Efficacy and Safety of Finerenone in Subjects With Type 2 Diabetes Mellitus and the Clinical Diagnosis of Diabetic Kidney Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Finerenone in Subjects With Type 2 Diabetes Mellitus and the Clinical Diagnosis of Diabetic Kidney Disease (NCT02545049), a randomized phase 3 study. The primary outcome is The First Occurrence of the Composite Endpoint of Cardiovascular Death, Non-fatal Myocardial Infarction, Non Fatal Stroke, or Hospitalization for Heart Failure..

2. Clinical Question

The trial’s stated purpose was to evaluate efficacy and safety in this population. The primary outcome combines cardiovascular death with non-fatal events and hospitalization for heart failure. Because the endpoint is a composite, its result concerns the combined first-event outcome; it does not, by itself, establish the pattern for each component considered separately. The registry record also lists kidney-related outcomes, all-cause hospitalization, all-cause mortality, and a urinary albumin-to-creatinine measure among secondary outcomes.

The main comparison is between randomized groups over follow-up, rather than a comparison of treatment taken outside the trial. The registry describes an average study duration of 41 months for the primary endpoint. The clinical question is therefore about the recorded first occurrence of the composite during the trial’s follow-up and the registry’s analysis of time to that occurrence.

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group, multicenter, event-driven Phase 3 study sponsored by Bayer. The registry lists random allocation, quadruple masking, and a parallel-group model. The masking designation is recorded as quadruple; the specific roles covered by that designation are not reported in the registry record.

The study compared finerenone with matching placebo, each in addition to standard of care. Enrollment is listed as 7352. The record gives a start date of 2015-09-17, primary completion on 2021-02-02, and a completed status. For the primary outcome, the posted analysis population is the full analysis set. The registry describes follow-up for the primary endpoint as extending from randomization to the first occurrence of the cardiovascular composite or censoring at the end of the study, with an average study duration of 41 months.

The record calls the study event-driven but does not report the original event target, the planned sample-size justification, or the design assumptions used to set enrollment. It also does not report the details of randomization procedures, the identities of the masked parties, or how standard of care was specified. Those details should not be inferred from the design labels alone.

Six outcome measures and six analyses are indicated in the registry record, with five secondary outcomes listed. The primary endpoint is a first-event composite; the other posted outcomes include renal composites, all-cause hospitalization and mortality, and change in urinary albumin-to-creatinine ratio. The registry’s stated average follow-up is not a guarantee that every participant had the same observation time.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Overall, 19381 participants were screened. Of them,11944 participants were screening failures and 7437 participants were randomized. 85 participants were prospectively excluded from the analyses due to critical GCP violations resulting in 7352 in full analysis set. 7341 participants received study treatment. 1 participant was assigned to placebo but received finerenone.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS)

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: Full analysis set (FAS): the FAS comprised all randomized participants (except those with critical GCP violations).

6. Primary Endpoint

The registry lists the primary outcome as The First Occurrence of the Composite Endpoint of Cardiovascular Death, Non-fatal Myocardial Infarction, Non Fatal Stroke, or Hospitalization for Heart Failure., measured over 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..

The primary endpoint is the first occurrence of a composite of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure. The registry reports the number of participants with a first occurrence and describes follow-up from randomization until the first event or censoring at the end of the study. The full analysis set is identified as the population for this outcome.

A composite endpoint combines clinically distinct events into one outcome. This can capture a broader range of events relevant to the trial’s question and can support an analysis of time to the first qualifying event. The result is about the composite as defined: it does not reveal, on its own, which component contributed most, whether component-specific effects were similar, or whether later events after the first event differed between groups. Component-specific primary results are not reported in the registry record.

Time-to-event methods can account for differing follow-up and censoring, unlike a simple comparison of the number of participants with an event divided by the number assigned to a group. The registry’s primary analysis uses a log-rank method and reports a hazard ratio. The posted event counts are useful descriptive information, but they do not replace that time-to-event comparison. The average study duration is reported as 41 months; individual follow-up lengths are not reported in the registry record.

7. Results Posted in the Registry

• Finerenone: 12.4% (458/3686) versus 14.2% (519/3666) on control; risk difference -1.7 percentage points (95% CI -3.3 to -0.2), risk ratio 0.88 (0.78–0.99); unadjusted two-proportion z-test p = 0.029.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

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.

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Finerenone: 9.5% (350/3686) versus 10.8% (395/3666) on control; risk difference -1.3 percentage points (95% CI -2.7 to 0.1), risk ratio 0.88 (0.77–1.01); unadjusted two-proportion z-test p = 0.069.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Finerenone and the control. The interval gives the range of effects that remain compatible with the data.

