a 95% confidence interval from 0.27 to 0.64. These figures describe the reported proportions and the uncertainty around each group’s estimate. They are not, by themselves, a formal estimate of the between-group effect.
The point estimates differ numerically, but a numerical contrast should not be treated as a demonstrated treatment effect. The registry’s comparison entry reports a risk difference of 0.0 with limits of 0.0 and 0.0, and its accompanying values are also zero. This entry does not provide an interpretable comparison consistent with the nonzero group proportions. The record lists no analyses, and it does not report a usable inferential test or model for the primary outcome. The zero-valued comparison therefore cannot be read as evidence that the groups had identical outcomes, nor can it be used to quantify the apparent difference between their posted proportions.
The confidence intervals for the two group estimates are broad and overlap. Such overlap is not a substitute for a formal test of the difference; nor does it establish that the groups are equivalent. A valid comparison would need a clearly reported effect estimate and its uncertainty, with an analysis that accounts for the randomized groups and the available outcome data. That inferential result is not reported in the registry record.
The primary endpoint description states that data were not collected on two Control participants. The record does not explain the reasons for missingness or describe imputation, exclusion rules, or sensitivity analyses. Because missing outcomes can affect an estimated proportion, the posted group summaries should be read together with this limitation. The group denominators shown for the outcome are not a substitute for a stated missing-data analysis.
These results come from a pilot trial. The registry describes the study as a pilot test, but does not report a sample-size justification or the intended precision for the primary estimate. The observed proportions can help describe what was recorded in this study, but the available registry information does not support a definitive claim about comparative effectiveness. The appropriate reading is descriptive: the posted group estimates differ, their uncertainty is substantial, and the registry record does not supply an interpretable statistical comparison.
9. Statistical Methods Explained
What does a proportion estimate?
A proportion is the fraction of people in a defined group who meet an outcome definition during a specified period. In this trial, the event was illness relapse during the period from immediate post-intervention to endline. The registry reports proportions of 0.27 for KUPAA plus standard of care and 0.45 for standard of care. A proportion is useful because it gives an understandable summary of how often an event was recorded within each group. It does not show when each event occurred, how long it lasted, or whether an individual had more than one episode.
Interpretation depends on a clear event definition and consistent measurement. The registry defines relapse using hospitalization or a new non-hospitalized illness episode with symptom recurrence after one month of controlled symptoms. The record does not describe the assessment procedures or whether outcome assessors were independent of treatment delivery. Masking was listed as none, so readers should not assume blinded outcome assessment.
What does a confidence interval express?
A confidence interval presents a range of values compatible with the data under the method used to calculate it. It communicates uncertainty around an estimate rather than a range containing a known fraction of individual outcomes. The reported 95% confidence interval for the KUPAA group is 0.13 to 0.46; for the control group it is 0.27 to 0.64. The intervals indicate that the group proportions are estimated with appreciable uncertainty. Their width reflects limited information for estimating each group’s event proportion, though the registry does not report a formal precision target.
A confidence interval for each group is different from a confidence interval for the difference between groups. The latter would directly address the comparative question. The registry comparison entry has a risk difference of 0.0 and limits of 0.0 to 0.0, despite the two nonzero group estimates. Since the record lists no analyses and does not explain this entry, it should not be treated as an interpretable effect estimate. No valid interval for the between-group effect is reported in the registry record.
Why is the analysis population important?
The outcome summary lists 33 participants in the KUPAA group and 31 in the control group. The primary outcome description says that data were not collected on two Control participants. The record does not say whether the analysis included everyone as randomized, only those with observed outcome data, or another population. Nor does it report whether missing outcomes were imputed or addressed in sensitivity analyses.
In general, an analysis that includes participants according to their assigned groups preserves the logic of randomization, while excluding people with unavailable outcomes can weaken that protection if missingness is related to prognosis or treatment. This is a general statistical consideration, not a claim about why data are missing in this trial. Without the analysis population and missing-data methods, the influence of the unavailable outcomes cannot be assessed from the registry record.
