Heart Failure Polypill at a Safety Net Hospital: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Heart Failure Polypill at a Safety Net Hospital (NCT06029712), a randomized phase 2 study. The primary outcome is Measured Adherence to GDMT by Pill Count.

2. Clinical Question

The rationale recorded for the study is that heart failure regimens can involve several medications, while combination pills have been used for other conditions. The trial focused on whether combining medications could affect adherence in a safety-net hospital setting. It was designed as a pilot, so its reported results should be read as an initial assessment in a limited study population, rather than as a definitive estimate applicable to all people with HFrEF.

3. Trial Design

This was a randomized, open-label, crossover pilot trial listed as Phase 2. Participants were assigned to a sequence involving the heart failure polypill and individual tablets. Because the trial had no masking, participants and study personnel were not blinded to the treatment period. The registry describes the primary outcome after a month of polypill use and after a month of individual tablet use.

The posted primary outcome results combine the individual-tablet periods from both assignment groups and combine the polypill periods. The registry therefore presents period-level summaries rather than separate results for each randomized sequence. It reports a mean adherence ratio for each treatment condition and a 95% confidence interval for each mean. A paired treatment contrast, crossover-model estimate, or estimate accounting for treatment sequence is not reported in the registry record.

The study was conducted at a single center according to the registry summary, and enrollment was 35. The registry status is completed, with a start date of 2024-02-27 and a primary completion date of 2024-09-17. The primary results give 27 observations for each treatment condition; those counts are not the same as the overall enrollment figure. The reason for the difference between enrollment and the reported condition-specific counts is not reported in the registry record.

The trial is described as a pilot intended to investigate adherence, not as a large confirmatory trial. The registry record does not report the planned crossover sequence details, washout arrangements, or the method used to account for within-participant correlation. Those omissions matter when interpreting a crossover study: the two treatment-period observations from one participant are not necessarily independent, and the order in which treatments are received may be relevant.

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

Analysis population for the primary outcome, as defined in the registry: As part of the crossover analysis, the individual tablet periods were combined from both assignment groups (individual tablets first and polypill first), and the polypill periods were combined.

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 Measured Adherence to GDMT by Pill Count, measured over The outcome will be measured 1) following a month of polypill use, and 2) following a month of individual tablet use..

The primary endpoint was measured adherence to guideline-directed medical therapy (GDMT) by pill count. For each prescribed class of GDMT, the registry describes calculating an adherence ratio from the number of pills missing relative to the number expected to be missing. The class-specific ratios were then averaged to create an overall adherence ratio. The outcome is expressed as a percentage, and pill counts could be performed in person or by videoconferencing.

Pill count is a way of assessing whether medication has been used over a specified period. It is an observable measure tied to returned medication, but it does not directly establish when or how each dose was taken. In general, adherence measures can capture different aspects of medication-taking behavior, and a pill-count estimate should be interpreted as the particular operational measure defined by the study.

The registry assessed this endpoint after a month of each treatment condition. The posted mean was 74.6 for individual tablets, with a 95% confidence interval of 66.6 to 82.5; for the polypill, the mean was 83.3, with a 95% confidence interval of 75.3 to 91.3. These are separate condition-level summaries. The registry does not report the paired mean difference or a confidence interval for that difference, so the separate intervals do not give the precision of a direct within-participant treatment comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it is a measured adherence ratio expressed as a percentage, and the registry reports a mean with a 95% confidence interval for each condition. A mean describes the average of the recorded values, while the confidence interval describes uncertainty in that average according to the analysis method. The posted mean for individual tablets is 74.6, with a 95% confidence interval of 66.6 to 82.5. The posted mean for the polypill is 83.3, with a 95% confidence interval of 75.3 to 91.3.

Those condition-specific summaries are not the same as the main randomized comparison. In a crossover design, each participant can be observed in both treatment conditions. A direct treatment effect can therefore be estimated from within-participant contrasts, with methods that account for the paired observations and the trial's sequence or period structure when appropriate. In general, using paired information can improve precision when measurements from the same person are related. It also means that an analysis treating all period observations as independent may not reflect the design. The registry record does not report a paired treatment estimate or the model used for the primary analysis.

A confidence interval is a measure of statistical uncertainty, not a statement about the probability that a particular hypothesis is true. A narrow interval indicates greater precision under the method and assumptions; a wide interval indicates less precision. The posted intervals are for separate means, so they cannot be substituted for an interval around the difference between treatment periods. The direct comparison's confidence interval and p-value are not reported in the registry record.

The registry's secondary heart failure admission outcome is a count of participants with an event. It reports zero of 31 for individual tablets and one of 30 for the polypill. A recalculated, unadjusted risk difference is 0.033, with a 95% confidence interval from -0.031 to 0.098 and p = 0.305. This comparison uses the posted counts and does not adjust for other factors. Its interval expresses substantial uncertainty relative to the small number of observed events. In general, when event counts are sparse, a small change in the observed count can have a noticeable effect on an estimated rate or risk difference.