All-cause Hospitalization

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Finerenone: 42.7% (1573/3686) versus 43.8% (1605/3666) on control; risk difference -1.1 percentage points (95% CI -3.4 to 1.2), risk ratio 0.97 (0.92–1.03); unadjusted two-proportion z-test p = 0.339.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Finerenone and the control. The interval gives the range of effects that remain compatible with the data.

All-cause Mortality

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Finerenone: 9.0% (333/3686) versus 10.1% (370/3666) on control; risk difference -1.1 percentage points (95% CI -2.4 to 0.3), risk ratio 0.90 (0.78–1.03); unadjusted two-proportion z-test p = 0.123.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Finerenone and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover kidney events, hospitalization, mortality, and urinary albumin-to-creatinine change. The posted time-to-event results for the renal composite, all-cause hospitalization, and all-cause mortality have confidence intervals that include the hazard-ratio null value. The registry record lists the urinary albumin-to-creatinine change outcome and another renal composite with a sustained eGFR decrease of at least 57%, but their detailed results are not included in the posted secondary results used for this interpretation.

Overall, the available secondary analyses do not provide clear statistical evidence of a difference for those endpoints based on their reported intervals and p-values. This does not establish that the groups are identical, nor does it negate the primary result. Each endpoint has its own definition, event frequency, and uncertainty. A primary composite result does not imply that every secondary outcome, or every component within a composite, follows the same pattern.

Secondary outcomes are often considered alongside the primary endpoint, but their evidential weight depends on the planned testing strategy. If many outcomes are tested, the chance of at least one apparently notable result can increase unless the analysis accounts for multiplicity. The registry record does not report whether a multiplicity adjustment was used, which outcomes were formally confirmatory, or whether testing followed a prespecified hierarchy. Accordingly, the secondary findings should be interpreted as supportive or exploratory descriptions rather than independent proof of a broad effect.

The overall pattern should therefore be read with the primary cardiovascular result in view but not collapsed into it. The available secondary estimates are less definitive, and unreported detailed results cannot be inferred from outcome titles alone. Their interpretation also depends on the full endpoint definitions and analysis methods, some of which are not described in the registry record.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• 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.
• All-cause Hospitalization
• All-cause Mortality
• Change in Urinary Albumin-to-creatine Ratio (UCAR) From Baseline to Month 4
• 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

9. Statistical Interpretation

10. Statistical Methods Explained

What does the hazard ratio describe?

The registry reports the primary analysis as a log-rank comparison and gives a hazard ratio of 0.87. A hazard ratio compares the event rate over time between groups, conditional on the time-to-event framework. It is not the same quantity as the difference in the proportion of participants who had an event by the end of follow-up. Interpreting it relies on the methods’ treatment of event times and censoring. The registry record does not provide all details needed to assess those assumptions.

The corresponding 95% confidence interval is 0.76 to 0.98. A confidence interval expresses uncertainty around an estimated effect: values within the interval are more compatible with the observed data under the method than values far outside it, while the interval does not remove uncertainty. For a hazard ratio, the value representing no relative difference is the ratio’s null value. The primary interval does not include that value. The reported p-value is 0.026.

How is a risk difference different?

The event counts can also be expressed as observed proportions in each group. An unadjusted re-calculation from the posted counts gives a risk difference of -0.017, with a 95% confidence interval from -0.033 to -0.002. Risk difference is the first group’s observed event proportion minus the comparator’s observed event proportion in this calculation. A negative estimate means the observed proportion was lower in the finerenone group. This calculation does not incorporate event timing or censoring, so it should not be read as a substitute for the reported log-rank hazard ratio.

The p-value shown with this unadjusted calculation is 0.029. It is a test result for the calculation’s statistical comparison, not the probability that the treatment works or that the result occurred by chance. The registry’s sponsor analysis and the count-based re-calculation use different summaries; their numerical estimates and p-values should not be expected to match exactly.

Why do time-to-event methods matter?

In a time-to-event outcome, participants may have different lengths of observation, and some may be censored before an event is recorded. A log-rank analysis uses information about when events occur and the follow-up available up to events or censoring. A simple event-count comparison treats the posted counts as proportions and does not model timing. The registry describes censoring at the end of the study and an average study duration of 41 months, but does not report individual follow-up distributions or further censoring details.

