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 Perianal Fistula (PAF) Validation and Burden of Illness Study (NCT03981939), a registered phase not stated study. The primary outcome is Objective 2: Mean Costs for Crohn's Disease Participants With and Without CPAF in ICES Database CPAF.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03981939 |
| Phase | Phase not stated |
| Condition | Inflammatory Bowel Disease, Perianal Fistula, Crohns Disease |
| Allocation | Not Stated |
| Design model | Not Stated |
| Masking | Not Stated |
| Enrollment | 3032 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Takeda |
| Results posted | Yes |
2. Clinical Question
Population
The study concerns participants diagnosed with Crohn's disease (CD), including participants with complex perianal fistula (CPAF) and participants without CPAF. The stated primary-outcome population is eligible cases and controls drawn from the Institute for Clinical Evaluative Sciences (ICES) database. A separate arm describes participants with CPAF from The Ottawa Hospital (TOH). The record reports enrollment of 3032, but does not report how that enrollment is divided among the groups or how many participants contributed to the primary outcome.
Intervention
No treatment or other clinical intervention is described. The intervention entry is “No intervention,” and the study uses retrospective observation of participant records. The question is about disease burden and health-care costs, rather than a comparison of assigned treatments.
Comparator
For the primary cost question, eligible CD participants with CPAF in the ICES database are considered alongside eligible CD participants without CPAF in that database. A separate TOH group includes participants with CPAF. The record does not establish that the TOH group is a like-for-like control for either ICES group, nor does it describe the matching procedure referred to in the official title.
Primary question
Objective 2 asks about the mean cost of CD among participants with and without CPAF in the ICES database. Costs are expressed in Canadian dollars and cover the period from the index date through the end of study, described as approximately 5 years post-index. The registry record describes cost by sector as well as total cost. It does not report the group means, their difference, or uncertainty around any comparison, so the direction and magnitude of the cost contrast cannot be assessed from the record.
3. Trial Design
This is a completed, retrospective observational study, not a randomized trial. The record describes a study intended to better understand disease burden and unmet need related to perianal fistulas within Crohn's disease in Ontario. No intervention was assigned. The allocation field is not stated, and the registry record specifically cautions that the arms are not a like-for-like control comparison.
Three groups are described. One consists of participants with CPAF from TOH; another consists of participants with CD and CPAF in the ICES database who met the case definition; the third consists of participants with CD but without CPAF in the ICES database who did not meet the case definition. Each group is described as observed retrospectively over the index period from April 1, 2007 to March 31, 2013. For the primary cost outcome, the stated analysis population is eligible ICES cases and controls, not the separate TOH group.
The record reports enrollment of 3032. It does not state the group-specific enrollment or the number included in the primary outcome analysis. The study start is listed as 2019-07-02, primary completion as 2019-12-17, and status as completed. These dates describe the registry study record; the observation window for the arms is separately specified as the index period.
The official title refers to procedure validation, a matched case control study, and a patient journey study. Details about the matching method, validation procedures, and patient-journey methods are not reported in the registry record. The record lists five outcome measures and no analysis entries. That absence means that the methods and results for inferential comparisons cannot be evaluated from the registry record alone.
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: Participants with a diagnosis of Crohn's disease (CD) with and without Complex Perianal Fistula (CPAF) were enrolled from The Ottawa Hospital (TOH) and Institute for Clinical Evaluative Sciences (ICES) database and were observed retrospectively in this study.
Overall Study
| Stage | CD Participants With CPAF - The Ottawa Hospital (TOH) | CD Participants Without CPAF (ICES Database) | CD Participants With CPAF (ICES Database) |
|---|---|---|---|
| STARTED | 549 | 1902 | 581 |
| COMPLETED | 549 | 1902 | 581 |
| NOT COMPLETED | 0 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: All participants identified from TOH with and without CPAF were included in the analysis. Participants could meet more than more case definition.
