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 Reducing Arrhythmia in Dialysis by Adjusting the Rx Electrolytes/Ultrafiltration, Study A (NCT03519347), a randomized na study. The primary outcome is Adherence With Proposed Interventions.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03519347 |
| Phase | NA |
| Condition | End Stage Renal Disease |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Double |
| Enrollment | 19 |
| Arms in the primary outcome | 4 |
| Lead sponsor | NYU Langone Health |
| Results posted | Yes |
2. Clinical Question
Population
People with end stage renal disease receiving chronic hemodialysis were the intended population. The registry record reports enrollment of 19 participants. It does not describe further eligibility details or provide a separate definition of the population included in each outcome analysis.
Intervention
The study evaluated point-of-care testing before dialysis to guide adjustments to dialysate potassium and bicarbonate. The strategies aimed to maximize potassium removal, minimize potassium flux, avoid alkalosis, or avoid acidosis. The potassium and bicarbonate strategies address different treatment priorities; the record does not establish that each is a single combined intervention.
Comparator
The randomized arms are the four named strategies. The posted outcomes compare some strategies with the potassium removal maximization strategy. A separate randomized standard-of-care arm is not listed. One secondary measure concerns how often a point-of-care-guided prescription differs from a standard-of-care-guided prescription, but that is not the same as a direct randomized comparison against a standard-of-care arm.
Primary question
Could the study carry out the proposed point-of-care-guided approach in practice, as measured by the percentage of dialysis sessions in which testing was completed and the dialysate was adjusted according to the algorithm? The study also aimed to estimate arrhythmia-related outcomes. The adherence endpoint addresses feasibility; it does not itself measure whether arrhythmias were prevented.
The registry record describes a feasibility-focused study that also sought estimates of the extent to which dialysis prescription strategies might affect abnormal heart rhythms. The primary outcome is implementation adherence, not a clinical event rate. That distinction matters: high adherence can show that a protocol was followed during observed sessions, but it cannot on its own establish a reduction in arrhythmias or other health outcomes.
The study involved four strategy labels, with two focused on dialysate potassium and two on dialysate bicarbonate. The record describes point-of-care testing as the means of guiding adjustments. It does not report enough protocol detail to reconstruct how the strategies were sequenced, how long each was used, or whether particular participants received every strategy in a specific order.
3. Trial Design
The registry record describes a randomized, double-blind, crossover study, and the official title calls it a phase 2, multicenter study. The phase field itself is recorded as NA. Enrollment was 19, and the status is completed. The record gives a start date of 2018-11-30 and a primary completion date of 2022-07-12.
A crossover design means that participants may receive more than one study condition over time, so comparisons can be made within the same person as well as across conditions, depending on the planned analysis. That general feature can be useful when participant characteristics vary substantially. It also makes sequence, period, and carryover considerations important. The registry record does not report the treatment sequence, period lengths, washout arrangements, or the statistical handling of these crossover features. No assumptions about those details can be made from the posted results.
The outcome-level participant counts are 16 or 17 for the four strategies. These counts are not the same as the overall enrollment count, and they should not be added together as though they necessarily represented distinct people. In a crossover study, strategy-level counts can reflect people observed under more than one condition. The registry record does not explain the difference between enrollment and the participant counts attached to individual outcomes, or specify how missing observations were handled.
The listed interventions include four prescription strategies, point-of-care testing, and a cardiac monitor. The arm descriptions explain the potassium and bicarbonate priorities, but they do not supply a complete protocol. The registry record does not report the randomization method, allocation ratio, blinding procedures, or a detailed analysis plan. The listed outcome timeframe is up to Week 24; additional timing details 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.
