This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
FIGARO-DKD was a randomized, parallel-group, quadruple-masked phase 3 treatment trial comparing finerenone with placebo in subjects with type 2 diabetes mellitus and the clinical diagnosis of diabetic kidney disease. The registry reports one primary endpoint and five additional posted statistical analyses.
| Feature | FIGARO-DKD |
|---|---|
| Trial name | FIGARO-DKD |
| ClinicalTrials.gov identifier | NCT02545049 |
| Phase | Phase 3 |
| Status | Completed |
| Therapeutic area | Nephrology |
| Condition | Diabetic Kidney Disease |
| Enrollment | 7352 |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Primary purpose | Treatment |
| Interventions | Finerenone (BAY94-8862); Placebo |
| Lead sponsor | Bayer |
2. Clinical Question
The central statistical question was whether randomized assignment to finerenone, compared with placebo, changed the occurrence of the registered primary cardiovascular composite endpoint in subjects with type 2 diabetes mellitus and the clinical diagnosis of diabetic kidney disease.
Population
Subjects with type 2 diabetes mellitus and the clinical diagnosis of diabetic kidney disease.
Intervention
Finerenone (BAY94-8862).
Comparator
Placebo.
Primary question
Does finerenone produce a difference in the time to first occurrence of the registered cardiovascular composite endpoint compared with placebo?
3. Trial Design
Finerenone
- Finerenone (BAY94-8862)
- 7352 total subjects were enrolled across the two arms.
Placebo
- Placebo
- 7352 total subjects were enrolled across the two arms.
4. Endpoints
The registry lists one primary endpoint and five additional posted statistical analyses. Four of the six posted analyses are time-to-event endpoints analyzed with a log-rank framework, while the urinary albumin-to-creatine ratio analysis is continuous and was analyzed with ANCOVA.
| Endpoint | Time frame | Type | Analysis |
|---|---|---|---|
| Primary: The First Occurrence of the Composite Endpoint of Cardiovascular Death, Non-fatal Myocardial Infarction, Non Fatal Stroke, or Hospitalization for Heart Failure. | From randomization up until the first occurrence of the CV composite endpoint, or censoring at the end of the study, with an average study duration of 41 months. | Time-to-event | Stratified log-rank test; stratified Cox proportional hazards regression for HR |
| The First Occurrence of the Composite Endpoint of Onset of Kidney Failure, a Sustained Decrease of eGFR ≥40% From Baseline Over at Least 4 Weeks, or Renal Death. | From randomization up until the first occurrence of the renal composite endpoint, or censoring at the end of the study, with an average study duration of 41 months | Time-to-event | Stratified log-rank test; stratified Cox proportional hazards regression for HR |
| All-cause Hospitalization | From randomization up until the first occurrence of the hospitalization due to any cause, or censoring at the end of study, with an average study duration of 41 months | Time-to-event | Stratified log-rank test; stratified Cox proportional hazards regression for HR |
| All-cause Mortality | From randomization up until death due to any cause, or censoring at the end of the study, with an average study duration of 41 months | Time-to-event | Stratified log-rank test; stratified Cox proportional hazards regression for HR |
| Change in Urinary Albumin-to-creatine Ratio (UCAR) From Baseline to Month 4 | From baseline up until Month 4 | Continuous | ANCOVA |
| The First Occurrence of the Composite Endpoint of Onset of Kidney Failure, a Sustained Decrease in eGFR of ≥57% From Baseline Over at Least 4 Weeks, or Renal Death | From randomization up until the first occurrence of the renal composite endpoint, or censoring at the end of the study, with an average study duration of 41 months | Time-to-event | Stratified log-rank test; stratified Cox proportional hazards regression for HR |
5. Statistical Methodology
Time-to-event analysis
The primary endpoint and four of the secondary analyses are time-to-event outcomes. The registry reports log-rank testing for the randomized comparison and states that a stratified Cox proportional hazards regression model was used to provide the hazard-ratio point estimate and corresponding two-sided 95% confidence interval.
