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 A Phase 3b Study of Erythropoietin Drugs Using a Specified Dosing Algorithm in Patients With Chronic Kidney Disease on Hemodialysis (NCT02504294), a randomized phase 3 study. The primary outcome is Percentage of Time When Participants Had Hemoglobin Levels Between 9 to 11 Gram Per Deciliter (g/dL).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02504294 |
| Phase | Phase 3 |
| Condition | Chronic Kidney Disease (CKD) |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 432 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Pfizer |
| Results posted | Yes |
2. Clinical Question
Population
Adults with chronic kidney disease receiving hemodialysis were the population of interest. The primary outcome analysis included participants who received at least one dose of study medication after randomization and were evaluable for that outcome. The registry record describes this group as the full analysis set, while noting that the number analyzed for the outcome refers to evaluable participants.
Intervention
The intervention arm was Epoetin Hospira. The study summary describes a comparison of a dosing algorithm for Hospira Epoetin with standard-of-care Epogen. The registry record lists IV iron as an intervention, but does not report its role or use in enough detail to characterize it further.
Comparator
The comparator was Epogen in the standard-of-care arm. The study was open label, so masking was not used. The registry record does not specify further details of the standard-of-care dosing approach in the information describing the posted outcomes.
Primary question
During Week 17 up to Week 24, what percentage of weeks did participants have hemoglobin levels between 9 and 11 g/dL, and how did that percentage compare between Epoetin Hospira and Epogen? This endpoint describes time within a stated hemoglobin range rather than a single hemoglobin measurement. The posted comparison is categorized as a non-inferiority or equivalence analysis, but the registry record does not report the margin needed to interpret such a claim formally.
3. Trial Design
This was a Phase 3b, randomized, open-label, parallel-group study in adults with chronic kidney disease on hemodialysis. The registry record reports enrollment of 432 participants and a completed study. It lists the study start as 2015-07-13 and primary completion as 2016-07-02. Those record dates describe the study timeline; they do not, by themselves, establish when every participant entered or completed treatment.
Participants were assigned to Epoetin Hospira or standard-of-care Epogen. The study summary frames the comparison as Hospira Epoetin using a specified dosing algorithm versus standard-of-care Epogen. Because masking was listed as none, the study was open label. The record does not provide additional detail about how the algorithm was implemented, how standard-of-care dosing was determined, or whether clinicians could make other treatment decisions during follow-up.
The primary outcome covered Week 17 up to Week 24. The registry reports outcome-evaluable counts of 178 for Epoetin Hospira and 173 for Epogen. These counts are distinct from the adverse-event populations, which were 212 and 206, respectively. They are also distinct from total enrollment: enrollment is not the same as the number contributing data to a particular outcome.
A secondary outcome measured change from baseline in weekly mean study medication dose over the final 8 weeks of treatment. The record also lists IV iron as an intervention, but does not provide enough information to describe its administration or its relationship to the two randomized arms. The allocation concealment process, detailed eligibility criteria, treatment duration beyond the stated endpoint timeframes, and sample-size justification are not reported in the registry record.
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: In this study, 432 participants were enrolled, however, only 418 participants were treated.
Overall Study
| Stage | Epoetin Hospira | Epogen |
|---|---|---|
| STARTED | 212 | 206 |
| COMPLETED | 156 | 158 |
| NOT COMPLETED | 56 | 48 |
Reasons for not completing this period:
| Reason | Epoetin Hospira | Epogen |
|---|---|---|
| Adverse Event | 15 | 12 |
| Withdrawal by Subject | 7 | 1 |
| Protocol Violation | 27 | 26 |
| Physician Decision | 1 | 2 |
| Lost to Follow-up | 6 | 7 |
Analysis population for the primary outcome, as defined in the registry: FAS included all participants who received at least 1 dose of study medication after randomization into the study. Here, "Number of participants analyzed (N)" signifies those number of participants who were evaluable for this outcome measure.
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) included all participants who received at least 1 dose of study medication after randomization into the study.
| Characteristic | Unit | Epoetin Hospira | Epogen | Total |
|---|---|---|---|---|
| Participants analysed | 212 | 206 | 418 | |
| Age, Continuous | Years | 60.5 (SD 13.96) | 59.3 (SD 14.23) | 59.9 (SD 14.09) |
| Sex: Female, Male: Female | Participants | 82 | 102 | 184 |
| Sex: Female, Male: Male | Participants | 130 | 104 | 234 |
6. Primary Endpoint
The registry lists the primary outcome as Percentage of Time When Participants Had Hemoglobin Levels Between 9 to 11 Gram Per Deciliter (g/dL), measured over Week 17 up to Week 24.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of Weeks |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was the percentage of weeks in which participants had hemoglobin levels between 9 and 11 g/dL, measured from Week 17 up to Week 24. It summarizes how often measurements fell within a prespecified range during a span of treatment, rather than focusing on a single measurement or only the final measurement. The unit is percentage of weeks.
