This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial facts on this page are restricted to the ClinicalTrials.gov record. The registry identifies the primary endpoint and confirms that results have been posted, but the ClinicalTrials.gov recordset does not contain formal effect estimates, confidence intervals, or p-values for that endpoint.
1. Trial at a Glance
PARADIGM-HF was a randomized, double-blind, phase 3 parallel-group trial evaluating LCZ696 200 mg BID versus enalapril 10 mg BID in patients with heart failure with reduced ejection fraction. The registered primary endpoint was the first occurrence of a composite of cardiovascular death or heart failure hospitalization.
| Feature | PARADIGM-HF |
|---|---|
| Trial name | PARADIGM-HF |
| NCT identifier | NCT01035255 |
| Phase | Phase 3 |
| Condition | Heart Failure With Reduced Ejection Fraction |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Primary purpose | Treatment |
| Enrollment | 8442 |
| Interventions | LCZ696 200 mg BID; Enalapril 10 mg BID |
| Trial status | Terminated |
| Lead sponsor | Novartis Pharmaceuticals |
| Sponsor type | Industry |
| Start | 2009-12 |
| Primary completion | 2014-05 |
2. Clinical Question
The registered clinical question is whether LCZ696 and enalapril differ with respect to morbidity and mortality in patients with heart failure with reduced ejection fraction, with the primary endpoint defined as the first occurrence of cardiovascular death or heart failure hospitalization due to heart failure.
Population
Patients with heart failure with reduced ejection fraction.
Intervention
LCZ696 200 mg BID.
Comparator
Enalapril 10 mg BID.
Primary question
How does the randomized comparison affect the first occurrence of cardiovascular death or heart failure hospitalization due to heart failure?
3. Trial Design
LCZ696
- LCZ696 200 mg BID
- Drug intervention
Enalapril
- Enalapril 10 mg BID
- Drug intervention
4. Endpoints
The registry identifies one primary endpoint. It is a time-bounded composite event endpoint expressed in the registry as a number of participants experiencing the first occurrence of either cardiovascular death or heart failure hospitalization.
| Endpoint | Registered definition | Time frame | Endpoint type |
|---|---|---|---|
| First occurrence of the composite endpoint | Number of participants that had first occurrence of the composite endpoint, which is defined as either cardiovascular (CV) death or heart failure (HF) hospitalization due to HF. | Up to 51 months | Binary |
5. Statistical Methodology
The ClinicalTrials.gov record confirms a randomized, parallel, double-blind phase 3 design and a primary composite endpoint assessed over up to 51 months. However, no formal statistical analyses were posted. Consequently, this page does not assign a hazard ratio, confidence interval, p-value, event count, or other treatment-effect estimate to the primary endpoint.
Why the primary endpoint is naturally a time-to-event problem
The registry definition centers on the first occurrence of cardiovascular death or heart failure hospitalization and specifies follow-up of up to 51 months. This structure contains more information than a simple yes/no outcome at one fixed visit because participants may experience the event at different times and may have different amounts of observable follow-up.
Here, T represents the time to the first occurrence of the composite endpoint. The survival function describes the probability of remaining free of the first event beyond time t.
Kaplan-Meier estimation
For a first-event endpoint followed over several months or years, Kaplan-Meier estimation is a standard way to describe the event-free experience over time. It allows participants who have not experienced the endpoint by their last available observation to contribute information up to that point through censoring.
Log-rank comparison
A randomized comparison of time-to-first-event curves can be performed with a log-rank test. A stratified version can be used when prespecified randomization or analysis strata need to be incorporated. The ClinicalTrials.gov record does not specify whether a log-rank test, a stratified log-rank test, or another formal comparison was used for PARADIGM-HF, so no specific test is attributed to the registry.
Cox proportional-hazards modeling
A Cox proportional-hazards model is commonly used to summarize a treatment effect for a time-to-event endpoint through a hazard ratio. If such an analysis were performed, the model would compare the estimated instantaneous event rates between randomized treatment groups while accounting for the observed event times and censoring.
An HR below 1 would indicate a lower estimated instantaneous event rate under LCZ696 relative to enalapril within the model. It would not by itself describe an absolute difference in the probability of experiencing the composite endpoint.
Composite endpoint analysis
The primary endpoint combines two clinically distinct events: cardiovascular death and heart failure hospitalization due to heart failure. A composite can increase the number of observed first events and therefore provide more statistical information than either component alone. At the same time, interpretation depends on the treatment effect across the individual components and on how clinically comparable those components are.
6. Planned Analysis
The registry states that results have been posted and that the primary endpoint has a reported results entry. However, the ClinicalTrials.gov record contains no formal statistical analyses for that endpoint. Accordingly, there is no supported numerical treatment comparison to reproduce on this page.
Primary endpoint analysis status
The ClinicalTrials.gov record contains no hazard ratio, confidence interval, p-value, or comparative event count for the primary endpoint.
For this type of endpoint, a complete statistical analysis would ordinarily retain the time from randomization to the first occurrence of cardiovascular death or heart failure hospitalization, identify participants who remain event-free at their last observation, and use an appropriate time-to-event comparison between randomized groups. A treatment-effect model such as a Cox proportional-hazards model could then provide a hazard ratio with a confidence interval.