This pairing is statistically coherent: the log-rank test addresses the evidence for a difference between the time-to-event distributions, while the Cox model supplies an interpretable relative effect estimate through the hazard ratio.
A hazard ratio is a relative time-to-event measure. It is not an absolute risk difference, a risk ratio, or a statement that a particular proportion of individual subjects benefited.
Stratified analysis
The registry specifically describes the primary analysis as a stratified log-rank test and states that a stratified Cox proportional hazards regression model was used for the hazard-ratio estimate and confidence interval. The ClinicalTrials.gov record does not identify the specific stratification variables, so none are added here.
Intention-to-treat framework
The primary analysis notes identify intention-to-treat analysis as a concept associated with the primary endpoint. The secondary all-cause hospitalization, all-cause mortality, and urinary albumin-to-creatine ratio analyses also identify intention-to-treat analysis in their analysis text where specified.
The important statistical principle is that randomization creates the treatment comparison. An intention-to-treat analysis preserves that assignment when evaluating efficacy, rather than redefining the treatment groups according to treatment received after randomization.
ANCOVA for the Month 4 ratio endpoint
The change in urinary albumin-to-creatine ratio from baseline to Month 4 was a continuous endpoint and was analyzed using ANCOVA. The registry reports the effect measure as a ratio of least squares means, with a two-sided 95% confidence interval.
Because the reported effect is a ratio, an estimate below 1 indicates a lower modeled Month 4 ratio in the finerenone group relative to placebo under the specified ANCOVA analysis. The estimate should not be relabeled as a raw mean difference.
6. Results
ClinicalTrials.gov reports six statistical analyses: one primary endpoint analysis and five secondary endpoint analyses. The primary endpoint and all five secondary analyses reported in the ClinicalTrials.gov record have formal statistical estimates.
Primary Endpoint: Cardiovascular Composite
The primary endpoint was the first occurrence of the composite of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure, measured from randomization until the first occurrence of the CV composite endpoint or censoring at the end of the study.
Hazard ratio for the primary cardiovascular composite
95% CI: 0.76–0.98 · P = 0.0264
Finerenone vs placebo · Full analysis set · Stratified log-rank test
| Element | Reported result |
|---|---|
| Analysis population | Full analysis set (FAS) |
| Groups compared | Finerenone vs Placebo |
| Method | Log Rank |
| Effect measure | Hazard Ratio (HR) |
| Estimate | 0.87 |
| 95% CI | 0.76–0.98 |
| P-value | 0.0264 |
| Hypothesis type | Superiority |
The reported HR of 0.87 means that the estimated instantaneous rate of experiencing the primary composite event was 0.87 times that in the placebo group under the stratified Cox model. Expressed as a relative difference, this corresponds to an estimated 13% lower hazard for finerenone relative to placebo.
The HR does not mean that 13% fewer subjects necessarily experienced the endpoint, nor does it imply that every individual subject had exactly a 13% reduction in risk. Hazard ratios summarize relative event rates over time under the fitted model.
The two-sided 95% CI of 0.76–0.98 describes the statistical uncertainty around the estimated hazard ratio. Its width indicates that the point estimate should not be treated as exact. The interval also lies below 1, which is consistent with the reported superiority-test result.
The P-value of 0.0264 addresses the evidence against the relevant null hypothesis under the statistical testing framework. It does not measure the magnitude or clinical importance of the effect. The HR and its confidence interval provide the effect-size information.
Because this is a Cox-model hazard ratio, interpretation also depends on the appropriateness of the proportional-hazards framework. The ClinicalTrials.gov record does not report a diagnostic or assessment of that assumption, so the single HR should be understood as a model-based summary rather than a complete description of the event curves.
Secondary Endpoint: 40% Renal Composite
Hazard ratio for the 40% renal composite
95% CI: 0.76–1.01 · P = 0.0689
Finerenone vs placebo · Stratified log-rank test
The endpoint was the first occurrence of onset of kidney failure, a sustained decrease of eGFR ≥40% from baseline over at least 4 weeks, or renal death.