As a general statistical concept, a time-in-range endpoint can describe the pattern of repeated measurements across a period. It can therefore reflect both the level of a measure and how consistently it falls within the defined range over that period. It does not, by itself, describe the size or duration of excursions outside the range, nor does it show each participant’s individual pattern. The registry record does not give the measurement schedule or explain how weeks without an available measurement were handled.
The reported arm estimates were 61.93% for Epoetin Hospira and 63.33% for Epogen. Their respective 95% confidence intervals were 57.51 to 66.17 and 58.71 to 67.72. These intervals express uncertainty around each arm’s estimated percentage. The registry record also posts a between-arm analysis categorized as non-inferiority or equivalence, but does not report the method or the decision margin. Without that margin and the relevant analysis details, the arm estimates alone do not establish a formal non-inferiority or equivalence conclusion.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary outcome estimates indicate that the percentage of weeks with hemoglobin between 9 and 11 g/dL was 61.93% in the Epoetin Hospira group and 63.33% in the Epogen group during Week 17 up to Week 24. The confidence intervals for the two arm estimates overlap. The posted between-arm analysis reports a difference whose confidence interval crosses zero. This means the interval includes values consistent with differences in either direction; it does not establish that the treatments are identical.
The registry categorizes the primary comparison as a non-inferiority or equivalence analysis. That label is not enough to determine whether a non-inferiority or equivalence criterion was met. Such an interpretation depends on the prespecified margin and the analysis plan, neither of which is reported in the registry record. A confidence interval that includes zero is not, by itself, evidence of equivalence: a formal conclusion requires judging the full interval against the chosen margin.
Precision is relevant to interpretation. A confidence interval gives a range of estimates compatible with the observed data under the statistical model and its assumptions. It is not a guarantee that the true value lies in that interval, and it does not measure clinical importance. The arm-level intervals provide uncertainty estimates for each percentage; the registry’s between-arm interval is the more direct quantity for judging the comparative estimate. Its span indicates that the posted results leave room for a range of possible between-arm effects.
The outcome counts were 178 for Epoetin Hospira and 173 for Epogen. The record defines the full analysis set as participants who received at least one dose after randomization, but clarifies that the count analyzed for this endpoint is the number evaluable for it. Thus, the endpoint estimates do not represent every enrolled participant. The registry does not explain why some participants were not evaluable or how missing measurements affected the analysis.
The primary result concerns a hemoglobin time-in-range measure over the stated weeks. It does not, on its own, establish effects on other clinical outcomes, describe individual responses, or determine the balance of benefits and harms. Nor does it settle how the dosing algorithm should be interpreted in settings beyond the trial population. The data should be read as an estimate with uncertainty, alongside the limitations of the registry’s reporting.
9. Statistical Methods Explained
The study used randomized, parallel-group allocation, with an open-label design. Randomization is intended, in general, to make treatment groups comparable on average at the point of assignment. It does not ensure that every characteristic is balanced in a particular trial. The registry record does not report the randomization sequence method, allocation concealment, or whether any stratification factors were used. Open-label means masking was not used; the record does not describe whether that affected treatment delivery, measurement, or analysis.
What the primary estimate represents
The primary outcome is a percentage of weeks within the specified hemoglobin range. Each arm’s reported value is an estimate of that percentage among participants evaluable for the outcome. The reported 95% confidence intervals describe statistical uncertainty around those arm estimates under the analysis used. In general, a confidence interval is calculated by a method that depends on the data structure and assumptions. The registry record does not report the primary outcome’s statistical method, the handling of repeated observations, or the assumptions used to calculate its intervals.
The posted primary comparison is labeled as a difference in percentage and classified as non-inferiority or equivalence. Its confidence interval includes zero. In general, a confidence interval for a difference that includes zero does not rule out no difference, but it also does not establish that the difference is sufficiently small to meet an equivalence or non-inferiority standard. That requires a margin set before interpreting the results and a suitable analysis. The margin and the detailed statistical plan are not reported in the registry record.
