COMBACTE-CDI Understanding the Burden of C. Difficile Infection: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for COMBACTE-CDI Understanding the Burden of C. Difficile Infection (NCT03503474), a registered phase not stated study. The primary outcome is The Number of Cases of CDI.

2. Clinical Question

The study’s broader aim, as described in the registry, was to quantify the burden of CDI and describe current management practices. The background identifies missed cases as a concern, including cases that may be missed because of diagnostic testing limitations or a lack of clinical suspicion. Central testing of samples sent to participating sites was intended to look for cases not identified through the original testing process.

The registry also describes a follow-up case-control study to collect information on outcomes and risk factors. It states that the data would be used to construct transmission models and examine cost effectiveness. These aims extend beyond the primary count of positive and negative test results. However, the registry record does not report the follow-up results, transmission-model findings or cost-effectiveness findings.

This is therefore a question about case identification and burden in the sampled population, not a randomized comparison of treatments. The positive and negative groups are defined by the laboratory result. The record does not present a treatment allocation or an intervention-versus-control effect. Nor does it provide the outcome counts needed to state the number or proportion of positive results.

3. Trial Design

COMBACTE-CDI is described as a large, complex, multicentre, multicountry study focused on the burden of Clostridium difficile infection and current management practices. The registry lists the University of Leeds as sponsor and gives the status as completed. The listed enrollment is 3240. This enrollment figure does not, by itself, establish the number of samples with a reported primary endpoint result or the endpoint denominator.

The registry describes approaching hospitals and laboratories that carry out diagnostic testing of samples from in-patients and community patients, including Long-Term Care Facilities patients. On selected study dates, samples sent to the sites were to be tested at a central laboratory for CDI, regardless of the test requested at the site. The stated sample-receipt period was July through December 2018, at the time of a diarrheal episode. The record gives the study start as 2018-06-11 and primary completion as 2018-12-31.

After identification of cases and negative controls, the study description refers to a follow-up case-control study for collecting outcome and risk-factor information. The registry’s primary outcome is the number of participants with a positive test result, alongside the number with a negative result. The arms are named CDI Cases and CDI Negative Controls, but the registry explicitly indicates that this is not a randomized comparison and that allocation is not stated. The two arms are defined by test result rather than assigned study treatment.

The registry lists no intervention other than “No intervention.” It does not report the number or characteristics of participating sites, the detailed recruitment procedures, or the follow-up duration. It also does not report a protocol-level statistical analysis plan. The study description says that the work was based on a previous study model, but details beyond that description 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: Residual samples received during the study period were tested for Clostridioides difficile infection (CDI) at the coordinating laboratory. Informed consent is not required for the use of anonymised residual material. The number of CDI cases and negative controls are the number of participants with a sample that returned a positive or negative test result, respectively. Duplicate samples were excluded and therefore not completed. Test result was not reported back to the laboratories.

Overall Study

Analysis population for the primary outcome, as defined in the registry: The number of CDI cases and negative controls are the number of participants with a sample that returned a positive or negative test result, respectively. Therefore the "CDI negative controls" arms did not return a positive test result.

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: A clinical report form was requested 6 months after collection of residual diarrhoeal samples for all positive CDI cases and a proportion of CDI negative controls (case-control study analysis population). The number of baseline participants is different from the participant Flow module, and is the number of CDI cases with completed clinical report form and an appropriate number of CDI negative controls from the pool of samples received to obtain a ratio of approximately 1:4 (Cases:Controls)

6. Primary Endpoint

The registry lists the primary outcome as The Number of Cases of CDI, measured over Samples were received between July and December 2018 (at time of diarrheal episode) and test performed during that timeframe to identify the number of CDI cases (positive test result) and the number of CDI negative controls (negative test result)..

The primary endpoint, “The Number of Cases of CDI,” counts participants whose sample returned a positive CDI test result at the coordinating laboratory. The registry also defines the CDI negative controls as participants whose sample returned a negative result. The outcome is measured in participants and is a count, rather than a continuous measurement or a time-to-event outcome.

The registry says that the count is intended to indicate how many participants had a positive result among the overall total of participants. It also explains that the proportion with a positive result could be inferred by relating the CDI cases to the participants in both named groups. That interpretation requires the actual positive and negative counts. Those endpoint counts are not reported in the registry record, so the proportion cannot be stated from the posted results.

As a general statistical point, a positive-result count can describe the number of detected cases in the tested population. A proportion adds a denominator and can help describe the frequency of positive results in that population. Neither quantity alone establishes the rate in a wider population unless the sampling frame and representativeness support that interpretation. The registry’s stated aim of quantifying disease burden makes the endpoint relevant to case identification, but the available record does not provide enough endpoint detail to quantify that burden.

The record describes testing during the sample-receipt period and at the time of a diarrheal episode. It does not report the endpoint counts, a confidence interval, or a detailed definition of the testing algorithm beyond a positive or negative test result. It also does not report how samples without a final result were handled.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a binary classification at the sample-result level: a positive test result identifies a CDI case, while a negative result identifies a CDI negative control. The registry expresses the outcome as the number of participants in each result-defined category. This makes the primary endpoint a count of participants with an event, rather than a comparison of a continuous measurement or a time-to-event endpoint. The registry lists no registered analyses.

