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 Efficacy and Safety of LCZ696 Compared to Valsartan, on Morbidity and Mortality in Heart Failure Patients With Preserved Ejection Fraction (NCT01920711), a randomized phase 3 study. The primary outcome is Cumulative Number of Primary Composite Events of Cardiovascular (CV) Death and Total (First and Recurrent) HF Hospitalizations..
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01920711 |
| Phase | Phase 3 |
| Condition | Heart Failure With Preserved Ejection Fraction |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 4822 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Novartis Pharmaceuticals |
| Results posted | Yes |
2. Clinical Question
Population
Patients with heart failure and preserved ejection fraction, described as New York Heart Association (NYHA) Class II-IV. The primary endpoint description specifies left ventricular ejection fraction (LVEF) of at least 45% and identifies the Full Analysis Set as the primary efficacy population.
Intervention
LCZ696, following a single-blind run-in that included valsartan and then LCZ696 before randomization. The registry describes the double-blind treatment period as lasting up to 57 months.
Comparator
Valsartan, an active treatment comparator. The registry describes the same run-in sequence in both arm descriptions; the full run-in eligibility criteria are not reported in the registry record.
Primary question
Whether the rate of the composite of cardiovascular death and total heart failure hospitalizations, including first and recurrent hospitalizations, differs between LCZ696 and valsartan during follow-up.
PARAGON-HF was designed to compare the two treatments on morbidity and mortality in patients with heart failure with preserved ejection fraction. Its primary endpoint combines cardiovascular death with heart failure hospitalizations. That question is broader than whether a treatment changes only the chance of a first hospitalization: the endpoint also includes recurrent hospitalizations, and it counts cardiovascular deaths as part of the composite.
The registry identifies this as a randomized, double-blind, parallel-group, active-controlled phase 3 study. It records enrollment of 4822. The primary endpoint description says the treatment arm with the lower event rate would be considered to have a successful response. However, the registry record does not report a primary efficacy estimate, confidence interval, or test result. The direction and precision of the observed comparison therefore cannot be assessed from the registry information.
The registry lists additional outcomes involving the Kansas City Cardiomyopathy Questionnaire, NYHA functional class, a composite renal endpoint, and all-cause mortality. Their inclusion indicates that the study assessed more than the primary event-count question, but their numerical results and analysis details are not reported in the registry record.
3. Trial Design
PARAGON-HF is a phase 3, randomized, double-blind, parallel-group study with an active comparator. The registry identifies Novartis Pharmaceuticals as the sponsor and lists the condition as heart failure with preserved ejection fraction. The recorded start date is 2014-07-18, primary completion date is 2019-06-07, and status is completed. These dates describe the registry record and do not, by themselves, establish when results were analyzed or reported.
The registry records enrollment of 4822. Its arm descriptions specify a single-blind run-in lasting 3-8 weeks, followed by a double-blind period lasting up to 57 months. During the run-in, patients were to start on valsartan 80 mg b.i.d. For 1-2 weeks and then LCZ696 100 mg b.i.d. For 2-4 weeks before randomization. The arm descriptions end before the full run-in eligibility criteria are given, so those criteria are not reported in the registry record.
A run-in can have design implications beyond the randomized comparison. In general, when eligibility for randomization depends on completing a preliminary treatment period, the randomized participants may represent a selected group. The available record does not provide enough detail to establish the precise selection process or its consequences in this study. The randomized comparison should therefore be understood as applying to the population that entered the trial's primary efficacy analysis, identified as the Full Analysis Set.
The primary outcome was a cumulative count of cardiovascular deaths and total heart failure hospitalizations, with first and recurrent hospitalizations included, over total follow-up time up to 57 months. This differs from a simple binary outcome that records only whether a participant experienced at least one event. The registry also lists 5 outcome measures and 10 analyses, but the mapping between them, analysis specifications, and numerical results are not reported in the registry record.
The adverse-event record lists arm-specific counts and an all-patients row. Its reported all-patients risk population is 4821, whereas overall enrollment is recorded as 4822. The registry does not explain this difference. The adverse-event populations should therefore be read according to the figures and labels recorded for those data, rather than assumed to be identical to the enrollment count.