Time-to-event estimates depend on the analysis population and the handling of incomplete follow-up. The primary endpoint population is listed as the full analysis set. The precise analysis rules, treatment of missing data, and details of censoring are not reported in the registry record. They should not be reconstructed from the summary results.

What does a confidence interval add?

A point estimate alone gives no sense of how uncertain the estimate is. The confidence interval shows a range of estimates compatible with the observed data and chosen method at the stated confidence level. Narrower intervals indicate greater precision than wider intervals, but precision is not the same as lack of bias or clinical importance. For the primary hazard ratio, the reported 95% interval is 0.76 to 0.98; for the count-based risk difference, the re-calculated 95% interval is -0.033 to -0.002. The two intervals refer to different effect measures.

For the secondary analyses, the renal composite hazard ratio interval is 0.76 to 1.01, the hospitalization interval is 0.90 to 1.04, and the mortality interval is 0.77 to 1.04. Those ranges include the respective null value for a hazard ratio. A result that is not clearly separated from the null does not demonstrate that groups are identical; it indicates that the data and analysis leave uncertainty about the direction or size of a difference.

What is the role of the count-based re-calculation?

The re-calculated risk differences are simple, unadjusted analyses of the posted summary counts. They provide a transparent descriptive comparison, but they may not reflect the trial’s full analysis plan, any stratification, or time-to-event structure. The registry’s own reported analysis is the log-rank hazard ratio. The discrepancy indicator is false, but the measures still differ in meaning because one is time-to-event and the other is a count-based proportion comparison. Neither p-value should be detached from its effect estimate, interval, endpoint definition, and analysis population.

Where results are tested across multiple endpoints, interpreting each p-value in isolation can overstate the strength of evidence. Whether and how multiplicity was handled is not reported in the registry record. Similarly, the record does not state whether interim analyses were conducted or whether any adjustment was planned for them. These missing methods details limit what can be concluded about the complete testing strategy.

11. Confidence Intervals and Precision

The primary sponsor-reported estimate is a hazard ratio of 0.87, with a 95% confidence interval from 0.76 to 0.98. This interval is on the relative time-to-event scale and does not include the null value for a hazard ratio. Its width conveys uncertainty around the estimate; it does not mean that every value in the interval is equally likely, nor does it account for possible bias or unreported analytical choices.

The unadjusted count-based risk difference is -0.017, with a 95% confidence interval from -0.033 to -0.002. This is a different measure on a proportion-difference scale. The limits are both below no difference, but the estimate should be interpreted as an unadjusted comparison of posted counts rather than an alternative time-to-event analysis. The two sets of intervals cannot be compared as though they were on the same scale.

For the secondary time-to-event analyses, the renal composite hazard ratio is 0.87 (95% confidence interval, 0.76 to 1.01), all-cause hospitalization is 0.97 (0.90 to 1.04), and all-cause mortality is 0.89 (0.77 to 1.04). Each interval includes the hazard-ratio null value, leaving uncertainty about the size and direction of the corresponding difference. Interval inclusion does not prove equivalence or absence of an effect.

Precision depends partly on the amount and timing of outcome information, and also on the analysis method. The registry gives an average study duration of 41 months but does not report the individual follow-up distribution or detailed missing-data handling. Confidence intervals describe sampling uncertainty under the analysis assumptions; they do not capture every limitation in the record.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size displays are planning illustrations based on the observed effect, not the trial’s original design calculations. For time-to-event outcomes, the figures represent total events required; for other outcomes, they represent participants per arm. The primary endpoint is time-to-event, so its figures are total events. The secondary count-based outcomes use participants per arm. This distinction matters: an event target is not a group enrollment target, and participants may contribute follow-up without experiencing an event.

The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The original trial’s own planning assumptions, including its sample-size justification and event target, are not reported in the registry record.

The 80% power column describes an illustrative design that would have an 80% probability of detecting the assumed effect under the stated assumptions if that effect were the true one. The 90% power column uses a higher probability of detection for that same assumed effect and therefore generally calls for more information. The rows for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect show how the planning requirement changes when the assumed effect is reduced. Smaller effects need more participants or events to detect with the same planned power.