Baseline Observation Period
| Stage | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 |
|---|---|---|---|---|
| STARTED | 5 | 4 | 4 | 6 |
| COMPLETED | 5 | 4 | 4 | 6 |
| NOT COMPLETED | 0 | 0 | 0 | 0 |
Interventional Crossover Period 1
| Stage | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 |
|---|---|---|---|---|
| STARTED | 5 | 4 | 4 | 6 |
| COMPLETED | 5 | 4 | 4 | 5 |
| NOT COMPLETED | 0 | 0 | 0 | 1 |
Reasons for not completing this period:
| Reason | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 |
|---|---|---|---|---|
| Death | 0 | 0 | 0 | 1 |
Interventional Crossover Period 2
| Stage | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 |
|---|---|---|---|---|
| STARTED | 4 | 4 | 4 | 5 |
| COMPLETED | 4 | 4 | 4 | 5 |
| NOT COMPLETED | 0 | 0 | 0 | 0 |
Interventional Crossover Period 3
| Stage | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 |
|---|---|---|---|---|
| STARTED | 3 | 4 | 4 | 5 |
| COMPLETED | 3 | 4 | 4 | 5 |
| NOT COMPLETED | 0 | 0 | 0 | 0 |
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.
| Characteristic | Unit | Intervention Sequence B1B2K1K2 | Intervention Sequence B2B1K2K1 | Intervention Sequence K1K2B1B2 | Intervention Sequence K2K1B2B1 | Total |
|---|---|---|---|---|---|---|
| Participants analysed | 5 | 4 | 4 | 6 | 19 | |
| Age, Continuous | years | 61.9 (SD 13.0) | 59.7 (SD 8.92) | 64.0 (SD 12.8) | 54.0 (SD 11.9) | 59.9 (SD 11.655) |
| Sex: Female, Male: Female | Participants | 1 | 1 | 0 | 1 | 3 |
| Sex: Female, Male: Male | Participants | 4 | 3 | 4 | 5 | 16 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 2 | 0 | 1 | 1 | 4 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 3 | 4 | 3 | 5 | 15 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 2 | 2 | 0 | 4 | 8 |
| Race (NIH/OMB): White | Participants | 1 | 2 | 3 | 1 | 7 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 2 | 0 | 1 | 1 | 4 |
| Region of Enrollment: United States | participants | 5 | 4 | 4 | 6 | 19 |
6. Primary Endpoint
The registry lists the primary outcome as Adherence With Proposed Interventions, measured over Up to Week 24.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Percentage of sessions |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is adherence with the proposed interventions, expressed as the percentage of sessions. The stated assessment is the percentage of sessions in which point-of-care testing was completed and dialysate was adjusted according to the algorithm, over a timeframe of up to Week 24. This measure was chosen to address whether the testing-and-adjustment procedure could be carried out during dialysis sessions.
The reported mean adherence is 88% for potassium removal maximization, with a 95% confidence interval of 82% to 92%. For potassium gradient minimization, the mean is 90%, with a 95% confidence interval of 85% to 94%. For alkalosis avoidance, it is 83%, with a 95% confidence interval of 77% to 88%; for acidosis avoidance, it is 80%, with a 95% confidence interval of 74% to 85%. The outcome-level participant counts range from 16 to 17.
A confidence interval describes uncertainty around an estimate under the methods used to construct it. These intervals give a range of values compatible with the observed adherence summaries and the interval procedure; they are not guarantees that the true value lies within the stated limits. The registry record does not report a formal comparison among the four adherence estimates. The endpoint is useful for assessing implementation, but it does not directly quantify arrhythmia prevention, safety, or comparative clinical benefit.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The primary results show that adherence was reported for all four strategies, with estimates between 80% and 90%. Their confidence intervals are not identical, but the registry record does not report a formal statistical comparison of adherence among strategies. The results therefore describe adherence within each strategy; they do not establish that one strategy was more feasible than another.
The secondary outcomes cover prescription changes relative to standard of care, atrial fibrillation duration, and potentially lethal arrhythmias, among other measures. Some posted mean differences and unadjusted p-values are available for selected strategy comparisons. These comparisons are summary-based recalculations, not analyses reported by the registry. They should be read as exploratory descriptions, particularly because the participant counts are small and multiple secondary outcomes and comparisons are present.
For the percentage of sessions in which the prescription differed from standard of care, the posted recalculation comparing potassium gradient minimization with potassium removal maximization is a mean difference of -14.6 percentage points, with a 95% confidence interval from -43.19 to 13.99 and an unadjusted p-value of 0.317. The interval includes no difference, so the summary does not provide precise evidence of a difference for this comparison. The alkalosis avoidance comparison is a mean difference of 34.4 percentage points, with a 95% confidence interval from 9.067 to 59.73 and an unadjusted p-value of 0.008. These results concern how often prescriptions differed from standard care, not arrhythmia events.