7. Safety
The ClinicalTrials.gov record includes serious adverse events by randomized arm. These are presented as affected participants divided by participants at risk.
| Safety measure | LCZ696 | Enalapril |
|---|---|---|
| Serious adverse events | 1937 / 4203 | 2142 / 4229 |
The numerator and denominator are reported exactly as reported in the registry by the trial data. Because no formal comparative safety analysis is reported, the figures should be treated as descriptive arm-level counts rather than as evidence of a statistically tested between-group difference.
8. Statistical Methods Explained
Why is a first-occurrence composite endpoint suitable for time-to-event analysis?
A first-occurrence endpoint records the earliest qualifying event after randomization. Because participants can experience that event at different times, the analysis can use the timing of the event rather than only whether an event ever occurred. This is particularly relevant when follow-up extends to "up to 51 months."
What does a hazard ratio mean for this trial?
A hazard ratio would summarize the relative instantaneous rate of experiencing the first cardiovascular death or heart failure hospitalization between LCZ696 and enalapril under a specified time-to-event model. An HR below 1 would represent a lower estimated event rate in the LCZ696 group, while an HR above 1 would represent a higher estimated event rate. A hazard ratio is not an absolute risk difference.
Why is the composite endpoint important?
The endpoint combines cardiovascular death and heart failure hospitalization due to heart failure. Combining events can increase the number of primary endpoint events available for analysis, but the resulting treatment effect represents the composite rather than automatically describing either component individually.
What would censoring mean here?
Censoring occurs when a participant's event-free follow-up ends before the primary endpoint is observed. For example, a participant who remains free of the composite endpoint through the last available observation contributes information through that time but does not provide an observed event time afterward. The ClinicalTrials.gov record does not describe the censoring rules, so this page does not impose a specific rule.
Why does randomization matter statistically?
Randomization creates the basis for comparing outcomes between treatment assignments without deliberately selecting patients into the two groups according to their prognosis. In a properly conducted randomized comparison, the treatment assignment is the principal variable being compared, while subsequent treatment exposure and follow-up can introduce additional considerations for analysis.
Why is the absence of a reported p-value important?
A p-value is a property of a specified statistical test and its null hypothesis. Because the ClinicalTrials.gov recordset does not provide the formal comparison, there is no supported p-value to report.
What does a confidence interval add?
If a hazard ratio or another treatment-effect estimate were reported, its confidence interval would describe statistical uncertainty around that estimate under the specified analysis framework. It would not describe the range of individual patient responses. Because no formal effect estimate is reported here, there is no confidence interval to interpret for the primary endpoint.
9. Analysis Population and Interpretation
The trial enrolled 8442 participants and used randomized allocation. The ClinicalTrials.gov record does not specify the detailed efficacy analysis population, per-protocol population, censoring set, or treatment-exposure rules used for the formal primary analysis.
| Population / feature | What the ClinicalTrials.gov record establishes |
|---|---|
| Overall enrollment | 8442 participants |
| Randomization | Randomized |
| Study arms | 2 |
| Design model | Parallel |
| Masking | Double |
| Formal efficacy analysis population | Not specified in the ClinicalTrials.gov record |
| Formal safety analysis population | Not specified in the ClinicalTrials.gov record; serious AE denominators are reported |
This distinction matters because an overall enrollment count is not necessarily identical to the denominator used for every analysis. The registry-reported serious-adverse-event denominators illustrate that analysis-specific populations can differ from the total enrollment.
10. Missing Data, Censoring, and Follow-Up
The primary endpoint has a follow-up window of up to 51 months. For a time-to-first-event endpoint, participants can have different observed follow-up durations. Proper analysis therefore requires explicit rules for censoring and follow-up termination.
Event observation
A participant who experiences cardiovascular death or heart failure hospitalization contributes the time to the first qualifying event.
Event-free follow-up
A participant who remains free of the endpoint contributes information through the last time at which event status is known.
Competing events
Because cardiovascular death is itself part of the composite, it is an endpoint event rather than simply a censoring event for the composite.
Registry limitation
The ClinicalTrials.gov record does not specify the detailed missing-data or censoring rules used in the formal analysis.
It would be inappropriate to infer an imputation method for the primary endpoint from the registry summary. Time-to-event analyses typically do not impute an unobserved event time simply because follow-up ends; instead, censoring rules define how the available follow-up contributes to the estimator.
11. Multiplicity and Statistical Error
The ClinicalTrials.gov record identifies one registered primary endpoint. They do not provide an alpha allocation, multiplicity hierarchy, interim-analysis schedule, stopping boundary, or detailed type I error-control strategy.
| Design topic | Supported information |
|---|---|
| Primary endpoint count | 1 |
| Primary endpoint type | Binary |
| Interim analysis | Not specified in the ClinicalTrials.gov record |
| Alpha-spending method | Not specified in the ClinicalTrials.gov record |
| Multiplicity adjustment | Not specified in the ClinicalTrials.gov record |
| Non-inferiority margin | Not specified in the ClinicalTrials.gov record |
| Bayesian methods | Not specified in the ClinicalTrials.gov record |
Because these design features are not included in the ClinicalTrials.gov recordset, no claim is made here about whether the trial used an interim efficacy boundary, hierarchical testing, alpha spending, or another formal error-control procedure.