For example, the primary display lists 6,032 total events for 80% power and 8,074 total events for 90% power when planning for 100% of the observed effect. These are hypothetical planning figures, not a description of the trial’s observed power. Power calculations do not measure the probability that the observed result is true, and they do not replace the confidence intervals that describe uncertainty around the estimated effects.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.142; treatment proportion p₁ = 0.124; average p̄ = 0.133.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4801 = 0.9410.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4799 = 0.4039.
4. n = (0.9410 + 0.4039)² / (-0.0173)² = 6,031.5, rounded up to 6,032 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.141571194762684                          # control proportion
p1 <- 0.124253933803581                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2545049)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 6,032 (80% power), 8,074 (90% power)
 75% of observed effect: n per arm = 10,871 (80% power), 14,553 (90% power)
 50% of observed effect: n per arm = 24,792 (80% power), 33,188 (90% power)
power.prop.test: 6031.5 
Simulated power at n = 6,032 per arm: 0.801 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry’s own primary analysis uses a log-rank method and reports a hazard ratio of 0.87, a 95% confidence interval from 0.76 to 0.98, and p = 0.026. That is the sponsor-reported time-to-event result. The posted counts separately support an unadjusted re-calculation of the risk difference: -0.017, with a 95% confidence interval from -0.033 to -0.002 and p = 0.029.

These estimates are not identical because they summarize the data differently. The hazard ratio incorporates event timing and censoring through a time-to-event analysis; the re-calculated risk difference compares observed event proportions from the posted counts. The latter is a useful supplementary descriptive analysis, not a replacement for the registry analysis or a reconstruction of the sponsor’s complete statistical model.

The discrepancy indicator is false, and the broad direction of the primary results is consistent across these summaries. Still, the registry’s own analysis remains the relevant reported trial analysis. The record does not provide enough detail to reproduce its full model or assess all implementation choices. Readers should distinguish the sponsor’s log-rank result from simple re-calculations and avoid treating their p-values as interchangeable.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and multiple secondary outcomes, but does not report a multiplicity-control procedure. When researchers test several outcomes, the probability of at least one result appearing notable by chance can rise. Common approaches include adjusting significance thresholds or specifying a hierarchy of outcomes, but whether either approach was used in this trial is not reported in the registry record. It would therefore be inappropriate to assume that every secondary p-value has the same confirmatory status as the primary result.

The record identifies the full analysis set for the primary outcome. Detailed analysis-population definitions for each secondary outcome, including exclusions and handling of treatment discontinuation, are not reported in the registry record. The registry also does not report the missing-data methods, the rules for censoring beyond the endpoint descriptions, or how deviations from assigned treatment were handled. These are important details when assessing how closely the estimates reflect the randomized comparison.

The study is described as event-driven, but the original target number of events, the sample-size justification, and any planned interim analyses are not reported in the registry record. The statistical plan, including any interim monitoring boundaries or adjustments, is also not reported. No assumptions about those procedures should be inferred from the event-driven label or the reported p-values.

Finally, the registry provides summary outcome counts and selected analysis results, but not the full statistical analysis plan or participant-level data. The available record supports interpretation of the stated estimates and intervals, while leaving uncertainty about some operational and analytical decisions. These omissions do not invalidate the reported results, but they limit independent assessment of the complete testing and missing-data strategy.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (3.1%), acute kidney injury (1.3%), cellulitis (1.1%), urinary tract infection (1.1%), type 2 diabetes mellitus (0.9%).

The registry’s safety figures report serious adverse events in 1158 of 3683 participants at risk in the finerenone group and 1215 of 3658 at risk in the placebo group. These are counts, not a causal comparison adjusted for time at risk or participant characteristics. The record also lists other adverse events in 1797 of 3683 and 1717 of 3658, respectively, and deaths in 339 and 383, respectively. The precise definitions and analysis methods for these safety summaries are not reported in the registry record.

The most frequently listed serious adverse-event terms include pneumonia (3.1%), acute kidney injury (1.3%), cellulitis (1.1%), urinary tract infection (1.1%), type 2 diabetes mellitus (0.9%), and hypoglycaemia (0.8%). These percentages are reported terms in the registry; their denominators and whether participants could contribute more than one term are not specified in the information available for interpretation. They should not be summed or treated as mutually exclusive categories.

Safety counts provide descriptive context but do not alone establish that an event was caused by study treatment, nor do they describe severity beyond the serious-event classification. A full safety assessment would require the event definitions, exposure duration, coding and attribution methods, and the broader adverse-event results, which are not reported in the registry record.

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

The primary result describes a difference in the time-to-event analysis of the first cardiovascular composite event between the randomized groups. The reported hazard ratio is 0.87, with a 95% confidence interval from 0.76 to 0.98 and p = 0.026. The posted event counts and unadjusted risk-difference calculation point in the same direction, but they describe observed proportions rather than event timing. The estimates are evidence about the trial’s defined population, endpoint, follow-up, and analysis.

