Non-Interventional Study Describing Direct Costs Related to Anti-coagulation Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Non-Interventional Study Describing Direct Costs Related to Anti-coagulation Treatment (NCT04435769), a registered phase not stated study. The primary outcome is CHA2-DS2-VASc Score at Baseline.

2. Clinical Question

The score is a risk-scoring measure for stroke and systemic emboli in participants with NVAF. Its presence as the primary outcome should be distinguished from the study’s stated cost objective: the registered primary outcome is a baseline clinical risk score, not a direct measure of treatment cost or a count of subsequent strokes. The registry describes two cohorts, one using apixaban and one using warfarin for secondary stroke or TIA prevention. Because treatment allocation is not stated, cohort membership should be understood as observed treatment, not as randomized assignment.

The record reports 109 enrolled participants, but does not report the primary outcome’s cohort-specific means or standard deviations. It therefore cannot answer how the baseline score was distributed in either cohort, nor whether the cohorts were similar in measured baseline risk. The clinical question is descriptive rather than a causal comparison: the record describes patients receiving the two treatments, but does not establish what would have happened to the same patients under the other treatment.

3. Trial Design

This completed study is identified as non-interventional and describes direct costs related to anticoagulation treatment. Its official title specifies patients with nonvalvular atrial fibrillation in secondary stroke prevention who were prescribed apixaban or warfarin. The registry describes two cohorts: one of patients using warfarin and one of patients using apixaban. It reports 109 enrolled participants. Pfizer is listed as the sponsor.

The study began on 2020-08-04, and the primary completion date is 2021-05-31. The registry status is completed. These record dates describe the study timeline; they do not establish the length of follow-up for every participant. The primary outcome timeframe is baseline, based on retrospective data retrieved in the study. The outcome population is described as participants who followed the visit 2 schedule at 6 months, plus or minus 10 weeks, met all inclusion criteria, and did not meet any exclusion criteria.

The design is not a randomized comparison. Allocation is not stated, and the registry explicitly cautions that the arms are not a like-for-like control comparison. Treatment choice and cohort membership therefore cannot be interpreted as if the study had assigned otherwise comparable participants to apixaban or warfarin. Differences between cohorts, if reported, could reflect differences in the patients or circumstances associated with their treatment, in addition to any treatment-related factors.

The registry lists 23 outcome measures and 0 analyses. A protocol-level explanation for the enrollment target, a sample-size justification, and a detailed statistical analysis plan are not reported in the registry record. The record does not state whether the study used interim analyses. It also does not report a randomized allocation procedure. Those absences limit what can be concluded about how the study was planned and how its results should be interpreted beyond the descriptions that are registered.

4. Participant Flow and Accounting

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

Overall Study

Analysis population for the primary outcome, as defined in the registry: PP population comprised of participants who followed visit 2 schedule at 6 months (+/- 10 weeks), met all inclusion criteria and did not meet any of the exclusion criteria.

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: PP population comprised of participants who followed visit 2 schedule at 6 months (+/- 10 weeks), met all inclusion criteria and did not meet any of the exclusion criteria.

6. Primary Endpoint

The registry lists the primary outcome as CHA2-DS2-VASc Score at Baseline, measured over Baseline (from retrospective data retrieved in the study).

The primary outcome is the mean CHA2DS2-VASc score at baseline, expressed in units on a scale, with standard deviation as the stated measure of dispersion. The score is used to estimate the risk of stroke and systemic emboli among participants with nonvalvular atrial fibrillation. It is calculated from the risk factors described in the registry: age categories, female sex, history of congestive heart failure, hypertension, stroke, TIA or thromboembolism, vascular disease, and diabetes mellitus. The registry states that the score ranges from 0-9, with 0 indicating low risk and 9 indicating high risk.

As a baseline measure, the score characterizes participant risk at the beginning of the outcome assessment; it is not itself a count of strokes or emboli occurring during follow-up. Reporting a mean describes the cohort’s average score, while the standard deviation describes how dispersed individual scores are around that average. These are complementary summaries: an average alone can conceal how varied the participants’ scores are.

Baseline risk scores can be useful for describing the composition of clinical cohorts. In a non-randomized study, they may help readers consider whether the treatment groups differ in an important measured characteristic. However, a single score cannot establish that the groups are otherwise comparable, and a baseline score is not evidence that one treatment caused a later outcome. The numerical baseline means and standard deviations are not reported in the registry record. The registered endpoint therefore identifies what was intended to be described, but does not give the values needed to assess the observed score distributions.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the endpoint summary represent?

