Randomized Controlled Trial to Test an Alzheimer's Family Caregiver Intervention in Vietnam: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Randomized Controlled Trial to Test an Alzheimer's Family Caregiver Intervention in Vietnam (NCT03587974), a randomized na study. The primary outcome is Zarit Burden Inventory-4.

2. Clinical Question

The study was described as a cluster randomized controlled trial intended to assess feasibility, acceptability and preliminary effectiveness. Those aims are broader than a test of an outcome difference alone: feasibility and acceptability concern whether the intervention can be delivered and received in the intended setting, while preliminary effectiveness concerns outcome patterns that may inform later research. The registry names Patient Health Questionnaire-4 and Alzheimer's Disease Knowledge Scale as secondary outcomes, but their numerical results are not reported in the registry record.

3. Trial Design

The registry describes a randomized, parallel-group, cluster randomized controlled trial. It lists single masking, but does not report who was masked or how masking was implemented. The study was sponsored by the University of California, Davis. Its status is COMPLETED, with a recorded start date of 2018-06-10 and primary completion date of 2018-10-31.

Enrollment is recorded as 60. The registry lists two arms: REACH-VN and Enhanced Control. REACH-VN involved an in-home psychosocial intervention delivered in 4-6 sessions over 2-3 months. Enhanced Control involved a single educational session about the nature of dementia. The clinical question is therefore specific to these intervention packages and to the caregiver burden outcome registered for assessment at 3 months.

Cluster randomization means that assignment is organized at the level of groups or clusters rather than treating every participant's assignment as wholly independent. The registry record does not identify the cluster unit, the number or sizes of clusters, or the procedures used to account for clustering in the sponsor's analysis. Those details matter because outcomes from people in the same cluster may be more alike than outcomes from people in different clusters. A simple analysis that treats every participant as independent can understate uncertainty when this dependence is present.

The record describes the purpose as determining feasibility and preliminary effectiveness of a culturally adapted psychosocial intervention for Alzheimer's family caregivers in Vietnam. It does not report a sample-size justification, recruitment procedures, details of allocation concealment, or further protocol information. The registered primary completion date and completed status describe the trial record; they do not, by themselves, establish that every intended analysis or outcome result was reported. No sponsor analysis estimates are 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

Reasons for not completing this period:

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 Zarit Burden Inventory-4, measured over 3 months.

The primary endpoint was the Zarit Burden Inventory-4, assessed at 3 months. It measures caregiver burden and is recorded as a score on a scale ranging from 0-16. Higher scores indicate worse outcome, meaning more burden. The registry specifies that the primary measure is a mean with standard deviation as the dispersion measure.

Caregiver burden is relevant to the stated purpose because the intervention was designed to support family caregivers, enhance their knowledge and skills, and reduce stress. A burden score provides a quantitative way to describe how caregivers report this aspect of their experience at the registered follow-up time. In general, a continuous score can be summarized by its mean and variability, and a comparison can estimate whether the group averages differ. The score alone does not describe every dimension of caregiver wellbeing, and the endpoint should be interpreted alongside the intervention's broader feasibility and acceptability aims.

The registry record does not report the observed mean or standard deviation for either arm, a between-arm estimate, or an uncertainty interval for the primary endpoint. It also does not report the primary outcome population in further detail. Without those results, the direction and precision of the endpoint comparison cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does cluster randomization change?

Cluster randomization assigns groups or clusters rather than relying on independent assignment of each participant. The registry identifies this design, but does not report what constituted a cluster. That missing detail limits the ability to reconstruct the analysis or assess how assignment operated. In general, participants sharing a cluster can have related outcomes because of common settings, contacts, or other shared influences. This within-cluster similarity is called correlation.

When outcomes are correlated within clusters, the effective information in the data can be less than an analysis assuming independent observations would suggest. A method that accounts for clustering adjusts uncertainty to reflect that dependence. Ignoring clustering can make standard errors too small and confidence intervals too narrow. This is a general explanation of why the analysis must match the design; it does not establish which specific statistical method the sponsor used. The registry does not report that method.

Why not recalculate the comparison from participant-level summaries?

The registry record flags that a simple individual-level re-analysis would be misleading for this cluster-randomized trial and specifies that only the sponsor's own analysis should be shown. This is a design-based caution. A recalculation based on individual-level summary statistics could treat correlated observations as independent and therefore give misleading uncertainty. No numerical sponsor analysis is reported in the registry record, so an adjusted estimate cannot be described or checked from these registry entries.

In general, an analysis appropriate to a cluster trial needs to reflect the assignment structure and the outcome correlation within clusters. Different valid methods can make different assumptions and may use different adjustments. The registry does not report details sufficient to identify the sponsor's model, its covariates, its handling of cluster sizes, or its approach to missing outcome data. None should be inferred from the design label alone.

What does the endpoint summary mean?