What can be concluded from the posted comparison?
A randomized design is intended to support a comparison between groups, but randomization alone does not provide the analysis result. The statistical comparison must be reported with an effect measure, uncertainty, and a clear account of how outcomes were analyzed. Here the arm-level proportions and intervals are available, but the listed comparison values are all zero and no analyses are listed. Those comparison values do not supply a usable estimate of the relationship between assignment and relapse.
The record also lists three outcome measures but zero analyses and zero posted secondary outcomes in the available information. It does not state which outcome was prioritized in a prespecified analysis plan, whether any multiplicity adjustment was intended, or whether interim analyses were conducted. Those omissions do not establish that such procedures were or were not used; they mean the details are not reported in the registry record.
How should planning illustrations be understood?
Any sample-size illustrations associated with these results are planning calculations, not findings from the randomized trial. The stated assumptions use a two-sided significance level of 0.05, equal allocation, and a normal approximation, with an effect defined as a fraction of the observed effect. Such calculations describe hypothetical future-study requirements under their assumptions. They are not observed power, do not change the uncertainty in this trial, and do not repair the absence of an interpretable inferential comparison. The registry record does not report the trial’s own sample-size justification or planning assumptions.
In general, calculations based on an observed effect can be unstable, particularly when the underlying estimate is imprecise. Planning for a smaller effect than the observed estimate can be a way to explore sensitivity to assumptions, but no specific planning effect or resulting participant count should be inferred unless it is explicitly reported. These concepts are general explanations of planning methods; they are not additional results from this trial.
10. Confidence Intervals and Precision
Precision concerns how closely an estimate is pinned down by the data. The primary-outcome interval for KUPAA plus standard of care runs from 0.13 to 0.46 around the estimate of 0.27. The control interval runs from 0.27 to 0.64 around the estimate of 0.45. Both ranges cover meaningfully different possible underlying event proportions, so the point estimates should not be mistaken for exact values.
These are confidence intervals for separate group proportions, not an interval for the treatment contrast. The registry does not report a usable confidence interval for the difference between groups. The comparison entry gives limits of 0.0 to 0.0, but that zero-width entry is not reconcilable with the posted arm-level estimates and is not accompanied by a valid analysis. It cannot be used to claim precise knowledge of the comparative effect.
The intervals also do not show the distribution of individual outcomes or identify which participants had missing data. The primary-outcome note says data were not collected on two Control participants, while the reason and handling of those data are not reported in the registry record. Missingness can add uncertainty or introduce bias depending on its causes and how it is handled; the available record does not establish either.
In general, confidence intervals are informative when interpreted in light of the design, analysis method, and target population. The registry does not report the interval-calculation method, a precision target, or the planned analysis population. As a result, the arm-level intervals communicate uncertainty in the reported summaries, but they cannot resolve the uncertainty about the comparative effect or the consequences of missing outcome data.
11. Multiplicity, Missing Data and Analysis Populations
The registry lists three outcome measures, but the information available for this analysis includes no posted secondary outcomes and lists zero analyses. The primary endpoint is identified as the proportion of patient participants experiencing illness relapse. The record does not report a complete statistical analysis plan, which outcomes were formally tested, or how multiple comparisons were handled. It is therefore not possible to determine whether multiplicity adjustment was planned or applied. This absence should not be interpreted as evidence that no such plan existed; the details are not reported in the registry record.
For the primary outcome, the registry specifically notes that data were not collected on two Control participants. The outcome summaries report 33 participants in the intervention arm and 31 in the control arm. The record does not give the reasons for missingness, the timing of missingness, or whether the missing participants differed from those with observed outcomes. It also does not report imputation, a complete-case rationale, or sensitivity analyses that test alternative assumptions about the missing outcomes.
The analysis population is not stated. Although the trial was randomized, the registry record does not say whether the primary outcome was analyzed according to randomized assignment or using another approach. Safety summaries use their own listed at-risk counts, which should not be assumed to define the primary-outcome population. The available information is insufficient to assess adherence to assignment, crossovers, protocol deviations, or exclusions from analysis.