The estimated HR of 0.87 corresponds to a lower estimated hazard in the finerenone group under the reported model. The 95% CI of 0.76–1.01 spans 1, so the interval includes the null value for a hazard ratio.
The reported P-value of 0.0689 is a measure of compatibility with the null hypothesis under the stated superiority-testing framework; it is not a measure of the size of the renal effect. The HR itself provides the effect estimate, while the confidence interval communicates its precision.
The registry notes an endpoint-testing sequence: if the treatment effect of the 40% renal composite endpoint was not significant, all other endpoints—namely all-cause hospitalization, all-cause mortality, change in UACR from baseline to Month 4, and the 57% renal composite endpoint—would be tested in an exploratory manner. This hierarchy is important when interpreting the secondary P-values rather than treating each one as an independent confirmatory test.
Secondary Endpoint: All-cause Hospitalization
Hazard ratio for all-cause hospitalization
95% CI: 0.90–1.04 · P = 0.3558
Finerenone vs placebo · Full analysis set · Stratified log-rank test
The HR of 0.97 is close to 1, indicating a small estimated relative difference in the instantaneous rate of first all-cause hospitalization between the randomized groups under the reported model.
The 95% CI of 0.90–1.04 includes 1 and provides a range of plausible hazard-ratio values around the point estimate. The P-value of 0.3558 does not measure the size of the effect; it addresses the statistical evidence under the stated hypothesis-testing framework.
The registry specifically identifies this analysis as exploratory if the preceding 40% renal composite treatment effect was not significant. Consequently, the nominal P-value should not be interpreted in isolation from the endpoint hierarchy described in the registry.
Secondary Endpoint: All-cause Mortality
Hazard ratio for all-cause mortality
95% CI: 0.77–1.04 · P = 0.1337
Finerenone vs placebo · Full analysis set · Stratified log-rank test
The HR of 0.89 indicates a lower estimated instantaneous rate of death from any cause in the finerenone group under the reported Cox model. It does not represent an 11% absolute reduction in mortality or an 11% reduction for every subject.
The 95% CI of 0.77–1.04 crosses 1, showing uncertainty that includes both a value below and a value above the conventional null hazard ratio. The P-value of 0.1337 quantifies statistical evidence under the stated testing framework, not the magnitude of the estimated association.
The registry identifies all-cause mortality as exploratory if the treatment effect of the 40% renal composite endpoint was not significant. That hierarchy matters when interpreting the nominal result.
Secondary Endpoint: Change in Urinary Albumin-to-creatine Ratio
Ratio of least squares means at Month 4
95% CI: 0.650–0.704 · P < 0.0001
Finerenone vs placebo · ANCOVA
The reported ratio of least squares means of 0.676 means that the modeled Month 4 urinary albumin-to-creatine ratio for finerenone was estimated at 0.676 times the corresponding placebo value under the ANCOVA analysis.
Equivalently, the modeled ratio is approximately 67.6% of the placebo value. This is a relative ratio, not a raw mean difference. It should therefore not be described as a reduction of 0.676 units.
The 95% CI of 0.650–0.704 is relatively narrow around the point estimate and describes uncertainty in the estimated ratio. The P-value of <0.0001 indicates strong statistical evidence under the stated hypothesis-testing framework, but it does not by itself establish the clinical importance of the magnitude.
The analysis population was subjects in the full analysis set with measurements available within the Month 4 time window. This differs conceptually from a time-to-event analysis of all randomized subjects because availability of a Month 4 measurement is part of the stated analysis population.
Secondary Endpoint: 57% Renal Composite
Hazard ratio for the 57% renal composite
95% CI: 0.60–0.99 · P = 0.0406
Finerenone vs placebo · Stratified log-rank test
This endpoint was the first occurrence of onset of kidney failure, a sustained decrease in eGFR of ≥57% from baseline over at least 4 weeks, or renal death.