The registry describes sample-size illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are planning illustrations rather than the trial's own design justification, and they are not observed power. The registry record does not report the trial's sample-size justification. No numerical planning sample sizes are reported in the registry record used for this explanation.

The record lists 20 outcome measures and 17 posted secondary outcomes. It records zero reported analyses. These registry entries do not establish whether analyses were performed but not entered, whether results were unavailable, or whether a specific statistical plan was followed. The primary outcome is reported as a mean with a confidence interval, but no details of the interval construction or model are reported in the registry record.

Multiplicity refers, in general, to the increased chance of false-positive findings when many outcomes or comparisons are examined. If secondary results are interpreted as a set of confirmatory tests, adjustment or a prespecified hierarchy may be important. The registry does not report a multiplicity adjustment, a hierarchy of hypotheses, or a statistical analysis plan. Unadjusted secondary estimates can still be informative descriptively, but they should not be treated as independent confirmation of a primary finding.

Missing data and analysis populations also affect interpretation. The primary period counts are 27 for each condition, while total enrollment is 35. The record does not explain the disposition of participants, the handling of incomplete pill counts, or whether the primary results use an intention-to-treat, per-protocol, or other analysis population. In general, complete-case results can be biased if the availability of measurements is related to treatment or adherence. The missing-data approach and the definition of the analysis population are not reported in the registry record.

For a crossover study, period effects, sequence effects, and carryover are potential design considerations. A period effect is a change over time that could affect outcomes regardless of treatment; a sequence effect concerns the order in which treatments are received; carryover refers to effects of an earlier treatment continuing into a later period. These are general statistical considerations, not claims that such effects occurred in this trial. The registry describes combining treatment periods across assignment groups but does not report tests or modeling for period, sequence, or carryover effects.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures, with 17 secondary outcomes posted, and records zero reported analyses. It does not report a statistical analysis plan, a hierarchy of outcomes, or a procedure for controlling the chance of false-positive findings across multiple tests. In general, testing several outcomes without adjustment can produce apparently notable results by chance. This concern does not make secondary outcomes uninformative, but it means that unadjusted results should not carry the same confirmatory weight as a prespecified primary analysis.

The primary results consist of condition-specific means and confidence intervals. A direct treatment comparison, along with its confidence interval or p-value, is not reported in the registry record. It is also not reported whether the primary comparison was prespecified in a protocol or analysis plan, or whether any alternative analyses were conducted.

The outcome counts vary across the posted results. The trial enrolled 35 participants, the primary results report 27 observations for each treatment condition, and the heart failure admission outcome reports 31 and 30 participants in the two conditions. The registry record does not explain these differences, describe participant flow, or state how incomplete outcome data were handled. The primary analysis population and missing-data method are not reported. Therefore, the effect of missing measurements on the reported means cannot be assessed from the registry record.

The primary outcome was measured in two treatment periods for a crossover study, but the record does not report how within-participant correlation was handled in the primary analysis. It also does not describe analyses for period, sequence, or carryover effects. These are general considerations for crossover designs, not evidence that any particular bias occurred. Without the analysis details, readers cannot determine whether the posted summaries reflect an analysis appropriate to the paired design.

The registry record also does not report a protocol-based sample-size justification or interim analysis plan. The sample-size assumptions described for planning illustrations are not the trial's own design rationale, and those illustrations are not observed power. Missing information about the analysis plan, missing data, and multiplicity limits how strongly the posted findings can support confirmatory conclusions.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: heart failure hospitalization (3.3%), non-heart failure hospitalization (3.2%).

The registry reports serious and other adverse events separately for the two treatment conditions. For individual tablets, 1 of 31 participants at risk had a serious adverse event, and 5 of 31 had an other adverse event. For the polypill, 1 of 30 participants at risk had a serious adverse event, and 6 of 30 had an other adverse event. The record reports zero deaths in each condition.

These are counts of participants affected, not a direct estimate that an event was caused by one treatment. The record does not state that the events were judged treatment-related. It also does not report a statistical comparison of adverse-event risk. In a crossover trial, participants may contribute exposure to both conditions, so condition-specific counts should be interpreted with the design and exposure periods in mind.

The registry lists “Heart Failure Hospitalization” at 3.3% and “Non-Heart Failure Hospitalization” at 3.2% among the top serious-event terms. The record does not give further detail about how those term percentages relate to the condition-specific participant counts. The listed adverse-event figures are descriptive safety information from a small pilot, not a precise comparison of safety between treatments.

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

The posted results describe average pill-count adherence of 74.6 during the individual-tablet period and 83.3 during the polypill period. The registry reports 95% confidence intervals for these means, but it does not report a direct paired comparison between periods. The mean values therefore describe what was observed in each condition; they do not, by themselves, quantify the treatment effect or establish its statistical precision.

The result does not establish that one regimen causes better adherence than the other. The study is a pilot crossover trial with 35 enrolled participants, and the primary results each include 27 observations. The record does not describe why the counts differ or how missing measurements were addressed. A direct analysis that accounts for the crossover design and the relationship between a participant's two measurements is not reported in the registry record.