For mean duration of atrial fibrillation, the alkalosis avoidance comparison with potassium removal maximization is -270.0 seconds, with a 95% confidence interval from -1338.0 to 798.2 and an unadjusted p-value of 0.62. The interval is wide and includes no difference. The acidosis avoidance comparison is -1251.0 seconds, with a 95% confidence interval from -1303.0 to -1199.0 and an unadjusted p-value of <0.001. Its unusually large separation relative to other summaries deserves careful interpretation alongside the reported zero standard deviation in the potassium removal maximization group. The registry record does not report a model or analysis that explains this pattern.
For potentially lethal arrhythmias, the posted recalculation for potassium gradient minimization versus potassium removal maximization is a mean difference of 0.0, with a 95% confidence interval from -0.901 to 0.901 and an unadjusted p-value of 1.0. The alkalosis avoidance comparison is -2.0, with a 95% confidence interval from -2.627 to -1.373 and an unadjusted p-value of <0.001. These summaries should not be interpreted as confirmatory treatment effects: the record describes the outcome as a count but posts means and standard deviations, and the analysis plan is not reported.
No registry-reported inferential analyses are listed for these secondary outcomes. The recalculated estimates do not account for a reported crossover model, multiplicity, or other potential design features. The strongest interpretation is descriptive: the posted results offer preliminary estimates that may guide future study planning, while leaving substantial uncertainty about how the strategies compare on clinical outcomes.
9. Statistical Methods Explained
Statistical comparisons in the posted secondary summaries are unadjusted recalculations from group-level means, standard deviations, and participant counts. They are not sponsor-reported analyses. A recalculated mean difference describes the difference between two reported group means in the direction stated by the comparison. A confidence interval gives a range of estimates compatible with the observed summaries under the calculation method. When an interval includes zero, the data do not rule out no mean difference under that method. When it does not include zero, that is not, by itself, proof of a clinically important or reproducible effect.
What do the confidence intervals describe?
A confidence interval reflects sampling uncertainty as well as the assumptions of its calculation. Wider intervals indicate less precision than narrower intervals. Precision depends in part on the amount of information in the data and the variability of the outcome. In this study, outcome-level counts are small, and several secondary summaries show substantial variability. The intervals should therefore be interpreted as uncertain estimates, not as exact boundaries on a treatment effect.
The primary adherence results are reported as means with 95% confidence intervals. The registry record does not state the method used to construct those primary intervals. The secondary recalculations have their own stated estimates and intervals, but the record does not report a model accounting for crossover sequence or repeated observations. A calculation using group summaries cannot recover the within-participant correlation that may be important in a crossover trial. This is a general limitation of reconstructing comparisons from aggregate summaries.
Why do unadjusted p-values need caution?
A p-value is a measure of how incompatible the observed data are with a specified null model, assuming that model and its assumptions are appropriate. It is not the probability that a treatment has no effect, and it does not measure the size or importance of an effect. An unadjusted p-value does not account for the number of outcomes or comparisons examined. When many comparisons are considered, some may appear unusual by chance. The registry record does not report a multiplicity strategy, so the posted unadjusted values should be treated as exploratory rather than as a confirmatory family of tests.
The registry record lists the arrhythmia count outcome using a mean and standard deviation. As a general statistical matter, event counts may be skewed, include many zeros, or vary more than a simple normal model expects. Over-dispersion means that the observed variability is greater than a model allows; ignoring it can make uncertainty estimates misleading. Count-based approaches may be more suitable in some settings, but the appropriate method depends on the outcome distribution and design. No such model is reported in the registry record, so the mean-based summaries should not be taken as evidence that a particular distributional model was used.
What do the planning illustrations mean?
The planning illustrations are based on observed effects and are not the trial's own sample-size justification. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They show participant requirements for 80% or 90% power when planning for 100%, 75%, or 50% of the observed effect. These are hypothetical planning scenarios, not observed power and not evidence that a future study will achieve the stated power.