12. Trial Design Features That Affect Statistical Interpretation
Randomization
Randomized allocation provides the fundamental basis for comparing the two treatment assignments with respect to the primary endpoint.
Double blinding
Double masking can reduce the potential for knowledge of treatment assignment to influence participant or investigator behavior and assessment.
Parallel design
Each randomized participant belongs to one of two parallel treatment groups rather than serving as their own treatment control.
Composite outcome
The primary endpoint combines cardiovascular death with heart failure hospitalization due to heart failure.
13. Clinical Biostats Interpretation of the Primary Endpoint
The primary endpoint measures whether a participant experiences the first occurrence of either cardiovascular death or heart failure hospitalization due to heart failure during a follow-up period of up to 51 months.
The composite endpoint is not a measure of cardiovascular death alone and is not a measure of hospitalization alone. A result for the composite cannot, by itself, establish the magnitude of treatment effect on each component.
No confidence interval for the treatment effect is reported in the ClinicalTrials.gov record. Consequently, the precision of the formal comparative estimate cannot be characterized from the ClinicalTrials.gov recordset.
A p-value requires a defined statistical test, null hypothesis, test statistic, and analysis population. None of the formal primary-endpoint statistical-analysis records are reported here. Therefore, no p-value is reported.
A first-event endpoint observed over months or years contains timing and censoring information. A time-to-event method can use that information directly, whereas a simple binary comparison at a single arbitrary time point can discard part of the available follow-up structure.
14. Limitations
- Formal effect estimates are absent from the ClinicalTrials.gov record: the dataset contains no hazard ratio, confidence interval, p-value, or comparative event count for the primary endpoint.
- Results are posted, but the statistical-analysis record is not reported: the registry data confirm that results exist without providing the formal comparison needed for numerical interpretation here.
- Component-specific results are unavailable in the ClinicalTrials.gov record: the primary composite contains cardiovascular death and heart failure hospitalization due to heart failure, but no component-specific estimates are provided.
- Analysis population details are limited: the ClinicalTrials.gov record does not specify the exact efficacy population, censoring set, or formal safety-analysis population.
- Missing-data and censoring rules are not reported: no specific imputation or censoring strategy should be inferred.
- Multiplicity and interim monitoring are not specified: the ClinicalTrials.gov record does not document alpha spending, interim boundaries, or a multiplicity hierarchy.
- Serious adverse events are descriptive here: arm-level affected/at-risk counts are reported, but no formal comparative safety analysis is provided.
- Composite interpretation: a treatment effect on the composite should not automatically be interpreted as an equivalent effect on both component outcomes.
15. Why This Trial Matters Statistically
PARADIGM-HF is a useful statistical teaching case because its registered primary endpoint illustrates the transition from a simple binary concept to a longitudinal time-to-event analysis. The registry labels the endpoint as binary, but its definition as the first occurrence during up to 51 months makes event timing and censoring central to a rigorous analysis.
| Concept | How it appears in PARADIGM-HF |
|---|---|
| Randomization | Participants were randomly allocated to two treatment arms. |
| Double blinding | The trial used double masking. |
| Parallel-group design | LCZ696 and enalapril were evaluated in separate randomized groups. |
| Composite endpoint | The primary endpoint combines cardiovascular death and heart failure hospitalization due to heart failure. |
| First-event analysis | The endpoint is defined by the first occurrence of the composite event. |
| Time-to-event methods | The up-to-51-month follow-up structure naturally motivates survival-analysis methods. |
| Censoring | Participants without an observed first event may contribute event-free follow-up until observation ends. |
| Analysis populations | Overall enrollment and serious-AE denominators are separately reported. |
| Statistical uncertainty | A formal confidence interval would be needed to quantify uncertainty around a treatment-effect estimate. |
16. Related Tutorials
Learn more about the methods used in this trial:
17. Related Calculators
18. Sources
- ClinicalTrials.gov: NCT01035255 — PARADIGM-HF.
- PubMed record: PMID 40839760.
- PubMed record: PMID 40265590.
- PubMed record: PMID 39563094.
- PubMed record: PMID 39439294.
- PubMed record: PMID 39210725.
Continue exploring clinical-trial statistics
Connect the design and endpoint concepts in PARADIGM-HF with deeper biostatistics tutorials and statistical calculators.
19. Record Summary
PARADIGM-HF was a randomized, double-blind, parallel-group phase 3 trial enrolling 8442 participants with heart failure with reduced ejection fraction and comparing LCZ696 200 mg BID with enalapril 10 mg BID. Its registered primary endpoint was the first occurrence of a composite of cardiovascular death or heart failure hospitalization due to heart failure, assessed over up to 51 months.
The statistical structure is particularly instructive because a registry endpoint labeled as binary can still represent a longitudinal first-event process. A rigorous analysis would ordinarily consider event timing, censoring, and the randomized comparison rather than treating all participants as if they had identical follow-up.