Haiti Community Health Workers (CHW) Adaptation: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Haiti Community Health Workers (CHW) Adaptation (NCT05091710), a non-randomized na study. The primary outcome is The Number of Participants Who Reported the HF Follow up Program by CHWs Was Acceptable.

2. Clinical Question

The study addresses two related but distinct questions. One is whether the adapted follow-up program is acceptable to patients who receive it. The other is whether the intervention can improve heart failure care and outcomes, including linkage to care. The primary endpoint is about acceptability, not a direct measure of linkage, clinical status, or comparative efficacy.

The distinction matters for interpretation. The standard-of-care group did not receive the follow-up program, so the primary acceptability outcome was not assessed in that group. It therefore does not create a like-for-like comparison between intervention and standard care. The historical group is relevant to the broader study context, but its presence alone cannot establish whether any differences in care or outcomes were caused by the program.

The registry describes the intended population and intervention but does not report the primary endpoint result, participant-level responses, or detailed characteristics of the groups. It also does not report the results for linkage of heart failure patients. The clinical questions are therefore identifiable from the record, while their answers cannot be determined from the results recorded there.

3. Trial Design

This is a completed pilot trial of a community health worker intervention for heart failure care in rural Haiti. The registry records a parallel-group design and non-randomized allocation, with no masking. The listed enrollment is 72. The record identifies Boston Medical Center as the sponsor and gives the study start as 2023-02-23 and primary completion as 2025-08-31.

The intervention group is described as discharged heart failure patients receiving a follow-up care program delivered by trained community health workers. The comparator is a historical reference group that received standard care before community health worker training. Because the groups are not randomized and are drawn from different periods, the comparison is not a like-for-like control comparison. Differences between them could reflect the intervention, differences in who entered each group, changes over time, or other factors not described in the record.

The study builds on prior adaptation work. The registry says the intervention was adapted for rural Haiti using the ADAPT-ITT framework and that its first 6 steps were completed in a prior study. The current protocol is described as covering training and testing of the adapted intervention. The record also describes gathering participant feedback on feasibility, appropriateness, and acceptability, alongside assessment of heart failure outcomes.

The primary outcome is a count of participants who considered the home visit acceptable, assessed at 90 days. The standard-of-care group did not receive the program, so this outcome was not assessed there. Linkage of heart failure patients is listed as a secondary outcome. The record lists 7 outcome measures but does not describe all of them in the information recorded. It lists 0 analyses, so the statistical methods used for any outcome are not reported in the registry record.

The registry status is completed, but completion status does not by itself establish that every planned measure was collected or analyzed. The record does not report the recruitment process, group assignment procedures in detail, or how missing observations were handled. These omissions limit how fully the design and its results can be evaluated.

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: 1 participant in the Follow up Care Program for HF Patients arm was lost to follow up prior to receiving the intervention.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Standard of Care arm did not receive the HF follow program so this outcome was not assessed in that group.

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 The Number of Participants Who Reported the HF Follow up Program by CHWs Was Acceptable, measured over 90 days.

The primary endpoint counts participants who reported that the heart failure follow-up program delivered by community health workers was acceptable. Patients were asked whether they found it acceptable to have a community health worker visit their home within 5 days of hospital discharge. Responses used a 5-point Likert scale: 1=poor, 2=fair, 3=ok, 4=good, and 5=great. The registry defines an acceptable response as “good” or “great.” The stated assessment timeframe is 90 days.

This endpoint addresses whether recipients viewed a specific feature of the intervention favorably. Acceptability is relevant to an intervention that depends on patients engaging with home visits; feedback can help assess whether the delivery approach is suitable for its intended setting. It is distinct from feasibility, which concerns whether delivery can be carried out, and from clinical effectiveness, which concerns outcomes such as care linkage or health status.

The endpoint is a count rather than a reported average rating. Turning the scale into an acceptable/not-acceptable classification makes the primary measure easier to describe as a number of patients meeting the stated threshold, but it also does not preserve distinctions among responses below or above that threshold. The record does not report how many participants met the threshold, how many answered the question, or the distribution across the 5 response options.