Planning from an observed effect in a small study can be unstable. An unusually large observed difference may produce very small calculated requirements, while a modest or noisy effect may produce much larger requirements. The zero standard deviation recorded for some groups can also affect summary-based calculations in ways that may not reflect realistic future variability. The illustrations should therefore be treated as sensitivity exercises around assumed effects, not as definitive sample-size recommendations.
General statistical interpretation also depends on the relationship between the endpoint and the research question. Adherence is a feasibility measure; arrhythmia outcomes are clinical or event-related measures. A high adherence estimate cannot substitute for evidence about clinical outcomes. Similarly, an apparent difference in an intermediate measure, such as prescription changes, does not establish that arrhythmia risk changed. Each outcome answers a different question and may require a different analysis.
Because this is described as a crossover study, a full analysis could need to account for participant-level pairing, period effects, treatment sequence, and possible carryover. These considerations are general features of crossover designs. The record does not report how they were handled, and the recalculated comparisons cannot supply that missing design information. They are best understood as simple contrasts of posted summaries rather than reconstructions of a complete trial analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry record identifies multiple primary and secondary outcome measures, and the posted results include several secondary strategy comparisons. It does not report a plan for controlling multiplicity, such as a hierarchy of hypotheses or an adjustment to account for testing multiple outcomes. Without that information, unadjusted p-values cannot be interpreted as though they were the result of a single prespecified test. A small p-value among many exploratory comparisons may overstate the strength of evidence if considered in isolation.
The record does not report how missing outcome data were handled, whether analyses used all available observations, or whether any imputation was used. It also does not give reasons for differences between the overall enrollment count and the outcome-level counts. These omissions matter because missingness can affect both the estimated outcome and its uncertainty, particularly if the probability of having an observed result is related to a participant's condition or experience.
The analysis populations are not defined in the registry record. It is not reported whether the results use an intention-to-treat population, a per-protocol population, or another set of participants. The registry record also does not describe exclusions after randomization, adherence-related exclusions, or the handling of participants who did not complete all crossover conditions. No participant flow details or missing-data sensitivity analyses are reported.
For a crossover study, the treatment sequence, period effects, and within-participant correlations may influence estimates and uncertainty. The record does not report whether these features were incorporated into analyses. The summary-based recalculations cannot adjust for them. Therefore, the displayed comparisons should not be treated as substitutes for a prespecified participant-level analysis. The absence of reported multiplicity and missing-data methods does not establish that no such methods were used; it means that they are not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Potassium Removal Maximization Strategy | 3 | 17 | 17.6% |
| Potassium Gradient Minimization Strategy | 3 | 16 | 18.8% |
| Alkalosis Avoidance Strategy | 3 | 16 | 18.8% |
| Acidosis Avoidance Strategy | 5 | 17 | 29.4% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Potassium Removal Maximization Strategy | 1 of 17 |
| Potassium Gradient Minimization Strategy | 0 of 16 |
| Alkalosis Avoidance Strategy | 1 of 16 |
| Acidosis Avoidance Strategy | 2 of 17 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Potassium Removal Maximization Strategy | 4 of 17 |
| Potassium Gradient Minimization Strategy | 3 of 16 |
| Alkalosis Avoidance Strategy | 1 of 16 |
| Acidosis Avoidance Strategy | 3 of 17 |
Most frequent serious adverse event terms
| Event term | Organ system | Potassium Removal Maximization Strategy | Potassium Gradient Minimization Strategy | Alkalosis Avoidance Strategy | Acidosis Avoidance Strategy |
|---|---|---|---|---|---|
| Vascular access site complication | Vascular disorders | 2 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Atrial Fibrillation | Cardiac disorders | 0 of 17 | 0 of 16 | 2 of 16 | 0 of 17 |
| COVID-19 | Infections and infestations | 0 of 17 | 0 of 16 | 1 of 16 | 1 of 17 |
| Peripheral Angiogram | Surgical and medical procedures | 0 of 17 | 0 of 16 | 0 of 16 | 1 of 17 |