The result also does not establish whether adherence differences translate into differences in health outcomes. Heart failure admissions were recorded as a secondary outcome, with zero participants among 31 in the individual-tablet condition and one among 30 in the polypill condition. The unadjusted estimate is imprecise, and the registry does not report a broader, definitive clinical-outcome comparison. The results should not be read as evidence of clinical benefit or harm.

More generally, an observed mean and its confidence interval describe data under a particular measurement and analysis approach. Pill count reflects a medication-use measure over the observed period, but it cannot on its own establish the timing or circumstances of every dose. The posted estimate is also limited to the study's population and setting; the registry does not report enough participant detail to assess how representative the analyzed group was.

The findings can be read as an initial description of pill-count adherence and safety events during two treatment conditions. The registry record does not report the inferential details needed to determine the uncertainty of the primary within-participant contrast. It also does not report enough analysis information to resolve questions about missing data, treatment sequence, or multiplicity. These limits call for cautious interpretation, rather than a conclusion that the treatment conditions are equivalent or definitively different.

13. Why This Trial Matters Statistically

This trial raises a statistical question that is important for interpreting many clinical studies: does the analysis match the design? In a crossover trial, participants experience both conditions, so the treatment comparison depends on within-participant information. Condition-specific means can describe the observed data, but they are not a substitute for the paired contrast that answers the randomized comparison. The registry's absence of a reported primary contrast makes that distinction especially visible.

The record also illustrates why denominators and analysis populations matter. Overall enrollment is 35, while the primary results report 27 observations for each treatment condition and the admission outcome reports 31 and 30 participants. Without participant flow and missing-data details, readers cannot determine how the posted results relate to everyone enrolled or whether incomplete measurements might affect interpretation.

Precision is another central issue. The primary outcome is a mean adherence ratio, and the registry reports confidence intervals for each condition. However, the uncertainty of a treatment difference depends on the joint pattern of the paired observations, which the separate intervals do not describe. The heart failure admission comparison, based on sparse counts, also has a wide range of values compatible with the estimate. These features limit how much can be concluded from the observed pattern.

Finally, the trial includes multiple secondary outcomes, but no multiplicity strategy or analysis plan is reported. A broad set of measurements can be valuable for characterizing a pilot study, while still requiring restraint when interpreting isolated secondary results. The statistical importance of the trial is therefore not only the observed adherence means; it is also the need for transparent reporting of paired estimates, analysis populations, missing-data handling, and the intended role of secondary outcomes.

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 principal limitation in interpreting the primary result is that the registry reports separate treatment-period means and confidence intervals, but not a paired treatment estimate or its uncertainty. Since the study is crossover, the within-participant comparison is central to evaluating the treatment contrast. The method used to account for the paired data is not reported in the registry record.

Enrollment was 35, while the primary condition-specific results each report 27 observations. The admission outcome reports 31 participants in one condition and 30 in the other. Participant flow, reasons for unavailable measurements, and the handling of missing data are not reported. This makes it difficult to assess the analysis population and whether the posted summaries could be affected by missing observations.

The registry record does not report details of the randomized treatment sequences, any washout period, or methods for assessing period, sequence, or carryover effects. These are specific considerations in interpreting crossover evidence. No claim can be made from the registry record that such effects occurred, but their handling is not documented in the posted information.

The study is described as a single-center pilot, and the registry does not report the achieved numbers of participants with and without HIV. It is therefore not possible to assess the subgroup composition or the precision of any subgroup results. The primary outcome uses pill count, which is a particular measure of adherence; additional adherence measures are listed among the secondary outcomes, but the primary estimate does not establish medication-taking behavior in every respect.

The record does not report a statistical analysis plan, sample-size justification, interim analysis plan, or multiplicity adjustment. It also records zero reported analyses. Finally, the available safety counts are descriptive and do not include a reported statistical comparison or treatment-relatedness assessment. These reporting gaps limit assessment of the robustness, generalizability, and safety interpretation of the results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Heart Failure Polypill at a Safety Net Hospital — NCT06029712.
• Registry record last updated 2025-12-31; status last verified 2025-12.
• Results first posted 2025-12-31.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT06029712, titled “Heart Failure Polypill at a Safety Net Hospital,” is a completed Phase 2 randomized, unmasked crossover pilot trial sponsored by the University of California, San Francisco. Enrollment was 35. The registry lists a start date of 2024-02-27 and a primary completion date of 2024-09-17.

The primary outcome was mean adherence to guideline-directed medical therapy by pill count after a month in each treatment condition. The posted mean was 74.6 for individual tablets and 83.3 for the polypill, with a 95% confidence interval for each mean. A direct paired treatment comparison is not reported in the registry record. Secondary outcomes include heart failure admission, treatment satisfaction, health status, component-level adherence, blood pressure, heart rate, and weight. The record also reports condition-specific adverse-event counts and zero deaths in both conditions.

The registry reports 20 outcome measures, 17 posted secondary outcomes, and zero reported analyses. Details of the primary comparison, missing-data handling, and multiplicity procedures are not reported.