The HR of 0.77 means that the estimated instantaneous event rate was 0.77 times that in the placebo group under the reported stratified Cox model. Expressed as a relative hazard difference, this corresponds to an estimated 23% lower hazard for finerenone relative to placebo.
The HR does not mean that 23% of subjects avoided the event or that every subject experienced the same relative reduction. It is a model-based relative time-to-event estimate.
The 95% CI of 0.60–0.99 communicates uncertainty around the estimate and remains below 1 at its upper limit. The P-value of 0.0406 provides evidence under the reported testing framework, but it is not an effect-size measure.
The registry identifies this endpoint as exploratory if the 40% renal composite treatment effect was not significant. Therefore, the nominal result should be interpreted in the context of the prespecified endpoint hierarchy rather than as an isolated confirmatory finding.
7. Results Summary
| Endpoint | Effect | 95% CI | P-value | Method |
|---|---|---|---|---|
| Primary CV composite | HR 0.87 | 0.76–0.98 | 0.0264 | Stratified log-rank |
| 40% renal composite | HR 0.87 | 0.76–1.01 | 0.0689 | Stratified log-rank |
| All-cause hospitalization | HR 0.97 | 0.90–1.04 | 0.3558 | Stratified log-rank |
| All-cause mortality | HR 0.89 | 0.77–1.04 | 0.1337 | Stratified log-rank |
| Change in urinary albumin-to-creatine ratio at Month 4 | Ratio of LS means 0.676 | 0.650–0.704 | <0.0001 | ANCOVA |
| 57% renal composite | HR 0.77 | 0.60–0.99 | 0.0406 | Stratified log-rank |
8. Statistical Methods Explained
Why was a log-rank test used for the primary endpoint?
The primary endpoint is a time-to-event outcome: each subject is followed from randomization until the first occurrence of the cardiovascular composite or censoring. The log-rank test is designed to compare event-time distributions between randomized groups while accommodating censoring. That makes it structurally appropriate for an endpoint defined by the time until the first qualifying event.
Why was a Cox model used in addition to the log-rank test?
The log-rank test supplies a hypothesis test, whereas the Cox proportional hazards model supplies an estimated treatment effect in the form of a hazard ratio and its confidence interval. FIGARO-DKD's registry analysis notes explicitly state that a stratified Cox proportional hazards regression model was used to provide the hazard-ratio point estimate and two-sided 95% confidence interval.
What does an HR of 0.87 mean?
An HR of 0.87 means that, under the fitted model, the estimated instantaneous event rate in the finerenone group was 0.87 times the corresponding rate in the placebo group. The simple transformation 1 − 0.87 = 0.13 gives an estimated 13% lower hazard. This is not the same as saying that 13% fewer subjects experienced the endpoint.
Why is the confidence interval more informative than the P-value alone?
The confidence interval gives both an effect estimate and information about its statistical precision. For the primary endpoint, the HR is 0.87 with a 95% CI of 0.76–0.98. A P-value such as 0.0264 does not communicate that range or tell the reader how close alternative effect sizes are to the point estimate.
Why was ANCOVA used for the Month 4 urinary albumin-to-creatine ratio?
The Month 4 urinary albumin-to-creatine ratio endpoint is continuous rather than time-to-event. ANCOVA provides a linear-model framework for comparing treatment groups while accounting for covariate information included in the model. The ClinicalTrials.gov record identifies ANCOVA as the analysis method and report the result as a ratio of least squares means.
What does a ratio of least squares means of 0.676 mean?
A ratio of 0.676 means the modeled value in the finerenone group was estimated to be 0.676 times the corresponding placebo value. It is therefore a relative measure. It should not be confused with a difference of 0.676 units, and its interpretation depends on the scale and model used for the endpoint.
Why does the endpoint hierarchy matter?
The registry analysis notes state that if the treatment effect of the 40% renal composite endpoint was not significant, all-cause hospitalization, all-cause mortality, change in urinary albumin-to-creatine ratio from baseline to Month 4, and the 57% renal composite endpoint would be tested in an exploratory manner. This means the inferential status of the later endpoints cannot be separated from the prespecified sequence.