The primary outcome is a continuous measure: CHA2DS2-VASc score at baseline. For a continuous measure, a mean is an average across the participants included in the analysis. The standard deviation describes the extent to which participant values vary around that mean. A mean and standard deviation answer different questions: the mean locates the center of the distribution, while the standard deviation describes its spread. Neither one alone reveals the full shape of the data or identifies every participant’s score.

The score is intended to estimate stroke and systemic embolism risk in people with NVAF. The registry describes its scale as ranging from 0-9 and lists the risk factors used to calculate it. The score’s role in this study is descriptive. A higher score represents higher estimated risk according to the scale description, but the score is not a direct observation of a stroke or embolism occurring during the study. Nor does it measure treatment cost, despite the study’s broader cost objective.

The outcome timeframe is baseline, using retrospective data retrieved in the study. The listed population is the per-protocol population, defined by following the visit 2 schedule at 6 months, plus or minus 10 weeks, meeting all inclusion criteria, and not meeting any exclusion criteria. This definition identifies a selected analysis population; it should not automatically be treated as identical to every participant enrolled. The registry does not report how many participants contributed baseline scores in each cohort.

How should a cohort comparison be interpreted?

In a randomized study, random allocation is intended to make treatment groups comparable on average, including with respect to characteristics that were not measured. This study is described as non-interventional, and allocation is not stated. The registry specifically says the arms are not a like-for-like control comparison. Thus, a difference between cohorts would be an observed association, not by itself a treatment effect. Differences in baseline risk, clinical circumstances, or other factors could contribute to an observed contrast.

In general, confounding occurs when factors associated with both treatment received and the outcome influence the observed relationship. Adjustment methods can sometimes address measured confounders, but they depend on the variables collected and the statistical model. They cannot guarantee correction for unmeasured factors. The registry lists 0 analyses, and an adjustment strategy is not reported in the registry record. No adjusted treatment comparison should be inferred from the cohort descriptions.

What do uncertainty measures usually add?

In general, a confidence interval expresses uncertainty around an estimate under the chosen model. A narrow interval indicates greater statistical precision than a wide interval, while the interval’s location indicates which values are compatible with the data and assumptions. A confidence interval does not account automatically for bias from non-random treatment allocation, missing data, or outcome measurement. No confidence interval for the primary score is reported in the registry record.

In general, a p-value measures how incompatible observed data are with a specified null hypothesis under a statistical model. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of a difference. Because 0 analyses are listed and no primary-outcome test is reported, there is no registry-reported p-value on which to base an inferential conclusion. The analysis plan, including any model used for the mean score, is not reported in the registry record.

Standard deviation should not be confused with standard error. In general, standard deviation describes participant-level variability, whereas standard error describes uncertainty in an estimated mean. The registry specifies standard deviation as the dispersion measure for the primary outcome. The numerical means and standard deviations themselves are not reported in the registry record, so neither the observed variability nor the precision of the group means can be evaluated from the posted information.

Why can over-dispersion matter?

General explanation: over-dispersion occurs when observed variation is greater than a statistical model assumes. It matters because a model that understates variation can produce confidence intervals that are too narrow and tests that appear more decisive than warranted. The concern depends on the outcome and model. The primary endpoint is a score summarized by a mean and standard deviation; a model for that outcome is not reported in the registry record. It would therefore be inappropriate to claim that over-dispersion was assessed or that a particular adjustment was used.

What do the sample-size illustrations mean?

The planning illustrations are described as calculations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; the target effect is a stated fraction of the observed effect. For time-to-event outcomes, the stated convention is a total number of events; for other outcomes, it is participants per arm. These assumptions describe planning illustrations, not the study’s own design. The sample-size justification and the method used to set enrollment are not reported in the registry record.