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 | LCZ696 | Valsartan |
|---|---|---|
| STARTED | 2419 | 2403 |
| COMPLETED | 2055 | 2030 |
| NOT COMPLETED | 364 | 373 |
Reasons for not completing this period:
| Reason | LCZ696 | Valsartan |
|---|---|---|
| Death | 347 | 356 |
| Lost to Follow-up | 13 | 14 |
| Withdrawal by Subject | 4 | 3 |
Analysis population for the primary outcome, as defined in the registry: Full Analysis Set - This was the primary efficacy population applied in efficacy analyses for all efficacy endpoints.
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: Randomized set totaled 4822 patients. Full Analysis Set consisted of 4796 patients: primary efficacy population used in the analyses for all efficacy endpoints, includes all randomized patients minus 26 patients excluded due to GCP violations. Safety set included all randomized patients minus 1 patient who did not receive study drug: 4821 patients
| Characteristic | Unit | LCZ696 | Valsartan | Total |
|---|---|---|---|---|
| Participants analysed | 2419 | 2403 | 4822 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 415 | 415 | 830 |
| Age, Categorical: >=65 years | Participants | 2004 | 1988 | 3992 |
| Sex: Female, Male: Female | Participants | 1247 | 1244 | 2491 |
| Sex: Female, Male: Male | Participants | 1172 | 1159 | 2331 |
| Race/Ethnicity, Customized: Caucasian | Participants | 1975 | 1958 | 3933 |
| Race/Ethnicity, Customized: Black | Participants | 52 | 50 | 102 |
| Race/Ethnicity, Customized: Asian | Participants | 297 | 310 | 607 |
| Race/Ethnicity, Customized: Native American | Participants | 28 | 23 | 51 |
| Race/Ethnicity, Customized: Pacific Islander | Participants | 0 | 1 | 1 |
| Race/Ethnicity, Customized: Other | Participants | 67 | 61 | 128 |
6. Primary Endpoint
The registry lists the primary outcome as Cumulative Number of Primary Composite Events of Cardiovascular (CV) Death and Total (First and Recurrent) HF Hospitalizations., measured over Total follow up time (up to 57 months).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | Events |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the cumulative number of cardiovascular deaths and total heart failure hospitalizations, including both first and recurrent hospitalizations. The registry gives the measurement unit as events and defines the timeframe as total follow-up time up to 57 months. It identifies the primary efficacy population as the Full Analysis Set.
Combining cardiovascular death and hospitalization represents more than one type of clinical event in a single outcome. Counting recurrent hospitalizations also retains information that a first-event-only analysis would discard. In general, this kind of endpoint can reflect the burden of repeated events, but it requires analysis methods that account for the fact that events from the same participant are related rather than independent observations. The registry record does not report the method used to analyze the count outcome.
The endpoint description frames the comparison in terms of event rates and says the arm with the lower rate would be considered to have a successful response. A cumulative event count alone is not enough to establish the size or precision of a treatment comparison. Interpretation would ordinarily depend on an effect estimate and its uncertainty, along with the analysis method and relevant population definition. Those primary efficacy results are not reported in the registry record.
This is a composite endpoint: a reported overall comparison would not, by itself, show which component contributed most to any difference. Component-specific results, including separate results for cardiovascular death and heart failure hospitalizations, are not reported in the registry record. The listed secondary outcome of all-cause mortality is distinct from the primary endpoint's cardiovascular-death component.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record gives the primary endpoint definition and study design, but it does not report a numerical primary efficacy result. There is no posted arm-specific primary event count or rate, treatment-effect estimate, confidence interval, or p-value in the registry record recorded for this analysis. As a result, neither the direction nor the magnitude nor the statistical precision of the primary comparison can be described.
The primary endpoint is a count of cardiovascular deaths and total heart failure hospitalizations, including recurrent hospitalizations. Its interpretation would depend on how the count was modeled, how follow-up time was handled, and how within-participant recurrence was addressed. Those analytical details are not reported in the registry record. It would be inappropriate to infer a treatment effect from the endpoint definition, trial completion status, or adverse-event counts.
In general, a confidence interval describes a range of effect values compatible with the data under the specified statistical model and assumptions. A narrow interval indicates greater statistical precision than a wide interval, but neither width nor direction can be assessed without an estimate and interval. A p-value, if reported, would address compatibility with a null hypothesis under a particular analysis; it would not establish clinical importance on its own. No such primary comparison statistics are reported in the registry record.