| Coronary artery stenosis | Cardiac disorders | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Encephalopathy | Nervous system disorders | 0 of 17 | 0 of 16 | 0 of 16 | 1 of 17 |
| Hyperkalemia | Metabolism and nutrition disorders | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Hyperparathyroidism | Endocrine disorders | 0 of 17 | 0 of 16 | 0 of 16 | 1 of 17 |
| Illeostomy closure | Surgical and medical procedures | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Kidney stones | Renal and urinary disorders | 0 of 17 | 0 of 16 | 0 of 16 | 1 of 17 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Potassium Removal Maximization Strategy | Potassium Gradient Minimization Strategy | Alkalosis Avoidance Strategy | Acidosis Avoidance Strategy |
|---|---|---|---|---|---|
| Kidney stones | Renal and urinary disorders | 1 of 17 | 1 of 16 | 0 of 16 | 1 of 17 |
| Vascular access site complication | Vascular disorders | 1 of 17 | 0 of 16 | 1 of 16 | 1 of 17 |
| Abcess | Infections and infestations | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Athralgia | Musculoskeletal and connective tissue disorders | 0 of 17 | 0 of 16 | 1 of 16 | 0 of 17 |
| Cramping | Vascular disorders | 0 of 17 | 0 of 16 | 0 of 16 | 1 of 17 |
| Gastroenteritis | Gastrointestinal disorders | 1 of 17 | 0 of 16 | 0 of 16 | 0 of 17 |
| Hypokalemia | Metabolism and nutrition disorders | 1 of 17 | 0 of 16 | 0 of 16 | 0 of 17 |
| Palpitations | Cardiac disorders | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
| Puncture wound | Injury, poisoning and procedural complications | 0 of 17 | 1 of 16 | 0 of 16 | 0 of 17 |
Serious events with the highest arm-level rate: atrial fibrillation (12.5%), vascular access site complication (11.8%), coronary artery stenosis (6.2%), covid-19 (6.2%), hyperkalemia (6.2%).
The adverse-event summaries report the number at risk and the number affected for serious events and other events under each strategy, as well as deaths. For serious events, the affected counts are 3, 3, 3, and 5 across the four strategy groups; the corresponding at-risk counts are 17, 16, 16, and 17. The other-event counts are 4, 3, 1, and 3, with the same respective at-risk counts. The listed death counts are 1, 0, 1, and 2.
These figures are descriptive. The registry record does not report whether events were attributed to a particular intervention, whether the same person could contribute events under more than one strategy, or whether the counts represent distinct people across the crossover periods. It also does not provide an adjusted comparison of safety outcomes between strategies. The serious-event terms listed include atrial fibrillation at 12.5%, vascular access site complication at 11.8%, and coronary artery stenosis, COVID-19, hyperkalemia, and illeostomy closure at 6.2% each. The record does not assign those term percentages to individual strategies.
The small participant counts and incomplete details limit comparisons. These summaries cannot establish whether an event was caused by a strategy or whether safety differed between strategies. They should be read as the events recorded in the registry, not as a comparative safety conclusion.
12. What the Result Does, and Does Not, Mean
The primary result is evidence about implementation: the registry reports mean adherence estimates for each of the four strategies, with confidence intervals. Those estimates indicate that the prescribed testing-and-adjustment procedure was carried out in a substantial share of sessions in the observed study. They do not show that the strategies prevented abnormal heart rhythms, reduced hospitalization, or changed mortality.
The study's design and outcomes reflect an early evaluation of feasibility and clinical estimates. The randomized crossover structure can be useful for comparing strategies, but a valid interpretation depends on design and analysis details that are not reported, including treatment sequence, period handling, and the analysis population. The outcome-level counts are small. Estimates from a small study can be imprecise and may vary substantially in a larger evaluation.
The secondary summaries offer possible signals, not settled answers. Some unadjusted recalculations show differences between strategy summaries, while others are compatible with no difference. These comparisons are not registry-reported analyses, do not account for a reported multiplicity plan, and may not reflect the crossover design. A low p-value does not establish clinical importance, and a confidence interval does not resolve missing protocol or analysis information.
The results also do not establish that any strategy is preferable to standard care. A separate randomized standard-of-care arm is not listed, and the outcome concerning differences from standard-of-care-guided prescriptions measures prescription divergence rather than a direct clinical comparison. The record does not report a sample-size justification or confirmatory statistical plan.