What does a count describe?

In general, a count reports how many observations meet a specified definition. Counts are easiest to interpret alongside the number of observations assessed, because a count alone does not show how common the event was in the relevant population. A proportion combines the event count with a denominator. Here, the registry describes an intended proportion based on positive results relative to participants in both result-defined groups, but the actual endpoint counts are not reported in the registry record. The proportion therefore cannot be calculated or interpreted from the posted results.

The listed enrollment is 3240, but the registry does not identify that enrollment as the number with a final primary endpoint result. It does not provide counts of samples tested, positive results, negative results, or samples without a result. In general, the denominator for an outcome must match the population to which that outcome applies. Using a general enrollment figure as a denominator without confirmation could misstate the frequency of positive results.

Why does the sampling process matter?

The study description says that samples sent to participating hospitals or laboratories on selected study dates were to be tested at a central laboratory, regardless of the test requested at the site. This approach is intended to look for cases that may otherwise be missed. In general, the population represented by a surveillance sample depends on which sites take part, which samples are included, and when they are collected. The registry describes hospitals and laboratories of interest, in-patient and community samples, and a sample-receipt period from July through December 2018. It does not report the participating site count, the full sampling schedule, or enough detail to establish how representative the tested samples were of a broader population.

Central testing can provide a common testing process across samples, but the record does not give the detailed laboratory methods, diagnostic performance characteristics or rules for resolving uncertain results. The primary endpoint is described using positive and negative test results. No further classification of test results is reported in the registry record.

What can a comparison establish?

The CDI Cases and CDI Negative Controls groups are defined by the outcome test result. They were not assigned as intervention and control arms, and the registry notes that this is not a randomized comparison. In general, comparisons between groups formed after observing an outcome can be useful for describing associations, but they do not have the same interpretation as comparisons created by random allocation. Differences in participant characteristics, sampling or follow-up could affect a comparison. The registry does not report an adjusted analysis or group-level outcome results for the planned follow-up study.

The study description refers to a follow-up case-control study to collect outcome and risk-factor data. It also mentions transmission models and cost-effectiveness work. These methods can address questions beyond counting positive tests, but their results depend on additional data and assumptions. No model specifications, risk-factor estimates, transmission-model outputs or cost-effectiveness results are reported in the registry record.

How should planning calculations be read?

In general, a sample-size calculation estimates how much information a future study might need to detect an effect under chosen assumptions. It is not an observed result and does not show the probability that the completed study found a real effect. The registry’s planning description says the illustration uses a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation, with the effect to detect expressed as a stated fraction of the observed effect. It also distinguishes participants per arm for ordinary outcomes from total events for time-to-event calculations. These are planning assumptions, not a statement of the study’s original design.

The registry does not report a randomized allocation, and its groups are defined by test result. The equal-allocation assumption used for a planning illustration should not be read as evidence that the observed groups were equal in size. Nor should any such planning estimate be described as observed power. The record does not report the observed endpoint effect needed to contextualize the planning illustration, because the primary outcome counts are absent.

In general, a confidence interval describes the range of values compatible with the observed data under a statistical model and conveys uncertainty beyond a point estimate. The width of such an interval depends on the amount of information and the analysis method. No confidence interval for the primary endpoint is reported, and the absent endpoint counts prevent an independent calculation. The record also gives no inferential analysis or model with which to assess uncertainty.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("CDI Cases", "CDI Negative Controls")
n      <- c(103, 3050)
events <- c(103, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  CDI Cases  103  103  100.0  96.5  100.0  96.4  100.0  96.5  100.0
ARM  CDI Negative Controls  3050  0  0.0  0.0  0.1  0.0  0.1  0.0  0.1

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome measure and reports zero analyses. It does not report a multiplicity strategy, such as a prespecified approach to controlling false-positive findings across multiple statistical tests. Because no inferential results are reported, the record does not support an assessment of how multiple comparisons were handled. The absence of a reported strategy should not be interpreted as evidence that one was or was not used outside the registry entry.

Missing-data handling is not reported in the registry record. The entry does not state how samples without a final test result, unavailable follow-up information, or incomplete participant records were addressed. It also does not report the number of missing observations for the primary endpoint or follow-up measures. In general, missingness can affect an estimate when the observations that are absent differ systematically from those observed; evaluating that possibility requires information about the amount and pattern of missing data and the analysis approach.

The record does not define analysis populations beyond describing the positive-result and negative-result categories. It does not report whether the primary count included all enrolled participants, only participants with tested samples, or another defined set. The listed enrollment of 3240 cannot be assumed to be the primary endpoint denominator. The record also does not report a separate population for the follow-up case-control study or explain how participants entered that component.