The registry describes an active-controlled, randomized, double-blind design. Randomization is intended to make treatment groups comparable on average at baseline, while blinding can reduce some sources of differential measurement or behavior. These are general design principles, not evidence that all potential sources of bias were eliminated. The record does not report enough information to assess adherence, loss to follow-up, outcome adjudication, or implementation of blinding.
Because the endpoint combines cardiovascular death and recurrent hospitalizations, an overall composite result, if available, would not establish that each component moved in the same direction. Component-specific estimates would be needed to understand their contributions. The record lists all-cause mortality as a secondary outcome, but numerical results for that outcome and for the other secondary outcomes are not reported in the registry record.
The adverse-event entries describe counts of participants affected and the most frequently listed serious adverse event terms. They do not substitute for the primary efficacy analysis, and the reported deaths in that portion of the record are not identified as cardiovascular deaths. The registry does not report a comparative analysis of these safety counts, nor does it provide enough detail to infer causation from the counts alone.
Overall, the posted registry information supports a description of the question, design, endpoint, and selected safety counts. It does not support a conclusion about the relative efficacy of LCZ696 and valsartan, the certainty of an efficacy estimate, or the balance of benefits and harms. Those conclusions would require results and analysis details not reported in the registry record.
9. Statistical Methods Explained
The statistical meaning of this trial depends on how its event-count endpoint was analyzed. The registry describes the endpoint as the cumulative number of cardiovascular deaths and total heart failure hospitalizations, including first and recurrent hospitalizations. It does not report the statistical model, effect measure, variance estimate, confidence interval, or hypothesis test used for the primary comparison. Without those details, the analysis cannot be reconstructed or evaluated from the registry record.
Why does a recurrent-event endpoint need special care?
General statistical explanation: when one participant can contribute multiple events, the observations are not necessarily independent. A participant who has one hospitalization may also be more likely to have another, so treating every event as if it came from a different person can understate uncertainty. Methods for recurrent events may account for within-participant dependence, follow-up time, or both. The appropriate method depends on the question and data structure. The registry record does not identify which approach was used.
Death also matters to the interpretation of hospitalization counts. General statistical explanation: a participant who dies can no longer experience later hospitalizations, so death can affect the opportunity to observe recurrent events. A composite that includes cardiovascular death addresses both event types in one outcome definition, but does not remove the need to explain the analysis strategy. The registry does not describe how death was incorporated into the primary statistical model, beyond including cardiovascular death in the endpoint.
What would an effect estimate and confidence interval contribute?
General statistical explanation: an effect estimate summarizes the contrast between randomized groups using a specified scale, while a confidence interval communicates uncertainty around that estimate under the model. For an event-count comparison, the estimate might be expressed using a rate-based or other model-dependent measure, but the specific measure used in this trial is not reported in the registry record. No primary estimate or confidence interval is reported, so the magnitude and precision of the comparison cannot be discussed.
A significance test answers a narrower question than an effect estimate: it assesses how compatible the observed data are with a null hypothesis, given the chosen model and assumptions. A p-value does not describe the size of an effect, and statistical significance alone would not establish clinical importance. The registry record does not report a primary p-value or its testing procedure.
What does the Full Analysis Set tell the reader?
The registry labels the Full Analysis Set as the primary efficacy population and says it was applied in efficacy analyses for all efficacy endpoints. This establishes the named population for the efficacy analyses, but the registry record does not give its operational definition, the number included in it, or how participants with missing outcome information were handled. The population label alone does not establish that every randomized participant contributed complete data or that all participants were analyzed under an identical missing-data rule.
General statistical explanation: analysis-population definitions can affect the interpretation of a randomized trial. An analysis that retains participants according to assigned group is often intended to preserve the randomized comparison, while other populations may focus on treatment exposure or protocol adherence. The registry record does not describe additional analysis populations or specify which, if any, were used for sensitivity analyses. It would be inappropriate to assign a more detailed meaning to the Full Analysis Set beyond the registry's description.
How should the event results be checked?
For a count endpoint, useful reporting ordinarily includes the group-specific event information, the comparison measure, the analysis population, follow-up handling, and uncertainty around the contrast. Recurrent events may also call for a description of how multiple events within a participant were accounted for. These are general reporting considerations, not claims about the study's actual analysis. The registry record does not report these primary analysis details or results.