This does not mean that every participant had the same outcome or that the result applies automatically to people outside the trial population. It does not show which component of the composite accounts for the overall pattern; component-specific primary results are not reported in the registry record. It also does not establish a clear difference for each secondary outcome. The available secondary time-to-event intervals include the hazard-ratio null value, and the record does not report multiplicity methods.

A statistically notable result is not the same as a complete measure of clinical importance. The point estimate and confidence interval describe relative event-time differences, while the count-based risk difference gives a separate unadjusted summary. A clinician or researcher considering the broader implications would also need context about absolute risk, harms, population characteristics, follow-up, treatment adherence, and other evidence. The registry summary alone does not supply all of that information.

The result should not be read as certainty, proof of mechanism, or a guarantee of an individual outcome. Confidence intervals reflect sampling uncertainty under the analysis method, and unreported details about missing data, censoring, multiplicity, and analysis procedures limit further appraisal. The registry findings are a statistical account of the posted trial results, not medical advice.

18. Why This Trial Matters Statistically

This trial illustrates why the choice of effect measure matters. A log-rank hazard ratio summarizes event experience over time, whereas an unadjusted risk difference derived from event counts compares observed proportions. Both can be informative, but they are not interchangeable. Their distinction is especially important when follow-up varies or censoring occurs, as in a time-to-event study.

The primary estimate is accompanied by a confidence interval and a p-value, allowing readers to assess both the direction and precision of the reported comparison. The interval provides more information than a binary label of statistically significant or not. At the same time, the secondary results demonstrate why a primary finding should not be generalized across every outcome: the available secondary intervals include the hazard-ratio null value, and details of multiplicity handling are not reported.

The sample-size illustrations also show how planning depends on the effect assumed to be worth detecting. Smaller assumed effects require more information for a specified power under the stated illustrative assumptions. These calculations are distinct from the trial’s original design process, which is not reported in the registry record, and they are not observed power.

Finally, the record offers a practical reminder that a completed randomized study can still have limits in its public statistical detail. The population and endpoint are defined, and the sponsor’s primary estimate is available, but aspects of the analysis plan, missing-data approach, and secondary testing strategy remain unreported. Clear interpretation separates what the results show from what the record leaves uncertain.

19. 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 registry record does not provide component-specific results for cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure. Consequently, the primary composite estimate cannot identify which component contributed most or whether the components had similar patterns. The registry also lists six outcome measures but provides detailed secondary results for only some outcomes in the information summarized for this analysis; results for the other listed measures are not reported in the registry record.

The average study duration is reported as 41 months, but individual follow-up distributions and detailed censoring information are not reported. The record identifies the primary population as the full analysis set, while further definitions of analysis populations and handling of treatment changes are not reported. Missing-data procedures, the full statistical analysis plan, and multiplicity control are also not reported in the registry record.

The study is labeled event-driven, but its original event target, sample-size justification, and any interim analyses are not reported. The planning illustrations based on observed effects should not be mistaken for the trial’s own design calculations. The serious adverse-event figures are descriptive; detailed definitions, exposure-adjusted comparisons, attribution, and the full safety analysis are not reported in the registry record.

These limitations do not erase the reported randomized comparison, but they constrain independent evaluation of its analytical context and the breadth of conclusions. The available evidence is strongest for the posted primary time-to-event estimate and should not be extended to unreported component findings or analysis details.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of Finerenone in Subjects With Type 2 Diabetes Mellitus and the Clinical Diagnosis of Diabetic Kidney Disease — NCT02545049.
• Registry record last updated 2022-04-15; status last verified 2022-04.
• Results first posted 2022-04-15.
• 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.

23. Record Summary

The registry describes a randomized, double-blind, placebo-controlled, parallel-group, multicenter, event-driven Phase 3 trial of finerenone in addition to standard of care versus matching placebo in people with type 2 diabetes and diabetic kidney disease. Enrollment is listed as 7352, and the study status is completed.

The primary endpoint is the first occurrence of cardiovascular death, non-fatal myocardial infarction, non-fatal stroke, or hospitalization for heart failure. The reported log-rank analysis gives a hazard ratio of 0.87 (95% confidence interval, 0.76 to 0.98; p = 0.026). An unadjusted re-calculation from posted event counts gives a risk difference of -0.017 (95% confidence interval, -0.033 to -0.002).

The available secondary time-to-event estimates are less conclusive, and the registry record does not report important details including multiplicity and missing-data procedures. The findings should be understood within the defined population, endpoint, follow-up, and limits of the posted record.