How to read confidence intervals
As a general explanation, a confidence interval reflects the uncertainty associated with estimating a population quantity from trial data, given a statistical procedure and its assumptions. A narrower interval indicates greater precision than a wider interval on the same scale. An interval does not assign a probability to the fixed, unknown true value after the data have been observed. Its interpretation also does not replace consideration of the magnitude of the estimate or its relevance to the study question.
For a comparative analysis, the interval around the difference is especially useful because it directly describes uncertainty about the contrast between groups. A zero difference is the reference point for no measured between-group difference on that scale. For a non-inferiority question, however, zero is not the only relevant boundary: the prespecified margin is central. The record does not provide that boundary, so the classification of the analysis cannot be translated into a verified conclusion about meeting a margin.
Secondary outcome and recalculation
The secondary outcome is change from baseline in weekly mean study medication dose over the final 8 weeks of treatment. The reported analysis uses ANCOVA and presents an adjusted least-squares mean difference, with a 95% confidence interval and a p-value. ANCOVA is a general regression-based method that can compare groups while accounting for covariates included in the model. The registry identifies the method but does not list the covariates, model specification, or assumptions.
A separate recalculation based on posted summary statistics is an unadjusted approximation, not the sponsor’s adjusted ANCOVA analysis. The recalculated comparison gives a mean difference of 1062.0 units per week, with a 95% confidence interval from -513.3 to 2637.0 and p = 0.186. This interval crosses zero. Such a recalculation can help illustrate the information in the group summaries, but it is not interchangeable with a model-based result that adjusts for covariates. The standard deviations used in the approximation are identified as approximate in the posted analysis, which is another reason to treat the recalculation as an illustration rather than a replacement for the sponsor’s analysis.
Planning illustrations
Sample-size planning illustrations use the observed secondary-outcome effect as a reference and consider planning for 100%, 75%, and 50% of that observed effect. For those scenarios, the stated participants per arm for 80% power are 796, 1,415, and 3,183, respectively; the participants per arm for 90% power are 1,065, 1,894, and 4,260, respectively. These are planning illustrations based on a two-sided significance level of 0.05, equal allocation, and a normal approximation. They are not observed power and are not the trial’s own sample-size justification. The registry record does not report the trial’s sample-size assumptions or justification.
In general, power is the probability that a specified statistical test will reject its null hypothesis under a particular assumed effect and model. A planning calculation asks what sample size would be needed under stated assumptions; it does not describe the probability that the completed trial’s observed result is true or reliable. Illustrations tied to an observed effect can be sensitive to the estimate used, especially when that estimate is uncertain. Planning for a smaller effect requires more participants under the stated assumptions. These calculations should therefore be read as hypothetical design scenarios, not as an assessment of trial success.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| not stated | Difference in Percentage -1.3975 | -7.6503 to 4.8553 | not reported |
The registry posts a primary comparison categorized as non-inferiority or equivalence, with a reported difference and 95% confidence interval. The interval crosses zero, but the registry record does not report the margin, statistical method, or decision rule needed to determine whether a formal non-inferiority or equivalence criterion was met. The registry classification should therefore be distinguished from a conclusion that can be independently checked using the information in the record.
For the secondary dose-change outcome, the sponsor’s reported analysis is an ANCOVA result. It gives an adjusted least-squares mean difference of -1062.0 units per week, a 95% confidence interval from -2643.2 to 519.2, and p = 0.187. The recalculation is a simple unadjusted approximation from posted summary statistics; it gives a mean difference of 1062.0 units per week, a 95% confidence interval from -513.3 to 2637.0, and p = 0.186. The comparison is not an exact replication of ANCOVA: it does not use the sponsor model’s covariate adjustment, and its variability estimate is approximate.
The sponsor’s own analysis takes precedence when interpreting the reported secondary result. The difference between the posted adjusted estimate and the unadjusted recalculation illustrates why the methods should not be treated as interchangeable. Both intervals include zero, but the recalculation does not supersede the sponsor’s analysis or establish a result beyond what the registry reports.
11. Multiplicity, Missing Data and Analysis Populations
The registry record defines the primary analysis population as participants who received at least one dose of study medication after randomization. For the primary outcome, it clarifies that the number analyzed is the number evaluable for that outcome. The posted outcome counts are 178 for Epoetin Hospira and 173 for Epogen. These differ from the adverse-event populations of 212 and 206 and from total enrollment of 432. The record does not explain the reasons for these differences.