The registry lists 5 outcome measures and 10 analyses. Those counts indicate that multiple outcomes or analyses were registered, but they do not reveal how many tests addressed the primary question, which analyses were primary or secondary, or whether any were prespecified sensitivity analyses. That analysis mapping is not reported in the registry record.
What is a planning illustration?
General statistical explanation: sample-size planning asks how much information a future study might need to detect a specified effect under assumptions about the outcome, variability, significance level, allocation, and analysis. A planning calculation based on an observed effect is an illustration, not evidence that the original study was designed around that effect. It is also not observed power, which is not a useful substitute for examining an effect estimate and its uncertainty after a trial.
The planning note associated with the record describes a two-sided significance level of 0.05, equal allocation, and a normal approximation. It says that the effect to detect is a stated fraction of the observed effect, and distinguishes total events for time-to-event illustrations from participants per arm for other outcomes. The numeric planning outputs and a numeric observed effect are not reported in the registry record used for this analysis. No planning sample-size value can therefore be quoted.
In general, over-dispersion occurs when count data vary more than a simple model assumes. If present and not accommodated, it can lead to uncertainty estimates that are too small. Whether over-dispersion was assessed or how it was handled in PARAGON-HF is not reported in the registry record. This is one reason that an outcome definition by itself is not enough to judge the statistical analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 5 outcome measures and 10 analyses. It also identifies the composite count of cardiovascular death and total heart failure hospitalizations as the primary endpoint, with several secondary outcomes: change in the Clinical Summary Score on the Kansas City Cardiomyopathy Questionnaire at Month 8; change in NYHA functional class at Month 8; first occurrence of a composite renal endpoint; and all-cause mortality. These entries show that the study addressed multiple outcomes, but do not specify the full analysis hierarchy.
Multiplicity arises when several outcomes, time points, comparisons, or analyses are tested. General statistical explanation: the more inferential tests considered, the greater the chance of at least one apparently unusual result under repeated testing if no adjustment or prespecified hierarchy is used. Trials may address this through a testing sequence, adjusted significance thresholds, or other strategies. The registry record does not report whether a multiplicity procedure was used, how outcomes were prioritized, or which tests were confirmatory.
The record does not report a statistical analysis plan, interim analyses, missing-data methods, or sensitivity analyses. It also does not describe outcome-specific participant counts or the handling of incomplete follow-up. Those omissions limit the ability to assess whether the reported analysis, if encountered elsewhere, follows prespecified methods or is sensitive to missing observations. No missing-data mechanism should be inferred from the registry entries alone.
The primary efficacy population is named as the Full Analysis Set, but the record does not give its detailed definition. That limits assessment of how the population relates to randomized assignment and whether participants with incomplete information were included. The registry's completion status does not resolve these questions.
11. 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 |
|---|---|---|---|
| LCZ696 | 1424 | 2419 | 58.9% |
| Valsartan | 1416 | 2402 | 59.0% |
| All Patients | 2840 | 4821 | 58.9% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| LCZ696 | 347 of 2419 |
| Valsartan | 357 of 2402 |
| All Patients | 704 of 4821 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 2% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| LCZ696 | 2005 of 2419 |
| Valsartan | 1995 of 2402 |
| All Patients | 4000 of 4821 |
Most frequent serious adverse event terms
| Event term | Organ system | LCZ696 | Valsartan | All Patients |
|---|---|---|---|---|
| Cardiac failure | Cardiac disorders | 340 of 2419 | 380 of 2402 | 720 of 4821 |
| Pneumonia | Infections and infestations | 162 of 2419 | 178 of 2402 | 340 of 4821 |
| Atrial fibrillation | Cardiac disorders | 162 of 2419 | 145 of 2402 | 307 of 4821 |
| Acute kidney injury | Renal and urinary disorders | 90 of 2419 | 110 of 2402 | 200 of 4821 |
| Cardiac failure congestive | Cardiac disorders | 86 of 2419 | 83 of 2402 | 169 of 4821 |
| Cardiac failure acute | Cardiac disorders | 85 of 2419 | 77 of 2402 | 162 of 4821 |