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: All eligible participants with CD, with CPAF or without CPAF from TOH and ICES database were included in the analysis.
| Characteristic | Unit | CD Participants With CPAF - TOH | CD Participants Without CPAF (ICES Database) | CD Participants With CPAF (ICES Database) | Total |
|---|---|---|---|---|---|
| Participants analysed | 549 | 1902 | 581 | 3032 | |
| Age, Continuous | years | 38.12 (26.3 to 48) | 31.83 (23 to 38) | 33.09 (23 to 40) | 34.34 (23 to 48) |
| Sex: Female, Male: Female | Participants | 304 | 903 | 279 | 1486 |
| Sex: Female, Male: Male | Participants | 245 | 999 | 302 | 1546 |
| Race and Ethnicity Not Collected | Participants | — | — | — | 0 |
| Region of Enrollment: Canada | Participants | 549 | 1902 | 581 | 3032 |
6. Primary Endpoint
The registry lists the primary outcome as Objective 2: Mean Costs for Crohn's Disease Participants With and Without CPAF in ICES Database CPAF, measured over From the index date up to end of study (approximately 5 years post-index).
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | canadian dollars |
| Endpoint classification | Continuous Endpoint |
The primary outcome is Objective 2: mean costs for CD participants with and without CPAF in the ICES database. Cost is measured in Canadian dollars. The registry record describes total disease-burden cost and costs by sector, including general physicians, specialists, all physician OHIP, home care services, hospital admissions, dialysis, cancer and radiation, emergency department, Ontario drug benefit, hospital outpatient clinics, and same day surgery. The stated summary measure is the mean, with standard deviation as the measure of dispersion.
The outcome covers the period from the index date to the end of study, described as approximately 5 years post-index. The index date is defined as the first date the participant meets the case definition for CD with CPAF during the index period. Eligible ICES cases and controls are the stated population for this analysis.
In general, a mean cost represents average expenditure across the participants included in an analysis, while a standard deviation describes how spread out individual costs are around that average. Mean costs are used to describe the overall cost burden, but they can be influenced by participants with unusually high costs. For cost data, the distribution and the chosen inferential method matter when interpreting the mean. The registry record does not report the mean costs, standard deviations, group denominators, or an estimated between-group contrast, so neither the size nor precision of the cost comparison can be judged.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a relevant observational comparison: health-care costs among eligible ICES participants with CD and CPAF versus eligible ICES participants with CD without CPAF. It specifies the outcome, currency, sectors, population, and follow-up interval. However, it does not report the numerical outcome results. There are no group means or standard deviations in the record, and no estimated contrast or confidence interval is reported. As a result, the posted results do not establish whether average costs were higher or lower in either group, or how large any difference might have been.
The description of a mean and standard deviation indicates the intended way to summarize a continuous cost measure, but it does not show the actual distribution of costs. In general, health-care costs may vary substantially between people, and a small number of high-cost observations can affect a mean. Without group-level estimates, distributional information, or inferential results, it is not possible to judge whether that issue was important in this study or how it was handled.
The study is retrospective and nonrandomized. Therefore, a difference in costs, if one were reported, would be an observed association between CPAF status and costs in the defined database population. It would not by itself establish that CPAF caused the difference. Participant characteristics, clinical history, access to care, and other factors could be related to both CPAF status and health-care costs. The registry record does not report an adjustment model or enough detail to assess the handling of such factors.
The record cautions that the arms are not a like-for-like control comparison. The primary outcome does specify ICES cases and controls, but matching details are not reported. The TOH CPAF group comes from a different source and is not included in the stated primary-outcome population. These distinctions matter: results from one group or data source should not be treated as interchangeable with another without supporting methods and estimates.
The registry lists enrollment of 3032, but does not report how many participants were included in this outcome analysis. It also lists five outcome measures and no analysis entries. The lack of reported estimates and analysis details prevents assessment of statistical precision, adjustment, missing-data handling, or the consistency of findings across outcomes. The study's stated objective provides context for the comparison, but the registry record alone does not permit a conclusion about the observed cost burden.
9. Statistical Methods Explained
The primary measure is a continuous outcome: cost in Canadian dollars. The registry record identifies the mean as the summary measure and the standard deviation as its dispersion measure. These describe different features. The mean gives an average across observations, while the standard deviation describes variability among those observations. Neither statistic on its own describes the full shape of the data. Because cost distributions can be asymmetric, interpretation of means may depend on whether a few very high costs are present. The registry record does not report distributional summaries or graphical diagnostics.
What does the mean cost comparison represent?
In general, a comparison of group means describes the difference in average outcome between the groups defined for analysis. For this study, those groups are eligible ICES participants with CD and CPAF and eligible ICES participants with CD without CPAF. A mean-cost contrast would describe the observed difference in average costs over the stated follow-up period. It would not, by itself, separate the association with CPAF from other differences between the groups. The registry record does not report the means, a contrast, or a method for estimating uncertainty around a contrast.