The standard-of-care group did not receive the follow-up program and was not assessed for this outcome. Thus, the endpoint is a measure of acceptability within the program group, not a comparison of acceptability between groups. The observed endpoint result is not reported in the registry record, so the degree of acceptability cannot be evaluated from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a non-randomized, parallel-group study with no masking. It does not report the statistical analysis methods used for the outcomes. The explanations below describe general statistical considerations for this design and these outcome types; they are not claims about analyses performed in this trial.

What does a count endpoint tell the reader?

The primary endpoint is a count of participants whose responses met a predefined acceptability threshold. A count is most informative when read together with the number of participants assessed. The count alone indicates how many met the definition, while the denominator indicates how many had an opportunity to contribute an answer. A proportion can also be useful for describing the share meeting the threshold, but no proportion should be inferred without the corresponding result and denominator. Neither the count nor a proportion, by itself, explains why some patients did or did not find the visit acceptable.

The underlying responses are ordered categories, from poor to great. Combining “good” and “great” into the acceptable category is a threshold-based summary. In general, this makes an endpoint easier to communicate but discards information about differences among the individual response categories. A reader would ideally see both the threshold result and the response distribution. The registry record does not report either the primary count or the distribution.

Why does the non-randomized comparator matter?

In a randomized comparison, assignment by chance is intended to make groups comparable on average, including factors that investigators did not measure. Here, allocation is non-randomized, and standard care is represented by a historical reference group. In general, this can leave differences between groups in age, severity, access to care, or other characteristics that also affect outcomes. Changes over calendar time may further distinguish a historical group from a later intervention group. These are potential sources of confounding, not evidence that any particular imbalance occurred in this study.

Adjusted statistical models can sometimes account for measured differences, but they cannot automatically remove bias from unmeasured factors or make non-comparable groups equivalent. The registry record does not report baseline characteristics, adjustment variables, or an adjusted analysis. It is therefore not possible to assess whether or how the study addressed group differences.

What does uncertainty mean?

In general, a confidence interval describes uncertainty around an estimated effect or rate under the assumptions of the analysis. A narrow interval is more precise than a wide interval, while an interval that spans a broad range leaves more uncertainty about the underlying value. A p-value addresses compatibility of data with a specified null model; it does not measure the size or importance of an effect. Neither a confidence interval nor a p-value can replace the estimate itself and its context. The registry record does not report these statistics for the outcomes discussed.

For a single-group acceptability count, an interval around a proportion would describe uncertainty in the population proportion represented by the observed responses, subject to the sampling and measurement assumptions. For a group comparison, an interval around a contrast would describe uncertainty in the difference or other chosen effect measure. The primary endpoint is not assessed in the standard-of-care group, so it does not support that kind of between-group contrast.

Why can outcome type affect the analysis?

Binary outcomes classify participants as having or not having an event, while counts record how many participants meet a condition. The registry describes the primary outcome as a participant count, with acceptability defined by two response categories. In general, analysis of a binary outcome depends on both the event count and the number assessed. If a count represents recurrent events per participant rather than participants with any event, methods that account for over-dispersion may be relevant: over-dispersion occurs when variability exceeds what a simple model assumes. That issue should not be presumed for this primary endpoint, which is described as participants meeting a response threshold.

The registry lists linkage as a secondary outcome, but its measurement details and results are not reported in the registry record. The appropriate analysis depends on how linkage was defined and recorded. The same is true for the other outcomes among the 7 listed measures. Their measurement scales, timepoints, and planned analytical methods are not reported in the record available for interpretation.

What can a planning illustration establish?