| Anaemia | Blood and lymphatic system disorders | 68 of 2419 | 67 of 2402 | 135 of 4821 |
| Urinary tract infection | Infections and infestations | 54 of 2419 | 68 of 2402 | 122 of 4821 |
| Acute myocardial infarction | Cardiac disorders | 60 of 2419 | 54 of 2402 | 114 of 4821 |
| Chronic obstructive pulmonary disease | Respiratory, thoracic and mediastinal disorders | 42 of 2419 | 67 of 2402 | 109 of 4821 |
Most frequent non-serious adverse event terms
| Event term | Organ system | LCZ696 | Valsartan | All Patients |
|---|---|---|---|---|
| Hypotension | Vascular disorders | 529 of 2419 | 379 of 2402 | 908 of 4821 |
| Renal impairment | Renal and urinary disorders | 287 of 2419 | 331 of 2402 | 618 of 4821 |
| Hypertension | Vascular disorders | 225 of 2419 | 314 of 2402 | 539 of 4821 |
| Hyperkalaemia | Metabolism and nutrition disorders | 238 of 2419 | 298 of 2402 | 536 of 4821 |
| Urinary tract infection | Infections and infestations | 243 of 2419 | 268 of 2402 | 511 of 4821 |
| Cardiac failure | Cardiac disorders | 225 of 2419 | 269 of 2402 | 494 of 4821 |
| Atrial fibrillation | Cardiac disorders | 248 of 2419 | 239 of 2402 | 487 of 4821 |
| Dyspnoea | Respiratory, thoracic and mediastinal disorders | 217 of 2419 | 236 of 2402 | 453 of 4821 |
| Dizziness | Nervous system disorders | 229 of 2419 | 192 of 2402 | 421 of 4821 |
| Nasopharyngitis | Infections and infestations | 207 of 2419 | 177 of 2402 | 384 of 4821 |
Serious events with the highest arm-level rate: cardiac failure (15.8%), pneumonia (7.4%), atrial fibrillation (6.7%), acute kidney injury (4.6%), cardiac failure congestive (3.6%).
The adverse-event record reports serious adverse events and other adverse events by treatment group, along with a row labeled All Patients. For LCZ696, it lists 2419 at risk, 1424 affected by serious adverse events, 2005 affected by other adverse events, and 347 deaths. For valsartan, it lists 2402 at risk, 1416 affected by serious adverse events, 1995 affected by other adverse events, and 357 deaths. The all-patients row reports 4821 at risk, 2840 affected by serious adverse events, 4000 affected by other adverse events, and 704 deaths.
These are registry counts, not a comparative safety analysis. The record does not specify in this information the observation period or statistical method for comparing these outcomes. The deaths are not identified as cardiovascular deaths, and should not be treated as the primary endpoint's cardiovascular-death component. The all-patients risk count is 4821, while enrollment is listed as 4822; the registry does not explain this difference.
The leading serious adverse event terms are listed as Cardiac failure at 15.8%, Pneumonia at 7.4%, Atrial fibrillation at 6.7%, Acute kidney injury at 4.6%, Cardiac failure congestive at 3.6%, and Cardiac failure acute at 3.5%. The registry record does not explain the denominator or provide a treatment-group breakdown for these term percentages in the information recorded. They should not be interpreted as comparative risks or attributed to treatment without additional details.
12. What the Result Does, and Does Not, Mean
The record establishes that PARAGON-HF was a randomized, double-blind, active-controlled phase 3 study comparing LCZ696 with valsartan in patients with heart failure and preserved ejection fraction. It identifies a primary composite endpoint that includes cardiovascular death and total heart failure hospitalizations, including recurrent hospitalizations. It also reports selected adverse-event counts and lists secondary outcomes.
What the record does not establish is the primary treatment effect. The primary efficacy estimate, its confidence interval, and its statistical test are not reported in the registry record. Without them, the relative event rate cannot be described, the direction of the comparison cannot be stated, and its precision cannot be assessed. Trial completion and enrollment do not fill that information gap.
Even if a composite comparison were available, it would not automatically show that cardiovascular death and hospitalizations contributed equally or changed in the same direction. Component-specific results would be needed for that interpretation. Nor would a result on the primary composite answer every secondary question about symptoms, functional class, renal outcomes, or all-cause mortality. Numerical secondary outcome results are not reported in the registry record.
The adverse-event counts provide descriptive information but do not establish a treatment-related cause or a statistically supported between-group difference. The record does not provide the comparative analysis needed to make those judgments. It also does not give enough information to assess the effect of missing data, adherence, or the run-in selection process on the efficacy analysis.