The listed cost domains include total cost and costs associated with multiple sectors of care. Sector-specific outcomes can help describe where costs arise, but multiple outcomes also create opportunities to examine many results. The record does not state whether total cost was a single prespecified primary analysis distinct from the sector measures, how sector results were compared, or whether any multiplicity procedure was used. Those details are not reported in the registry record.
Why does cost variability matter?
As a general statistical point, over-dispersion means that observed variability is greater than a model assumes. When cost data are highly variable or skewed, an analysis that relies on a simple normal approximation may represent uncertainty poorly unless its assumptions are reasonable for the data and design. Alternative approaches may be considered in some settings, but the registry record does not identify the method used for the cost comparison or provide data to assess its assumptions. It is therefore not possible to say whether over-dispersion was present or how it was handled.
A confidence interval, in general, expresses the uncertainty associated with an estimate under the analysis assumptions. A narrow interval indicates greater statistical precision than a wide interval, but neither determines whether an estimate is clinically or economically important. No confidence interval for the primary cost outcome is reported in the registry record. Nor are test statistics or p-values reported. Consequently, the statistical precision and compatibility of the data with different possible cost contrasts cannot be evaluated.
What do the planning assumptions mean?
The registry record describes sample-size illustrations based on a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustrations are framed around detecting a stated fraction of an observed effect. These are planning assumptions, not a description of the study's actual allocation or proof that the study was designed around those assumptions. The registry record does not report a sample-size justification for the study or the numerical planning estimates.
In general, a planning sample size depends on the assumed effect, outcome variability, significance threshold, power target, and allocation. An illustration based on a fraction of an observed effect explores how a future study might be sized under selected assumptions; it does not establish the power of this retrospective study. The registry record does not provide the observed cost effect needed to assess such illustrations, and the planning calculations are not observed power.
For non-time-to-event outcomes, the stated illustrations are expressed as participants per arm. This outcome is a continuous mean-cost measure, not a time-to-event outcome. The registry record does not provide numerical sample-size results or identify a formal power calculation used to design the study. No new sample-size estimate can be inferred from the enrollment figure alone because group allocation, outcome variability, and an observed effect are not reported for this analysis.
The registry lists no analysis entries. Thus the record does not specify the statistical test or model, any covariate adjustment, how repeated or sector-specific measures were handled, or whether sensitivity analyses were conducted. It also does not report the assumptions used in the primary analysis. General statistical principles can explain why these details matter, but they cannot substitute for the methods and results of this particular comparison.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists five outcome measures, but no analysis entries. It does not describe which outcomes were designated primary for analysis beyond the named Objective 2 cost measure, how the remaining outcomes were prioritized, or whether statistical tests were conducted across multiple outcomes. A multiplicity adjustment, hierarchy of testing, or other approach to interpreting multiple comparisons is not reported in the registry record. Without that information and without reported estimates, the effect of multiple outcome assessments on interpretation cannot be evaluated.
Missing-data methods are also not reported. The record does not state whether cost information was complete for all eligible cases and controls, how incomplete records were handled, or whether any imputation or sensitivity analysis was used. It does not provide the number analyzed for this outcome or the denominators for the comparison. Enrollment of 3032 is reported for the study overall, but that figure does not establish how many participants contributed to this particular analysis.
The primary-outcome population is described as all participants eligible as cases and controls from the ICES database. The registry record does not provide further analysis-population definitions, exclusions, or details on how eligibility was applied. Nor does it explain whether participant inclusion differed by cost sector or follow-up availability. These omissions limit assessment of the completeness and representativeness of the analyzed data. The record's lack of analysis details should not be interpreted as evidence that missingness or multiplicity was absent; those matters are not reported in the registry record.
11. What the Result Does, and Does Not, Mean
The stated research question concerns the average health-care costs associated with CD among people with and without CPAF in the ICES database. Its value is descriptive: it identifies the population, outcome, cost sectors, and period over which costs were intended to be assessed. The registry record does not contain numerical primary-outcome findings. It therefore does not show the average costs in either group, the size or direction of a difference, or the precision of any estimate.
Even if a cost contrast were reported, the retrospective observational design would require careful interpretation. In general, participants with and without CPAF may differ in ways that also relate to health-care use. An unadjusted difference would describe those groups as observed, rather than establish a causal effect of CPAF. The registry record does not describe an adjustment strategy, and it cautions that the arms are not a like-for-like control comparison.