General explanation: a planning calculation asks how much information might be needed to detect a specified effect under stated assumptions. It is sensitive to the assumed effect, variability, allocation, and analysis method. A calculation using an effect estimated from observed data is not independent confirmation of that effect. It also is not observed power, which is not a useful substitute for examining the estimate and its uncertainty. The illustrations do not show that this non-interventional study was designed with those assumptions or that its cohorts satisfy equal allocation. The record reports 0 analyses, so no inferential analysis corresponding to the illustrations is reported.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Apixaban", "Warfarin")
n   <- c(40, 63)
m   <- c(4.6, 4.3)
sp  <- c(1.5, 1.7)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Apixaban  40  4.600  1.500  4.120  5.080
CARM  Warfarin  63  4.300  1.700  3.872  4.728

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 23 outcome measures and 0 analyses. It does not report a plan for controlling multiplicity, such as a hierarchy of primary and secondary tests or an adjustment for testing multiple outcomes. Without such details, the family of hypotheses to be considered and the interpretation of any nominal p-values cannot be determined. No p-values for the primary outcome are reported in the registry record, so there is no basis for assessing statistical significance or multiplicity-adjusted inference.

The missing-data approach is not reported in the registry record. In particular, the record does not state how missing baseline scores or incomplete follow-up were handled, whether participants were excluded from particular outcome summaries, or whether any imputation method was used. The defined per-protocol population includes participants who followed the visit 2 schedule at 6 months, plus or minus 10 weeks, met all inclusion criteria, and did not meet any exclusion criteria. That definition does not, by itself, explain the amount or pattern of missing data or establish how missingness may affect the summaries.

The registered outcome population and the overall enrollment should also be kept distinct. The registry reports 109 enrolled participants, but the number contributing to the primary outcome by cohort is not reported in the registry record. No analysis population beyond the stated per-protocol definition is described in the available record, and no sensitivity analysis is reported. It is therefore not possible to assess how results might change under different assumptions about incomplete records or population eligibility.

These omissions matter because missingness and multiple outcome testing can affect both the interpretation and the apparent precision of results. The registry does not report enough analysis detail to determine whether these issues were evaluated. That uncertainty is separate from the non-randomized design, which independently limits causal interpretation.

12. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: stroke/tia (9.0%), intracranial bleeding (3.0%), cardiac ischemia (2.4%), gi bleeding (1.5%), intraparenchymal hematoma of the lower lobe (1.5%).

For apixaban, the registry reports 42 participants at risk for both serious and other adverse events; 2 were affected by serious adverse events and 4 by other adverse events. For warfarin, it reports 67 participants at risk for both categories; 10 were affected by serious adverse events and 0 by other adverse events. Deaths are reported as 0 for apixaban and 1 for warfarin. These are descriptive counts in treatment cohorts, not results from randomized safety groups.

The registry also lists serious-event terms with percentages: Stroke/TIA, 9.0%; Intracranial bleeding, 3.0%; Cardiac ischemia, 2.4%; GI bleeding, 1.5%; Intraparenchymal hematoma of the lower lobe, 1.5%; and Diarrhoea, 1.5%. The registry record does not report the denominators or counting rules for these percentages, or whether an individual could contribute more than one event term. They should not be treated as directly comparable cohort-specific rates.

These figures do not establish that a treatment caused or prevented an event. The arms are not a like-for-like control comparison, and the registry does not state allocation. Differences in exposure, participant characteristics, or event ascertainment could affect the counts. The posted safety information is descriptive and does not substitute for a causal comparative safety analysis.

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

The record establishes that a completed, non-interventional study followed cohorts of patients with NVAF receiving apixaban or warfarin for secondary stroke or TIA prevention. It records a total enrollment of 109 and identifies baseline CHA2DS2-VASc score as the primary outcome, with the mean and standard deviation as the intended summary. It also states that the broader objective was to describe direct costs related to the two anticoagulation treatments.

However, the primary score values are not reported in the registry record. There is no reported cohort mean or standard deviation with which to describe the observed baseline risk distribution. The record therefore does not establish whether one cohort had a higher average score, how variable the scores were, or how precisely any cohort mean was estimated. The absence of those numerical results prevents a quantitative interpretation of the primary outcome.

The registry’s descriptions also do not demonstrate that the treatment cohorts were comparable. The study was non-interventional, allocation is not stated, and the record cautions that the arms are not a like-for-like control comparison. Even if a numerical difference in an outcome were available, it would not on its own demonstrate that treatment caused that difference. A descriptive cohort contrast and a causal treatment effect are different claims.

The safety counts and serious-event terms provide descriptive information, but do not resolve that limitation. Counts of affected participants and deaths do not establish causation, and the record does not state the denominators and event-counting details needed to interpret the listed term percentages as comparative rates. The listed 23 outcome measures and 0 analyses likewise do not supply missing estimates or an analysis plan.