In general, a sample-size calculation uses assumptions about the effect to detect, variability, allocation, and statistical error rates to illustrate the amount of information a future study might need. A planning illustration based on an observed effect is not the same as a prospective design justification, and it is not observed power. It should not be read as evidence that an effect is real or that a study was adequately designed to find it. The registry describes its planning illustrations as based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It also states that non-time-to-event illustrations are expressed as participants per arm. The observed outcome effects needed to interpret those illustrations are not reported in the registry record.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Follow up Care Program for HF Patients")
n      <- c(29)
events <- c(24)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Follow up Care Program for HF Patients  29  24  82.8  64.2  94.2  65.5  92.4  65.3  92.3

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 7 outcome measures, including the primary acceptability endpoint and linkage of heart failure patients as a secondary outcome. It lists 0 analyses. That entry does not establish whether analyses were performed outside the registry record; it means that analysis details are not documented in the record being interpreted.

When a study evaluates multiple outcomes or conducts multiple statistical tests, the chance of at least one apparently unusual result can increase. In general, multiplicity procedures can manage that issue, but the choice depends on which outcomes are confirmatory and how the analyses are organized. The registry record does not report a multiplicity strategy, a hierarchy among outcomes beyond identifying the primary outcome, or how multiple comparisons were handled. No claim about adjustment or lack of adjustment in actual analyses can be made.

Missing-data methods are also not reported in the registry record. It does not state how many participants completed the acceptability question, whether any responses were missing, or how missing observations were handled. For outcomes measured after enrollment, missingness can affect both the amount of information and the interpretation of results, particularly if participants with missing data differ from those with observed data.

The record does not describe analysis populations, such as which participants were included in each outcome analysis or how protocol deviations were treated. It also does not report an analysis plan, imputation approach, or sensitivity analyses. These details would be important for understanding how results were generated and how robust they might be to assumptions about missing observations. Their absence should be read as a limitation of the registry record, not as proof that the investigators did not use such methods.

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 5% reporting threshold.

The registry’s adverse-event records list 0 serious adverse events, 0 other adverse events, and 0 deaths in the follow-up care program group. They list the same recorded counts in the standard-of-care group. The records identify 30 participants at risk in the program group and 36 at risk in the standard-of-care group for both serious and other adverse events.

These entries describe the events recorded in the registry; they do not establish that no adverse events occurred outside the recording process or that all possible harms were assessed. The record does not report event definitions, ascertainment procedures, severity grading beyond the serious/other distinction, or a detailed observation window for these safety entries. Although the primary endpoint has a 90 days timeframe, the record does not state that this is the safety observation period.

Zero recorded events should therefore be interpreted narrowly. It is a statement about the counts entered in the registry, not proof of absence of risk. Without more detail on monitoring, completeness, and the period over which events were collected, the sensitivity of the safety assessment cannot be judged. The record also does not list serious adverse event terms. The absence of listed terms is consistent with the recorded zero serious events but does not add information about how events were sought or classified.

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

The study asks whether a community health worker follow-up program adapted for rural Haiti is acceptable to patients and whether it can improve heart failure care. The registry describes a program intended to support discharged patients through home follow-up, peer support, and navigation of the health system. It identifies acceptability as the primary endpoint and linkage of heart failure patients as a secondary outcome.

What the record does not establish is the observed result. The primary acceptability count is not reported, and the registry does not report results for linkage or other heart failure outcomes. Without outcome estimates, it is not possible to describe the size or direction of any observed effect, or the precision with which it was measured. The study’s completed status does not fill that gap.

The primary endpoint also has a limited scope. A favorable report about a home visit would describe patient acceptability under the registry’s definition; it would not by itself demonstrate improved linkage, better clinical outcomes, or comparative benefit. Conversely, acceptability is not a substitute for evaluating whether the program can be delivered consistently or whether it changes outcomes. Those are related but separate questions.

The historical standard-of-care group does not make the primary acceptability outcome comparative, because that group did not receive the intervention and was not assessed on this endpoint. For other outcomes, non-randomized allocation means that a difference between groups could reflect pre-existing differences or changes over time as well as the intervention. The record does not report analytical adjustments that would clarify this concern.