Missing-data methods are not reported in the registry record. In particular, it does not describe how missing hemoglobin measurements or incomplete follow-up were handled, whether any imputation was used, or whether sensitivity analyses assessed alternative assumptions about missing data. As a general matter, missing observations can affect an estimate when the likelihood or pattern of missingness is related to outcomes or treatment. Without the analysis details, the direction or extent of any such effect cannot be determined.
The record also does not report interim analyses, a statistical analysis plan, a sample-size justification, or a procedure for controlling multiplicity across outcomes or analyses. The presence of one posted secondary outcome does not establish how many analyses were planned or how results were prioritized. As a general statistical concern, multiple testing can raise the probability of chance findings unless the analysis plan accounts for it. The registry does not provide enough information to assess whether such an adjustment or hierarchy was used.
For a non-inferiority or equivalence interpretation, the missing details are especially important: the margin, analysis population requirements, and decision rule are not reported. The record’s analysis label alone does not establish that those criteria were met. These reporting gaps limit independent assessment of the primary comparison and prevent a full appraisal of the trial’s planned inferential framework.
12. 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 |
|---|---|---|---|
| Epoetin Hospira | 66 | 212 | 31.1% |
| Epogen | 64 | 206 | 31.1% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Epoetin Hospira | 10 of 212 |
| Epogen | 21 of 206 |
Most frequent serious adverse event terms
| Event term | Organ system | Epoetin Hospira | Epogen |
|---|---|---|---|
| Cardiac arrest | Cardiac disorders | 2 of 212 | 6 of 206 |
| Pneumonia | Infections and infestations | 7 of 212 | 1 of 206 |
| Condition aggravated | General disorders | 4 of 212 | 3 of 206 |
| Fluid overload | Metabolism and nutrition disorders | 4 of 212 | 3 of 206 |
| Hyperkalaemia | Metabolism and nutrition disorders | 5 of 212 | 1 of 206 |
| Acute myocardial infarction | Cardiac disorders | 2 of 212 | 2 of 206 |
| Chest pain | General disorders | 3 of 212 | 1 of 206 |
| Sepsis | Infections and infestations | 3 of 212 | 1 of 206 |
| Urinary tract infection | Infections and infestations | 2 of 212 | 2 of 206 |
| Acute respiratory failure | Respiratory, thoracic and mediastinal disorders | 2 of 212 | 2 of 206 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Epoetin Hospira | Epogen |
|---|---|---|---|
| Dyspnoea | Respiratory, thoracic and mediastinal disorders | 7 of 212 | 11 of 206 |
| Nausea | Gastrointestinal disorders | 3 of 212 | 13 of 206 |
Serious events with the highest arm-level rate: pneumonia (3.3%), cardiac arrest (2.9%), hyperkalaemia (2.4%), condition aggravated (1.9%), fluid overload (1.9%).
The posted serious adverse event figures show 66 affected participants among 212 at risk in the Epoetin Hospira group and 64 among 206 at risk in the Epogen group. For other adverse events, the corresponding counts are 10 among 212 and 21 among 206. These are counts of participants affected, not counts of all event episodes. The registry record does not provide a statistical comparison for these figures, so they should not be interpreted as evidence of a tested between-group difference.
The record lists serious adverse event terms including Pneumonia (3.3%), Cardiac arrest (2.9%), Hyperkalaemia (2.4%), Condition aggravated (1.9%), Fluid overload (1.9%), and Cardiac failure congestive (1.5%). It does not make clear in the posted information how these term-level percentages map to the individual treatment arms, so they should not be assigned to one arm. Death counts are not reported in the registry record.
As a general statistical point, adverse-event counts depend on the population at risk, duration and completeness of observation, and the way events are defined and collected. The open-label design and the absence of detail about event ascertainment also limit what can be inferred from the posted figures. These data describe reported events; they do not, on their own, establish causation or provide a complete comparison of safety.
13. What the Result Does, and Does Not, Mean
The primary result estimates how often participants’ hemoglobin levels fell within the stated range during the specified part of treatment. The arm estimates are close in numerical value, and the posted between-arm confidence interval includes zero. These observations describe the trial’s estimates and their uncertainty; they do not show that the groups were identical. A range of between-group effects remains compatible with the posted interval.
The registry’s non-inferiority or equivalence classification cannot be interpreted as a confirmed formal conclusion without the margin and decision rule. A nonsignificant difference, or an interval that includes zero, is not equivalent to evidence of equivalence. Conversely, the interval does not prove a difference. Interpretation depends on the study question, the prespecified threshold, the analysis population, and the statistical method, details that are not fully reported.