Thus, the record is informative about the study’s purpose, cohort structure, registered primary outcome, and selected safety counts. It does not provide the numerical primary outcome results or a randomized treatment comparison. Conclusions about comparative effectiveness, causal safety differences, or the size and precision of a baseline-score contrast are not supported by the registry record.

14. Why This Trial Matters Statistically

This study illustrates an important distinction in clinical statistics: a study objective, a registered outcome, and a reported result are not interchangeable. The stated objective concerns direct costs related to anticoagulation treatment, while the primary outcome is a baseline CHA2DS2-VASc score. The score describes estimated stroke and systemic embolism risk; it is not a direct cost measure. Recognizing that distinction helps readers avoid assuming that a listed primary outcome answers every question named in a study’s title or objective.

It also illustrates why study design determines the strength of an interpretation. A non-interventional comparison of patients receiving different treatments can describe observed cohorts. Without randomized allocation, a difference cannot be attributed to treatment alone. The registry explicitly cautions that these arms are not a like-for-like control comparison. Statistical adjustment, if undertaken, would not automatically remove confounding, and an adjustment plan is not reported in the registry record.

Transparent result reporting is essential for evaluating an outcome. The registry names the primary score and its intended summaries, but the numerical means and standard deviations are not reported. Without estimates, readers cannot assess the group distributions or their precision. The record’s listing of 23 outcome measures alongside 0 analyses further underscores that outcome registration is not the same as reporting an inferential analysis.

Finally, safety counts and sample-size illustrations require careful framing. Counts describe recorded events but do not establish causation in non-randomized cohorts. Planning calculations based on an observed effect are not observed power and do not show that the study was designed to detect that effect. These distinctions are broadly useful when interpreting registry records: design, reporting completeness, and the meaning of statistical summaries all matter as much as the outcome labels themselves.

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.

A central limitation is the absence of numerical results for the primary outcome. Although the registry identifies baseline CHA2DS2-VASc score, the mean and standard deviation for either cohort are not reported in the registry record. The cohort-specific number of participants contributing to that endpoint is also not reported. This prevents assessment of the baseline score distributions and their precision.

The non-interventional design limits causal interpretation. Treatment allocation is not stated, and the registry warns that the cohorts are not a like-for-like control comparison. The record does not report an adjustment strategy or enough detail to evaluate measured or unmeasured confounding. Thus, a difference between cohorts, if reported elsewhere, could not automatically be attributed to treatment based on this record.

The study’s stated focus on direct costs is not matched by a numerical cost result among the registry entries reported in the registry record. While the record identifies 23 outcome measures, it lists 0 analyses. The statistical analysis plan, sample-size justification, interim-analysis approach, and handling of missing data are not reported in the registry record. Multiplicity procedures and sensitivity analyses are also not reported. These omissions make it difficult to evaluate how outcomes were selected, modeled, or protected against analytical bias.

The per-protocol outcome population is defined by compliance with the visit 2 schedule at 6 months, plus or minus 10 weeks, and by meeting eligibility criteria. The record does not quantify exclusions or explain how missing observations were addressed. The safety information gives affected-participant counts and serious-event term percentages, but event definitions, percentage denominators, and counting rules are not reported in the registry record. The safety counts are therefore limited descriptive evidence, not a complete comparative assessment.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Non-Interventional Study Describing Direct Costs Related to Anti-coagulation Treatment — NCT04435769.
• Registry record last updated 2023-02-08; status last verified 2022-05.
• Results first posted 2023-02-08.
• 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.

19. Record Summary

NCT04435769 is a completed non-interventional study sponsored by Pfizer. It enrolled 109 participants with NVAF in cohorts receiving apixaban or warfarin for secondary stroke or TIA prevention. The study began on 2020-08-04, with primary completion on 2021-05-31. Its stated objective was to describe direct costs related to anticoagulation treatment.

The primary outcome was baseline CHA2DS2-VASc score, summarized as a mean with standard deviation. Numerical results for this outcome are not reported in the registry record. The record lists 23 outcome measures and 0 analyses. Its safety information reports affected-participant counts, deaths, and selected serious-event terms, but does not establish a causal comparison between cohorts.

The registry states that allocation is not stated and cautions that the cohorts are not a like-for-like control comparison. Statistical analysis details, missing-data methods, multiplicity procedures, and a sample-size justification are not reported in the registry record. Interpretation is therefore limited to the study description and the descriptive safety information recorded.