The primary measure is continuous and is specified as a mean with standard deviation. A mean describes the average score in a group, while a standard deviation describes how dispersed individual scores are around that average. In general, the mean can convey a group-level pattern, and the standard deviation helps describe variation among participants. Neither quantity alone is a between-group effect estimate: that requires a contrast between groups, with uncertainty assessed using an analysis that reflects the design.

The Zarit Burden Inventory-4 ranges from 0-16, and a higher score means more caregiver burden. An estimated difference in means would therefore need to be read with that direction in mind. But the registry record does not report group means, standard deviations, or a primary between-arm estimate. No direction or size of the observed difference can be stated.

How should the planning assumptions be read?

The registry's sample-size note describes calculations as planning illustrations based on the observed effect, not as the design rationale for the trial. It specifies a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation, with the target effect set as a stated fraction of the observed effect. For non-time-to-event outcomes, the note defines the planning quantity as participants per arm. These assumptions explain the intended interpretation of such calculations, but the numerical estimates and the observed effect are not reported in the registry record.

In general, a planning calculation asks how much information might be needed to detect a chosen effect under stated assumptions. Changing the target effect or assumptions can change the resulting sample-size illustration. Such an illustration does not show the trial's achieved power, does not replace the observed outcome comparison, and does not establish that the trial was adequately powered. The registry explicitly distinguishes the planning illustration from the trial's own design.

What is not available for statistical appraisal?

The registry records zero analyses, while naming a primary outcome and secondary outcomes. Numerical outcome summaries, model details, uncertainty intervals, and test results are not reported in the registry record. The sample-size justification, interim analyses, and full statistical plan are also not reported in the registry record. Without these details, readers cannot reproduce the sponsor analysis, assess its assumptions, or determine how planned analyses relate to the posted outcome information.

General statistical interpretation should distinguish an estimate from its uncertainty and from a test of a null hypothesis. The estimate describes the size and direction of a contrast; uncertainty describes how precisely it is determined; a test addresses compatibility with a specified null model. These are related but not interchangeable. For this trial, the registry's available outcome information does not include the numerical quantities needed to make those judgments for the primary measure.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("REACH-VN", "Enhanced Control")
n   <- c(25, 26)
m   <- c(2.9, 5.5)
sp  <- c(2.4, 3.4)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  REACH-VN  25  2.900  2.400  1.909  3.891
CARM  Enhanced Control  26  5.500  3.400  4.127  6.873

11. Multiplicity, Missing Data and Analysis Populations

The registry names the Zarit Burden Inventory-4 as the primary outcome and lists Patient Health Questionnaire-4 and Alzheimer's Disease Knowledge Scale as secondary outcomes. It reports three outcome measures and zero analyses. Numerical findings for the primary and secondary measures are not reported in the registry record.

Multiplicity is the statistical issue that arises when several outcomes, time points, or comparisons are tested. In general, conducting many tests can increase the chance of at least one apparently notable result, so a prespecified analysis plan may describe how outcomes are prioritized or how inference is adjusted. The registry record does not report a multiplicity strategy, the detailed analysis plan, or interim analyses. It is not possible to determine whether or how any adjustment was planned or used.

Missing data are also relevant to interpreting a trial outcome. In general, the amount and pattern of missingness, the reasons for missing measurements, and the methods used to address them can affect estimates and uncertainty. The registry record does not report endpoint completion, missing outcome counts, reasons for missingness, or the sponsor's handling of missing data. The enrollment figure alone does not establish how many outcome observations contributed to an analysis.

Analysis populations are not described in detail. The registry does not report whether the sponsor used an intention-to-treat population, a per-protocol population, or another definition for the primary analysis. It also does not state how cluster membership, participant eligibility after randomization, or adherence to the intervention was handled analytically. These omissions prevent a full assessment of which participants and clusters contributed to the outcome estimate.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry lists serious adverse events, other adverse events, and deaths separately for REACH-VN and Enhanced Control. For each arm, 30 participants are listed as at risk for serious adverse events and 30 as at risk for other adverse events. The reported number affected is 0 for serious adverse events and 0 for other adverse events in each arm, and deaths are recorded as 0 in each arm. No top serious-event terms are listed.

These figures describe what the registry reports, not a complete account of every possible safety experience. A reported zero means that the registry records no events in that category for the listed arm; it does not explain how events were sought, defined, monitored, or attributed. Those methods are not reported in the registry record. The record also does not give narrative descriptions of event surveillance or follow-up for adverse events.

Safety counts should be read separately from the primary caregiver burden outcome. They do not provide evidence about the size or precision of an outcome difference, and they do not resolve the missing primary or secondary outcome results. The registry's zero event counts are descriptive entries and should not be used to infer that an intervention has no risks.

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

The trial record establishes the research question, the cluster randomized parallel-group design, the two intervention conditions, and the planned primary outcome. It also states that caregiver burden was assessed with the Zarit Burden Inventory-4 at 3 months and gives the scale's range and direction. These details clarify what comparison was intended and how a score should be interpreted.