Overall, the posted data describe adherence and exploratory outcome patterns in this completed study. They can inform which questions deserve further testing and what uncertainties a larger, fully specified analysis should address. They should not be read as medical advice or as proof that a treatment works or does not work.
13. Why This Trial Matters Statistically
This study raises a useful statistical question that precedes the usual efficacy question: can a lab-guided, session-by-session prescription strategy be implemented consistently during hemodialysis? The primary endpoint directly targets that feasibility issue. Without evidence that the procedures can be carried out, a larger trial of clinical outcomes may be difficult to interpret because poor implementation could obscure the effect of the intended strategy.
The study also illustrates why feasibility and clinical outcomes must remain distinct. Adherence summarizes whether procedures were followed; arrhythmia measures summarize clinical observations. A successful result on one does not automatically imply a result on the other. Keeping those estimands separate avoids overstating what an implementation measure can demonstrate.
Its crossover design makes within-participant comparisons potentially informative, but it also demands design-aware analysis. Period, sequence, carryover, and repeated observations can affect uncertainty. The registry record does not report the analysis approach for these features, so summary-based contrasts cannot fully represent the design. This is a practical reminder that an appealing design does not remove the need for a transparent analysis plan.
Finally, the study provides a case for treating small-trial secondary findings as groundwork. Observed effects can help formulate hypotheses and generate planning scenarios, but sample-size illustrations derived from observed effects are sensitive to random variation. Clear reporting of analysis populations, missingness, multiplicity, crossover methods, and prespecified comparisons would make later evidence easier to interpret and compare.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
Several limitations are specific to interpreting this registry record. First, overall enrollment was 19, whereas the outcome-level counts range from 16 to 17. Because the design is crossover, those counts should not be treated as independent groups or added to estimate the number of unique participants. The registry record does not explain the differences or report participant flow.
Second, key details needed to evaluate the crossover are not reported: treatment sequence, period duration, washout, potential carryover, and the analysis of repeated measurements. Randomized and double-blind labels alone do not reveal how those features were implemented. The official title describes a phase 2 study, while the registry phase entry is NA.
Third, the primary endpoint is an implementation measure, while several secondary outcomes concern clinical events or arrhythmia measures. The record does not report a prespecified statistical plan tying these outcomes together. It also does not report multiplicity handling, missing-data methods, analysis populations, or the methods used for the primary confidence intervals.
Fourth, several secondary comparisons are recalculations from aggregate summaries rather than registry-reported analyses. Aggregate data cannot capture participant-level pairing in a crossover design. The posted count outcome is summarized using means and standard deviations, and some groups have zero reported standard deviation. These features can make simple normal-based comparisons fragile; the record does not describe a count model or explain the zero variability.
Finally, the registry does not list a separate randomized standard-of-care arm, despite reporting an outcome about prescription differences from standard care. The event summaries do not clarify attribution or overlap across crossover conditions. These reporting gaps limit the strength of comparative and causal interpretations.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Reducing Arrhythmia in Dialysis by Adjusting the Rx Electrolytes/Ultrafiltration, Study A — NCT03519347.
- Registry record last updated 2023-08-15; status last verified 2023-07.
- Results first posted 2023-08-15.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The completed registry record describes a randomized, double-blind crossover study with enrollment of 19 participants and a primary timeframe up to Week 24. Its official title calls it a phase 2, multicenter study; the phase entry is NA. The central aim was to evaluate feasibility of point-of-care-guided dialysate potassium and bicarbonate adjustment and to estimate effects on arrhythmia-related outcomes.
The primary endpoint was adherence, reported as mean percentages with 95% confidence intervals for four strategies. Secondary outcomes included prescription differences from standard care, atrial fibrillation duration, potentially lethal arrhythmias, enrollment, hospitalization, and mortality. Outcome-level counts were 16 or 17. Registry-reported inferential analyses are not listed, and several posted pairwise estimates are unadjusted summary-based recalculations.
The results are best understood as feasibility information and exploratory outcome estimates. The registry record does not report important analysis details, including handling of crossover features, missing data, multiplicity, or analysis populations. Those omissions, together with the small outcome-level counts, limit comparative conclusions.