In general, a trial result is most informative when the estimated effect is considered alongside its uncertainty, the endpoint definition, analysis population, missing-data approach, and prespecified testing strategy. Several of these details are not reported in the registry record. Accordingly, the available information supports a careful description of the trial and its registered question, but not a conclusion about comparative efficacy or a complete benefit-risk assessment.
13. Why This Trial Matters Statistically
This trial is statistically notable for using a recurrent-event composite rather than an endpoint limited to the first hospitalization. The primary outcome combines cardiovascular death with total heart failure hospitalizations, counting first and recurrent hospitalizations. That choice makes the estimand—the precise treatment contrast a study aims to estimate—important to interpretation. In general, a total event count asks about a different aspect of outcome burden than the time until a first event.
Repeated observations from one participant are related. General statistical explanation: methods for recurrent events need to reflect that dependence, and choices about follow-up and death can affect what the estimate means. The registry gives the endpoint definition but does not report the model or effect measure. This makes the study a useful example of why a clinically clear endpoint description is not a substitute for statistical analysis details.
The design also includes a single-blind run-in before randomization, followed by a double-blind treatment period. Such a sequence can matter for the population to which a randomized comparison applies, particularly if trial entry after the run-in depends on eligibility or tolerability. Here, the run-in criteria are not fully reported in the registry record, so the implications cannot be quantified or fully assessed.
Finally, the registry lists several outcomes and analyses, while its record does not specify the multiplicity strategy or provide the primary efficacy statistics. This illustrates why outcome hierarchy, missing-data methods, analysis populations, and uncertainty estimates are essential parts of transparent trial reporting. The available record supplies enough information to identify the central statistical question, but not to judge the result of the primary comparison.
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 largest limitation for interpretation is that the registry record does not report the primary efficacy result. It provides neither a group comparison nor an effect estimate, confidence interval, or p-value. The primary outcome is described as a cumulative event count, but the analysis method and its assumptions are not reported. The record therefore cannot establish the direction, magnitude, or precision of the primary comparison.
The record does not describe the statistical analysis plan, interim analyses, multiplicity adjustment, or missing-data strategy. It lists 5 outcome measures and 10 analyses, but does not map the analyses to outcomes or explain their inferential status. The Full Analysis Set is identified as the primary efficacy population, but its operational definition and size are not reported in the registry record.
The run-in description gives the treatment sequence and duration, but ends before the full eligibility criteria are stated. This limits assessment of who proceeded to randomization and how the randomized population relates to all patients entering the run-in. No conclusion about selection or its effect on the comparison can be drawn from the incomplete description alone.
For safety, the record includes descriptive counts and serious adverse event term percentages, but does not report a comparative statistical analysis or clarify the denominator for the term percentages in the information recorded. The all-patients risk count of 4821 differs from the enrollment figure of 4822, without an explanation in the registry record. These limitations prevent a complete assessment of safety comparisons and of the relationship between the efficacy and adverse-event populations.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Efficacy and Safety of LCZ696 Compared to Valsartan, on Morbidity and Mortality in Heart Failure Patients With Preserved Ejection Fraction — NCT01920711.
- Registry record last updated 2020-09-29; status last verified 2020-09.
- Results first posted 2020-09-29.
- 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.
18. Record Summary
PARAGON-HF (NCT01920711) is recorded as a completed, randomized, double-blind, parallel-group, active-controlled phase 3 study sponsored by Novartis Pharmaceuticals. It compared LCZ696 with valsartan in patients with heart failure with preserved ejection fraction. Enrollment is listed as 4822; the recorded start date is 2014-07-18 and primary completion date is 2019-06-07.
The primary endpoint was the cumulative number of cardiovascular deaths and total heart failure hospitalizations, including recurrent hospitalizations, over follow-up up to 57 months. The Full Analysis Set is named as the primary efficacy population. The registry also lists secondary outcomes concerning the Kansas City Cardiomyopathy Questionnaire, NYHA functional class, a composite renal endpoint, and all-cause mortality.
The registry record does not report the primary efficacy estimate or its statistical uncertainty, nor does it describe the analysis model, missing-data methods, multiplicity strategy, or full analysis plan. It reports descriptive adverse-event counts and serious-event terms. These details identify the trial question and some safety information but do not establish the comparative efficacy result.