Accordingly, the record supports a description of the trial’s purpose, design, and endpoint definitions, but not a conclusion about the intervention’s observed acceptability or efficacy. The recorded zero adverse-event and death counts should likewise be understood as registry entries, not as evidence that risk was absent. A fuller interpretation would require the outcome results, denominators for each measure, uncertainty estimates, missing-data information, and analysis methods, none of which are reported for the relevant outcomes in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of how an intervention can be evaluated during adaptation to a new setting. The registry describes prior adaptation of a heart failure intervention for rural Haiti and a subsequent pilot focused on training and testing the adapted program. In this context, acceptability is a meaningful implementation question: an intervention that depends on home visits must be considered from the perspective of the people asked to receive them.

The endpoint also illustrates the difference between an implementation measure and an effectiveness measure. The primary outcome counts participants who gave responses meeting a specified acceptability threshold. Linkage to care is listed separately as a secondary outcome. Keeping these questions distinct helps prevent a positive finding on one dimension from being mistaken for evidence on another. The registry does not report results for either outcome, so their empirical answers remain unknown from this record.

Design features shape what can be learned. A historical standard-of-care group offers a reference for outcomes that are measured in both groups, but non-randomized allocation leaves the comparison open to confounding and changes over time. Meanwhile, the primary acceptability outcome is assessed only in the intervention group, so it is descriptive rather than comparative. These features make it especially important to distinguish feasibility and acceptability evidence from causal evidence about clinical outcomes.

The record also demonstrates why outcome definitions and analysis reporting matter. The response threshold is explicit, but the result, denominator, response distribution, uncertainty, and missing-data handling are not reported. Without those elements, readers cannot assess how representative or precise the acceptability estimate would be. In a pilot, transparent reporting of these quantities would help clarify what was learned and what remains uncertain before any larger evaluation.

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.

Several limitations are specific to the trial description and results recorded in the registry. Allocation is non-randomized, and the comparator is a historical group receiving standard care before community health worker training. This structure does not provide a randomized, concurrent comparison. Group differences or changes over time could influence any comparative outcome, and the registry does not report baseline characteristics or methods used to address confounding.

The primary endpoint is assessed only among patients receiving the follow-up program. Standard-of-care participants did not receive the program, so they were not assessed for acceptability. The endpoint cannot therefore support a between-group comparison. In addition, acceptability is a patient-reported response to a specific question, and the record does not provide the observed responses or how many participants answered.

The registry does not report results for the primary endpoint, the listed linkage outcome, or the other outcome measures. It also does not report confidence intervals, statistical tests, analysis populations, missing-data methods, or the statistical analysis plan. The record lists 7 outcome measures and 0 analyses, leaving the analytical treatment of outcomes undocumented there. The status is completed, but this does not resolve the lack of reported results.

Safety interpretation is similarly limited. The registry records zero serious adverse events, other adverse events, and deaths in each group, but does not describe event ascertainment, definitions, or the safety observation period. These counts cannot establish that all harms were detected or that risk was absent.

The study is described as a pilot and as an assessment of feasibility, appropriateness, acceptability, and efficacy. The record does not provide additional details on recruitment, participant characteristics, intervention fidelity, or the delivery of the program. Without those details and the outcome results, readers cannot evaluate representativeness, implementation, or the precision of the findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Haiti Community Health Workers (CHW) Adaptation — NCT05091710.
• Registry record last updated 2026-11-05; status last verified 2026-10.
• Results first posted 2026-11-05.
• 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

The record describes a completed, non-randomized, parallel pilot trial of a community health worker follow-up program for heart failure care in rural Haiti. The standard-of-care comparator is a historical reference group, and masking is listed as none. The registry lists enrollment of 72 and identifies Boston Medical Center as sponsor.

The primary endpoint is the count of participants who considered a home visit within 5 days of hospital discharge acceptable, defined as a “good” or “great” response on a 5-point Likert scale. Its timeframe is 90 days. The standard-of-care group did not receive the program and was not assessed for this outcome. Linkage of heart failure patients is listed as a secondary outcome.

The registry does not report the primary endpoint result or results for linkage and other outcomes. It lists 7 outcome measures and 0 analyses. Safety records list zero serious adverse events, other adverse events, and deaths in both groups. Because allocation is non-randomized and the comparator is historical, comparative outcome interpretation would require caution even if results were available.