9. Hazard Ratios, Confidence Intervals, and P-values
The six posted analyses illustrate three different statistical questions that should be kept separate: how large is the estimated effect?, how precisely is it estimated?, and how much statistical evidence is present against the null hypothesis?
Effect size
The hazard ratio or ratio of least squares means describes the estimated relative difference between randomized groups.
Precision
The 95% confidence interval communicates uncertainty around the reported effect estimate.
Statistical evidence
The P-value addresses the null hypothesis under the specified testing framework. It does not measure effect size.
Analysis context
Endpoint hierarchy, analysis population, censoring, model assumptions, and exploratory status all affect interpretation.
For HR 0.87, the corresponding relative hazard difference is 13%. For HR 0.77, it is 23%. These are transformations of the reported estimates, not new trial results.
A hazard ratio below 1 does not automatically imply a clinically important effect. The magnitude, precision, endpoint definition, time frame, analysis population, and multiplicity framework all matter. Conversely, a P-value above a conventional threshold does not prove that treatment effects are exactly zero; the confidence interval provides information about the range of effects compatible with the data and model.
10. Intention-to-Treat and Analysis Populations
The registry explicitly associates intention-to-treat analysis with the primary endpoint and several secondary analyses. The primary endpoint analysis uses the full analysis set (FAS). All-cause hospitalization and all-cause mortality also specify the full analysis set.
| Analysis | Population reported |
|---|---|
| Primary cardiovascular composite | Full analysis set (FAS) |
| 40% renal composite | Not specified in the ClinicalTrials.gov record |
| All-cause hospitalization | Full analysis set |
| All-cause mortality | Full analysis set |
| Month 4 urinary albumin-to-creatine ratio | Subjects in full analysis set with measurements available within the time window of Month 4 |
| 57% renal composite | Full analysis set |
The Month 4 analysis illustrates why analysis-population language matters. A time-to-event analysis can retain subjects through censoring, whereas a continuous Month 4 analysis requires a measurement within the specified time window. The registry therefore defines a more restricted population for that particular analysis.
11. Safety Results
The ClinicalTrials.gov record reports serious adverse events by randomized arm as affected subjects divided by subjects at risk.
| Safety measure | Finerenone | Placebo |
|---|---|---|
| Serious adverse events, affected / at risk | 1158 / 3683 | 1215 / 3658 |
Finerenone
1158 / 3683 subjects were reported as affected by serious adverse events among those at risk.
Placebo
1215 / 3658 subjects were reported as affected by serious adverse events among those at risk.
The ClinicalTrials.gov record does not provide a formal statistical comparison, confidence interval, or P-value for these serious-adverse-event counts. The numbers should therefore be presented descriptively rather than converted into an unreported hypothesis test.
12. Multiplicity and Endpoint Hierarchy
Multiplicity is particularly important in FIGARO-DKD because the registry describes a sequential relationship between the 40% renal composite endpoint and several other secondary endpoints.
| Endpoint | Role in the ClinicalTrials.gov record | Reported effect | Interpretive status |
|---|---|---|---|
| Cardiovascular composite | Primary | HR 0.87; 95% CI 0.76–0.98; P = 0.0264 | Primary superiority analysis |
| 40% renal composite | Secondary | HR 0.87; 95% CI 0.76–1.01; P = 0.0689 | Part of stated testing sequence |
| All-cause hospitalization | Secondary | HR 0.97; 95% CI 0.90–1.04; P = 0.3558 | Exploratory if 40% renal composite not significant |
| All-cause mortality | Secondary | HR 0.89; 95% CI 0.77–1.04; P = 0.1337 | Exploratory if 40% renal composite not significant |
| Month 4 urinary albumin-to-creatine ratio | Secondary | Ratio 0.676; 95% CI 0.650–0.704; P < 0.0001 | Exploratory if 40% renal composite not significant |
| 57% renal composite | Secondary | HR 0.77; 95% CI 0.60–0.99; P = 0.0406 | Exploratory if 40% renal composite not significant |
This is a useful example of why "P < 0.05" is not a sufficient description of a clinical-trial result. Statistical inference depends on the prespecified testing sequence and the role assigned to each endpoint.