The endpoint is a measure of hemoglobin time within a defined range, not a complete account of health outcomes, individual experience, or overall treatment effects. The secondary outcome concerns study medication dose and is not a substitute for the primary endpoint. The serious adverse-event figures provide descriptive counts, but the registry does not report a formal statistical comparison for them.
The results apply to the adults with chronic kidney disease on hemodialysis studied in this randomized trial, subject to the limitations of the reported populations and analyses. The record does not support conclusions about groups not described in the trial or about outcomes that were not posted. No treatment recommendation follows from these statistical results alone. Their meaning is bounded by the endpoint definitions, uncertainty, missing analysis details, and the registry’s reporting.
14. Why This Trial Matters Statistically
This trial raises a useful statistical distinction between comparing observed group estimates and establishing non-inferiority or equivalence. Similar-looking arm estimates do not, by themselves, demonstrate that two treatments are equivalent. Formal non-inferiority and equivalence questions require a margin and an analysis designed to evaluate that margin. Because the registry record labels the primary analysis as non-inferiority or equivalence but does not report the margin or decision rule, readers can see the estimates and uncertainty without being able to verify the formal claim.
The outcome also illustrates the value of defining what a percentage represents. The primary measure is the percentage of weeks within a hemoglobin range over a particular treatment period. It captures a pattern across time rather than a single endpoint measurement. At the same time, the record does not describe measurement frequency or the handling of missing weeks, details that matter for understanding how the measure was constructed.
The trial also provides a practical example of why adjusted sponsor analyses and recalculations from summary statistics should be distinguished. The secondary ANCOVA result accounts for the model used by the sponsor, while the recalculation is unadjusted and uses approximate variability. The latter can be a useful check on the scale and uncertainty of a comparison, but it is not a replacement for the reported adjusted analysis.
Finally, the record highlights how analysis populations shape interpretation. Enrollment, outcome-evaluable counts, and adverse-event populations are not interchangeable. Clear reporting of who was included, how missing data were handled, and how multiple outcomes were interpreted is essential to assess how robustly a trial answers its stated question.
15. 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 leaves several trial-specific details unavailable. It does not report the non-inferiority or equivalence margin, the primary analysis method, or the statistical decision rule. Without these, the posted classification of the primary analysis cannot be independently checked against a formal criterion. The record also does not state the sample-size justification, so it is not possible to assess the planned precision or power for the primary question from the registry information.
The primary endpoint counts are smaller than enrollment, and the record defines the analyzed number as participants evaluable for the outcome. Reasons for non-evaluability and methods for handling missing measurements are not reported. The adverse-event populations differ from the primary outcome populations, but the record does not explain those differences. These population distinctions limit direct comparisons across efficacy and safety summaries.
The open-label design means participants and study personnel were not masked. The registry does not report how outcome assessment was conducted, or whether measurement and analysis procedures could mitigate potential influence of knowing treatment assignment. The record also does not provide details of the dosing algorithm, standard-of-care dosing decisions, or the role of IV iron.
Only one secondary outcome is posted, and the record does not describe multiplicity control or a prespecified hierarchy. The serious-event term percentages are not clearly assigned to arms in the posted information, and deaths are not reported in the registry record. These limitations do not negate the reported estimates, but they constrain how fully the study’s comparative results and safety profile can be assessed.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: A Phase 3b Study of Erythropoietin Drugs Using a Specified Dosing Algorithm in Patients With Chronic Kidney Disease on Hemodialysis — NCT02504294.
- Registry record last updated 2018-06-13; status last verified 2018-05.
- Results first posted 2018-06-13.
- 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.
19. Record Summary
NCT02504294 is a completed Phase 3b randomized, open-label, parallel-group study of Epoetin Hospira versus standard-of-care Epogen in adults with chronic kidney disease receiving hemodialysis. The registry reports enrollment of 432. The primary outcome was the percentage of weeks with hemoglobin between 9 and 11 g/dL from Week 17 up to Week 24. The reported arm estimates were 61.93% and 63.33%, with 95% confidence intervals of 57.51 to 66.17 and 58.71 to 67.72, respectively.
The primary comparison is categorized as non-inferiority or equivalence, but its margin and detailed method are not reported in the registry record. One secondary outcome, change in weekly mean study medication dose, has a sponsor-reported ANCOVA analysis; an unadjusted recalculation is only an approximation and does not replace it. Missing-data handling, multiplicity procedures, the sample-size justification, and several design details are not reported. Serious adverse-event counts are posted, but the registry does not provide a formal comparison or clearly assign the listed serious-event term percentages to treatment arms.