What the record does not establish is the numerical result of that comparison. The arm-specific means and standard deviations, sponsor's estimated effect, and uncertainty around the estimate are not reported in the registry record. Therefore, the evidence available in this record does not show whether the groups differed on caregiver burden, how large any difference was, or how precisely it was estimated. The same limitation applies to the named secondary outcomes.

This distinction is important because a trial description is not an outcome result. Randomization describes how treatment assignment was intended to be generated; it does not guarantee comparable groups in every respect or provide a measured effect. Completion status and enrollment also do not indicate the direction, magnitude, or statistical uncertainty of an outcome contrast.

The stated goal includes feasibility and acceptability as well as preliminary effectiveness. The intervention description and study setting identify what was tested, but numerical feasibility and acceptability findings are not reported in the registry record. Consequently, readers cannot use this record alone to assess how acceptable the intervention was or how successfully it was delivered.

In general, clinical importance and statistical evidence are distinct questions. A score contrast may be estimated precisely or imprecisely, and its practical meaning depends on context as well as the scale. The registry record does not report a contrast, uncertainty interval, or information to establish a clinically important threshold. No conclusion about clinical benefit or lack of benefit follows from the information recorded. The appropriate reading is limited to the design and outcome definitions, with the primary and secondary results unavailable in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because its cluster randomized design makes the unit of assignment and the unit of analysis important. In general, outcomes among participants in the same cluster may be correlated. Accounting for that dependence is necessary for an uncertainty estimate that reflects the design. The registry itself cautions against a simple individual-level re-analysis, so a standard comparison that assumes all observations are independent would not be a reliable substitute for the sponsor's analysis.

The primary measure also illustrates why outcome definition matters. The Zarit Burden Inventory-4 is a continuous score with a specified direction: higher scores indicate more burden. This makes the intended interpretation straightforward once arm results and a between-arm estimate are available. But without the numerical estimate and its uncertainty, knowing the measure and its direction cannot answer whether a difference was observed or how precise it was.

The study's stated aims include feasibility and acceptability alongside preliminary effectiveness. Those aims make clear that an intervention trial can be informative beyond a single outcome contrast. However, conclusions about implementation or acceptability require relevant results, which are not reported in the registry record. The registry names secondary measures but does not report their results either.

Finally, the sample-size note distinguishes retrospective planning illustrations from the trial's own design. This distinction is important in statistical reporting: a calculation based on an observed effect can illustrate future planning assumptions, but it is not a measure of achieved power and cannot replace the outcome estimate. Because the numerical illustration and observed effect are not reported in the registry record, no further quantitative appraisal is possible.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most consequential limitation for interpretation is the absence of numerical outcome results in the registry record. Although a primary endpoint and secondary measures are identified, there are no reported group summaries or sponsor estimates with which to assess direction, magnitude, or precision. The registered purpose includes feasibility and acceptability, but corresponding findings are not reported in the registry record.

The cluster structure is incompletely described. The record identifies a cluster randomized design but does not state the cluster unit, the number or sizes of clusters, or the method used to account for clustering. It also does not report the sponsor model, its assumptions, or the analysis population. Because an individual-level re-analysis would be misleading, those omissions particularly limit independent statistical appraisal.

Further information about the outcome data is not reported in the registry record, including endpoint completion, missingness, and how missing observations were handled. The record does not describe allocation concealment, who was masked, or the detailed intervention delivery and adherence data. These details should not be inferred from the single-masking designation or the intervention descriptions.

The registry record does not report a sample-size justification, statistical plan, interim analyses, or multiplicity strategy. It also does not provide numerical results for the secondary outcomes. These limitations mean that readers cannot assess whether the reported analyses followed a prespecified plan or how uncertainty across outcomes was addressed. The safety entries are descriptive and are not accompanied by event-monitoring methods.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Randomized Controlled Trial to Test an Alzheimer's Family Caregiver Intervention in Vietnam — NCT03587974.
• Registry record last updated 2020-07-22; status last verified 2020-07.
• Results first posted 2020-07-22.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT03587974 is a completed, randomized, parallel-group cluster trial sponsored by the University of California, Davis. The registry records enrollment of 60, a start date of 2018-06-10, and a primary completion date of 2018-10-31. It lists single masking, without specifying who was masked.

The trial compared REACH-VN, an in-home psychosocial intervention delivered in 4-6 sessions over 2-3 months, with an enhanced control consisting of a single session about the nature of dementia. The primary outcome was the Zarit Burden Inventory-4 at 3 months; higher scores on its 0-16 range indicate more caregiver burden. Patient Health Questionnaire-4 and Alzheimer's Disease Knowledge Scale were listed as secondary outcomes.

Numerical primary and secondary outcome results and the sponsor's statistical analysis are not reported in the registry record. The record reports zero serious adverse events, other adverse events, and deaths in each arm.