13. Missing Data and Censoring
The time-to-event endpoints are explicitly defined from randomization until the first occurrence of the relevant endpoint or censoring at the end of the study. Censoring is therefore part of the endpoint framework rather than an incidental data-cleaning step.
For the continuous Month 4 urinary albumin-to-creatine ratio analysis, the registry specifies an analysis population consisting of subjects in the full analysis set with measurements available within the Month 4 time window. This indicates that measurement availability is directly relevant to that analysis population.
14. Stratified Analysis and Model Assumptions
The registry uses the phrase stratified log-rank test for the time-to-event comparisons and states that a stratified Cox proportional hazards regression model produced the hazard-ratio estimate and two-sided 95% confidence interval.
Stratification can be useful when the analysis needs to account for prespecified strata rather than treating every subject as though they came from one homogeneous risk set. The important point for interpretation is that the reported HR is a model-based estimate from the stratified Cox framework.
The hazard ratio associated with a treatment indicator is represented by exp(β). In the reported FIGARO-DKD analyses, the stratified Cox model was used to estimate the treatment hazard ratio.
The proportional-hazards assumption is relevant because a single Cox HR is most naturally interpreted when the relative hazard is reasonably stable over time. The ClinicalTrials.gov record does not report a formal assessment of that assumption. The reported HRs should therefore be understood as the estimates from the specified model, without adding an unsupported claim about model diagnostics.
15. Primary Endpoint Interpretation in Detail
The primary HR of 0.87 summarizes the estimated relative difference in the instantaneous rate of the first cardiovascular composite event between finerenone and placebo under the stratified Cox model.
It does not provide the absolute probability of experiencing the composite endpoint, the number needed to treat, the duration of benefit for an individual subject, or the proportion of subjects who personally experienced a 13% reduction.
The 95% CI of 0.76–0.98 describes uncertainty around the estimated HR. It does not represent the range of treatment effects experienced by individual subjects.
The P-value of 0.0264 summarizes evidence against the relevant null hypothesis under the stated statistical framework. It is not a probability that the treatment is effective, and it is not a measure of the size of the treatment effect.
The primary endpoint is a composite. A subject qualifies through the first occurrence of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure. The HR therefore summarizes the composite endpoint as defined, rather than each component separately.
16. Comparing the Time-to-Event Results
The reported hazard ratios range from 0.77 to 0.97 across the four time-to-event secondary analyses and the primary cardiovascular composite. These estimates should not be treated as interchangeable because the endpoints have different clinical definitions and different testing roles.
| Endpoint | HR | 95% CI | Statistical reading |
|---|---|---|---|
| Primary cardiovascular composite | 0.87 | 0.76–0.98 | CI below 1; P = 0.0264 |
| 40% renal composite | 0.87 | 0.76–1.01 | CI includes 1; P = 0.0689 |
| All-cause hospitalization | 0.97 | 0.90–1.04 | CI includes 1; P = 0.3558 |
| All-cause mortality | 0.89 | 0.77–1.04 | CI includes 1; P = 0.1337 |
| 57% renal composite | 0.77 | 0.60–0.99 | CI below 1; P = 0.0406 |
The table demonstrates why effect size and statistical evidence should be read together. For example, the 40% renal composite and primary cardiovascular composite have the same point estimate, 0.87, but different confidence intervals and P-values. The difference arises from the information and endpoint-specific data contributing to each analysis.
17. Trial Timeline
Trial start
FIGARO-DKD began according to the registry profile.
Primary completion
The registry profile gives 2021-02-02 as the primary completion date.