These uncertainties matter because outcome selection, missing-data handling, and analysis-population definitions can affect both estimates and the confidence placed in them. The group-level proportions remain descriptive of the posted summaries, but the registry record does not provide enough detail to reconstruct or verify the primary analysis.
12. 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 |
|---|---|---|---|
| KUPAA Intervention + Standard of Care | 0 | 33 | 0.0% |
| Control - Standard of Care | 2 | 33 | 6.1% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| KUPAA Intervention + Standard of Care | 0 of 33 |
| Control - Standard of Care | 1 of 33 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| KUPAA Intervention + Standard of Care | 0 of 33 |
| Control - Standard of Care | 0 of 33 |
Most frequent serious adverse event terms
| Event term | Organ system | KUPAA Intervention + Standard of Care | Control - Standard of Care |
|---|---|---|---|
| Patient who left home temporarily/missing (relapse related) | Psychiatric disorders | 0 of 33 | 2 of 33 |
Serious events with the highest arm-level rate: patient who left home temporarily/missing (relapse related) (6.1%).
The registry’s serious adverse event summary lists 33 participants at risk in each arm. It reports zero serious adverse events in the KUPAA intervention group and two in the control group. Deaths are listed as zero in the intervention group and one in the control group. These are descriptive counts in the registry record; they do not establish that an event was caused by, or prevented by, an intervention.
The record identifies “Patient who left home temporarily/missing (relapse related)” as a top serious adverse event term, with 6.1% reported. It does not provide enough detail to determine how that term relates to the primary relapse outcome or to attribute it to a particular group. Other adverse events are listed as zero in both groups. The registry does not provide narratives, timing, severity details beyond the serious-event designation, or an assessment of causality.
Safety counts should be interpreted separately from the primary relapse proportions. The record’s safety at-risk counts are not the same as a documented primary-outcome analysis population, and the registry does not explain any relationship between the two. With a pilot trial and limited event counts, these figures describe what was recorded but do not provide a precise comparative safety estimate. No individual treatment decision should be based on these registry summaries alone.
13. What the Result Does, and Does Not, Mean
The results establish that the registry posted group-level estimates for the primary relapse outcome during follow-up. The estimate was 0.27 for KUPAA plus standard of care and 0.45 for standard of care, with the stated 95% confidence intervals. This is a descriptive account of the reported outcomes in the trial groups.
The results do not establish a statistically reliable difference between the groups. Although the arm-level point estimates are numerically different, the registry does not provide an interpretable inferential comparison. Its comparison entry reports zero for the risk difference and its limits, values that do not align with the posted arm-level proportions. The record lists no analyses, and it does not report a usable test or model for the primary outcome. The zero entry is not evidence that the groups were equivalent.
The findings also do not show that the observed proportions will apply to other settings, populations, or forms of care. The trial was described as a pilot of a culturally tailored model in Tanzania. Information about participant characteristics, implementation fidelity, recruitment, and the wider representativeness of the sample is not reported in the registry record. The available outcome information cannot establish the model’s effects on outcomes beyond the specified relapse measure.
Missing primary-outcome data for two Control participants add uncertainty. The record does not state why their data were unavailable or how missingness was addressed. Nor does it provide a defined analysis population or sensitivity analysis. Confidence intervals around each group proportion do not account for every possible uncertainty about outcome ascertainment, missingness, or analysis choices.
Overall, the registry supports a cautious descriptive reading of a randomized pilot: group proportions and their intervals are reported, but the comparative effect remains unresolved in the posted analysis information. The results neither demonstrate benefit nor establish absence of benefit. A fuller interpretation would require a transparent group comparison, analysis methods, missing-data handling, and further evidence beyond this pilot record.