The outcome covers costs from the index date to the end of study, described as approximately 5 years post-index. It includes total cost and costs across named sectors. This scope can address the economic burden recorded in those settings, but it does not, on its own, measure every dimension of disease burden or unmet need. The record's overall purpose includes those broader topics, while the primary outcome specifically concerns costs.
The registry records 0 serious events and 0 other events in the fields for each group. These entries should be read as registry-recorded values, not as proof that no adverse events occurred: the record does not explain the collection or reporting context for those fields, and the study is retrospective with no intervention. Similarly, the reported overall enrollment does not reveal the number analyzed for the cost outcome. The available information supports describing the study question and design, but not a conclusion about the magnitude, precision, cause, or practical importance of a cost difference.
12. Why This Trial Matters Statistically
This study addresses a statistical problem that arises in observational health-services research: describing costs for people with a condition alongside costs for a comparison group drawn from health records. Mean cost is interpretable as an average, but cost data can be variable and potentially asymmetric. Clear reporting of group sizes, means, dispersion, analysis methods, and uncertainty is important for readers assessing how stable and comparable such estimates are.
The registry record also illustrates why population and data source need to be explicit. It distinguishes a TOH group with CPAF from ICES groups with and without CPAF, while identifying the ICES cases and controls as the population for the primary cost outcome. That distinction prevents an assumption that every group contributes to the same analysis. The stated caution that the arms are not a like-for-like control comparison further emphasizes the need to interpret observational contrasts in light of how participants were selected.
For economic burden questions, the choice to report total and sector-specific costs can help describe resource use across different parts of care. At the same time, multiple outcomes make transparent analysis plans and multiplicity handling valuable. Missing-data methods and denominators are also important because cost estimates depend on which participants have usable information and what period their records cover.
The registry identifies these questions but does not report the numerical cost estimates or analysis details needed to answer them. The study is therefore statistically relevant as a clearly framed retrospective cost-comparison question, while the registry record alone cannot establish the observed burden or the precision of a comparison. This distinction between a defined outcome and a reported result is central to interpreting clinical and health-services research.
13. 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 main limitation for interpreting the primary outcome is that numerical results are not reported in the registry record. Group means, standard deviations, contrasts, confidence intervals, and p-values are absent. The group-specific numbers included in the analysis are also not stated. Without those details, neither the direction nor the size and precision of the comparison can be assessed.
The design is retrospective and nonrandomized, and the registry record cautions that the arms are not a like-for-like control comparison. Although the primary outcome focuses on ICES cases and controls, the matching method referenced in the official title is not described. The record does not report covariate adjustment or other methods for addressing differences between participants with and without CPAF. Causal attribution would therefore be unsupported by the information available.
Definitions and analytic details are incomplete for evaluating the cost outcome. The record defines the index date and lists cost sectors, but does not explain the construction of total cost in further detail, the statistical model, distributional checks, or the treatment of incomplete cost records. It does not report how many participants contributed data through the stated follow-up period or whether follow-up differed between groups. The five outcome measures and absence of analysis entries leave multiplicity and outcome prioritization unclear.
The stated overall enrollment of 3032 does not resolve these issues because group-specific and outcome-specific denominators are not reported. The adverse-event fields list zero values, but collection and reporting context are not described. These limitations concern what can be established from the registry record; they do not determine what may be available in other study documentation.
14. Related Tutorials
15. Related Statistical Calculators
16. Sources
- ClinicalTrials.gov: Perianal Fistula (PAF) Validation and Burden of Illness Study — NCT03981939.
- Registry record last updated 2021-04-21; status last verified 2021-03.
- Results first posted 2021-04-21.
- 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.
17. Record Summary
NCT03981939 is a completed, retrospective study of perianal fistula burden in Crohn's disease. The record describes three groups: participants with CPAF from TOH, participants with CD and CPAF from ICES, and participants with CD without CPAF from ICES. The index period is April 1, 2007 to March 31, 2013. Enrollment is listed as 3032, with study start on 2019-07-02 and primary completion on 2019-12-17.
The primary outcome, Objective 2, is mean cost in Canadian dollars for eligible ICES cases and controls, from the index date through the end of study, described as approximately 5 years post-index. Total and sector-specific costs are described. Numerical outcome results, group-specific analysis counts, and statistical analysis details are not reported in the registry record. The registry lists five outcome measures and no analysis entries. It also records zero serious and zero other events in the listed group fields, without further reporting context.