Registry status
The trial is listed as completed, with results posted and six statistical analyses reported in the ClinicalTrials.gov record.
18. Limitations
- Registry-derived scope: this page uses only the trial data posted on ClinicalTrials.gov for FIGARO-DKD. Details not contained in that data are intentionally not reconstructed from outside knowledge.
- No arm-specific enrollment: the ClinicalTrials.gov record reports total enrollment of 7352 and two arms but do not report the number randomized to each arm.
- Composite endpoint: the primary endpoint combines cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, and hospitalization for heart failure. A single HR for the composite does not establish that every component has the same effect.
- Proportional-hazards assumption: the Cox model is a model-based approach. The ClinicalTrials.gov record does not report a formal assessment of proportional hazards.
- Multiplicity: the registry explicitly describes an endpoint hierarchy in which several secondary endpoints would become exploratory if the 40% renal composite treatment effect was not significant.
- Month 4 analysis population: the ANCOVA analysis includes subjects in the full analysis set with measurements available within the Month 4 window, which differs from the population description for several time-to-event analyses.
- Safety statistics: serious adverse events are reported as affected/at-risk counts by arm, without a formal statistical comparison.
- Unreported design details: the ClinicalTrials.gov record does not specify a crossover procedure, interim-analysis design, alpha-spending method, non-inferiority margin, or Bayesian analysis.
19. Why This Trial Matters Statistically
FIGARO-DKD is a useful teaching example because the ClinicalTrials.gov record connects several core clinical-trial methods within one study: randomized allocation, quadruple masking, time-to-event endpoints, stratified log-rank testing, stratified Cox regression, hazard ratios, confidence intervals, intention-to-treat analysis, and ANCOVA for a continuous endpoint.
| Concept | How it appears in FIGARO-DKD |
|---|---|
| Randomization | The trial uses randomized allocation in a parallel design. |
| Blinding | The registry specifies quadruple masking. |
| Time-to-event endpoints | The primary cardiovascular composite and several secondary endpoints are analyzed from randomization to event or censoring. |
| Log-rank test | Reported for the primary and four secondary time-to-event analyses. |
| Stratified analysis | The registry specifies stratified log-rank testing and a stratified Cox model. |
| Hazard ratio | Used for the primary cardiovascular composite and time-to-event secondary endpoints. |
| Confidence interval | Two-sided 95% confidence intervals accompany the reported effect estimates. |
| Intention-to-treat | Identified in the analysis text for the primary and several secondary analyses. |
| ANCOVA | Used for the Month 4 urinary albumin-to-creatine ratio endpoint. |
| Multiplicity | The registry specifies an exploratory-testing consequence tied to the 40% renal composite result. |
20. Related Tutorials
Learn more about the methods used in this trial:
21. Related Calculators
22. Sources
- ClinicalTrials.gov: FIGARO-DKD, NCT02545049.
- PubMed: PMID 31665733.
- PubMed: PMID 34449181.
- PubMed: PMID 42234437.
- PubMed: PMID 41827014.
- PubMed: PMID 41679125.
Continue through the Clinical Biostats statistical pathway
Use the related tutorials and calculators to explore the survival-analysis, regression, inference, and clinical-trial methods represented in FIGARO-DKD.
23. Record Summary
FIGARO-DKD provides a compact teaching example of how a randomized clinical trial can combine time-to-event methodology with a continuous endpoint analysis. The primary cardiovascular composite was evaluated using a stratified log-rank test, with a stratified Cox proportional hazards model supplying an HR of 0.87 and a two-sided 95% CI of 0.76–0.98; the reported P-value was 0.0264. Secondary analyses included additional time-to-event endpoints and an ANCOVA analysis of the urinary albumin-to-creatine ratio at Month 4.
The most important statistical lesson is that these results must be interpreted in layers. The effect estimate describes the relative treatment effect, the confidence interval describes statistical uncertainty, and the P-value describes evidence under the specified testing framework. The endpoint hierarchy further determines whether secondary findings should be viewed as confirmatory or exploratory.