14. Why This Trial Matters Statistically
This trial is statistically notable because it pairs a randomized parallel design with an explicitly pilot-oriented question: whether a culturally tailored family psychoeducation model could be tested in a setting that includes traditional and biomedical understandings of mental illness and relatives as co-facilitators. Pilot studies can contribute information about feasibility and outcome measurement, but the registry record does not report detailed feasibility results or a design justification. The primary outcome provides a concrete endpoint for describing relapse over follow-up.
The record also illustrates why group estimates and comparative inference are not interchangeable. Separate proportions with confidence intervals describe each arm, but readers need a coherent estimate of the between-group effect to answer the randomized comparison. Here, the posted comparison entry is zero-valued despite nonzero arm estimates, and no analyses are listed. Reporting an interpretable estimate and its uncertainty is essential to make the randomized design statistically useful.
Missing outcome data are another important feature. The record identifies two Control participants without collected primary data, yet does not state how this affected analysis. Clear reporting of the analysis population, reasons for missingness, and sensitivity analyses would help readers judge whether the group summaries are robust to plausible assumptions.
Finally, the record distinguishes a planned intervention from the information needed to assess its implementation. It describes joining sessions, a workshop, and weekly family psychoeducation, but does not provide detailed delivery or adherence results. Statistical interpretation depends not only on the assigned groups and outcome counts, but also on transparent reporting of how the intervention was delivered, outcomes were measured, and analyses were selected. These reporting gaps limit what can be learned from the posted results without diminishing the value of a pilot question that addresses a context-specific model of care.
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.
- Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.
The registry record offers limited detail for evaluating the primary comparison. It reports arm-level relapse proportions and confidence intervals, but the comparison entry contains only zero values and does not provide an interpretable effect estimate. The record lists zero analyses. The method used to calculate the group intervals and the planned primary analysis are not reported in the registry record.
Two Control participants had no collected primary-outcome data. The reason, timing, and handling of the missing data are not described, and no sensitivity analysis is reported. The record does not identify the analysis population or explain whether outcomes were analyzed according to randomized assignment. These omissions prevent assessment of how missing observations may have influenced the reported control proportion.
Outcome ascertainment also has limits in the available description. Relapse is defined by hospitalization or a non-hospitalized episode after a period of controlled symptoms, but the record does not report who assessed events, how events were verified, or whether assessment was standardized across arms. Masking is listed as none. The available record therefore does not allow assessment of possible differences in event detection.
The intervention is described in broad terms, including joining sessions, a group workshop, and weekly family psychoeducation. Detailed delivery duration, adherence, fidelity, and variation across groups are not reported in the registry record. Participant characteristics, recruitment details, and information needed to judge representativeness are also not included in the registry record for this analysis.
The record lists three outcome measures but no posted secondary outcomes and no analyses. It does not report the statistical plan, multiplicity handling, interim analyses, or sample-size justification. These limitations concern what the registry record documents; they should not be interpreted as proof that the study team did not conduct additional procedures.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Family Psychoeducation for Adults With Psychotic Disorders in Tanzania — NCT04013932.
- Registry record last updated 2022-02-04; status last verified 2022-01.
- Results first posted 2022-02-04.
- 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.
19. Record Summary
NCT04013932 is a completed randomized, parallel-group pilot trial titled “Family Psychoeducation for Adults With Psychotic Disorders in Tanzania.” It evaluated KUPAA family psychoeducation plus standard of care against standard of care. The record lists enrollment of 66, no masking, and Duke University as sponsor.
The primary outcome was the proportion of patient participants experiencing relapse from immediate post-intervention to endline, approximately 6-7 months after the intervention. The posted estimate was 0.27 for KUPAA plus standard of care, with a 95% confidence interval of 0.13 to 0.46, and 0.45 for control, with a 95% confidence interval of 0.27 to 0.64. The registry notes that primary-outcome data were not collected on two Control participants.
The record does not provide an interpretable inferential comparison, lists no analyses, and does not report the statistical plan or missing-data handling. The reported arm-level estimates are therefore descriptive, and the comparative result remains uncertain.