No statistical analysis plan, adjusted analysis, sensitivity analysis or multiplicity-adjusted result is reported in the registry record. These absences limit assessment of prespecification and robustness. The registry’s description of intended study aims does not substitute for results or a documented analysis strategy.

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

The registry record establishes that COMBACTE-CDI was designed to examine the burden of CDI, detect cases through central testing of samples, and describe management practices. It defines a positive test result as a CDI case and a negative test result as a CDI negative control. It also describes follow-up data collection on outcomes and risk factors. These details explain what the study set out to measure, but they do not state how many positive results were observed.

Without the primary endpoint counts and a confirmed denominator, the posted information does not quantify the proportion of tested participants with CDI. It does not establish the frequency of CDI in all hospitals, communities, or the wider region. Generalizing from a tested sample to a broader population depends on how samples and sites were selected and on the population represented. The registry describes the sampling concept but does not provide the details or results needed to assess that generalizability.

The result-defined groups are not randomized treatment arms. The record lists no intervention and notes that the arms are not a like-for-like control comparison. Therefore, the group labels do not support a causal claim about treatment or a claim that one group’s outcome was caused by its CDI test result. The follow-up outcomes and risk-factor findings are not reported.

The adverse-event entries list 14 deaths among CDI Cases and 24 among CDI Negative Controls. These are counts without reported denominators, timing, causes or a comparative analysis. They do not establish a difference in mortality or explain why deaths occurred. The registry also lists no serious adverse-event terms.

More broadly, the registry entry is informative about study purpose and outcome definitions, but it does not present the primary results or analyses. It cannot establish the number of missed cases, the level of disease burden, the performance of a diagnostic approach, the effectiveness of any intervention, or the results of planned transmission and cost-effectiveness work. Those findings are not reported in the registry record.

13. Why This Trial Matters Statistically

This study addresses a statistical challenge in case detection: counts based on routine diagnostic processes may not capture every case if some samples are not tested or cases are not recognized. The registry describes testing samples centrally regardless of the test originally requested at the site, with the aim of looking for missed CDI cases. That design makes clear why the definition of the sampled population and the denominator matters when interpreting a count of positive results.

The study also illustrates the distinction between a descriptive surveillance outcome and a randomized treatment effect. The primary endpoint counts positive test results, while the named comparison groups are formed by those test results. They are not randomized arms. Reading the result labels as if they were assigned treatment groups would change the question and invite an unsupported causal interpretation.

Another statistical issue is the relationship between an overall enrollment figure and the number actually contributing to an endpoint. The record lists enrollment of 3240, but does not report the positive and negative counts or establish an endpoint denominator. Clear reporting of tested samples, final result status and analysis populations is important for interpreting a count and any proportion based on it.

The registry also mentions follow-up, risk-factor work, transmission models and cost-effectiveness analyses. Each involves questions and assumptions beyond the primary count. Since their results and methods are not reported in the registry record, their planned role should not be confused with evidence already available. In statistical reporting, transparent separation of aims, observed results and analyses helps readers identify what conclusions the record can support.

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.

The main limitation for interpretation is that the registry record does not report the primary endpoint counts. It describes what constitutes a CDI case and a negative control, but provides neither the number of positive results nor the number of negative results. It also does not report the number of samples without a final result. Consequently, the primary outcome and the proportion of positive results cannot be quantified from the posted record.

The listed enrollment of 3240 does not resolve that gap. The record does not establish that all enrolled participants had a sample included in the primary endpoint or that enrollment is the appropriate denominator for the stated outcome. The number of participating sites and the details of sample selection are also not reported. These omissions prevent assessment of how closely the tested sample reflects the population the study intended to characterize.

The groups are defined by laboratory result, and the registry says the arms are not a like-for-like control comparison. Allocation is not stated. The record therefore does not support a randomized comparison or a causal interpretation of differences between the groups. It does not report detailed follow-up results, adjustment for participant characteristics, or the statistical methods used for the planned case-control and modelling work.

Deaths are listed as 14 for CDI Cases and 24 for CDI Negative Controls, but the registry does not report the relevant denominators, timing, causes or analysis. These entries cannot be interpreted as comparative risks. The registry reports zero analyses and does not report methods for missing data or multiplicity. It also does not report a statistical analysis plan or confidence intervals for the primary outcome.

Finally, the registry’s stated aims include transmission modelling and cost-effectiveness work, but no findings from those activities are reported. These limitations concern what can be concluded from the registry record; they do not establish whether additional information exists elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: COMBACTE-CDI Understanding the Burden of C. Difficile Infection — NCT03503474.
• Registry record last updated 2022-02-09; status last verified 2022-01.
• Results first posted 2022-02-09.
• 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.

18. Record Summary

COMBACTE-CDI is a completed study sponsored by the University of Leeds, with listed enrollment of 3240. Its stated purpose was to quantify the burden of CDI and describe current management practices. The registry describes central testing of samples from in-patient and community settings, including Long-Term Care Facilities patients, and follow-up case-control data collection on outcomes and risk factors.

The primary outcome counts participants with positive CDI test results and identifies participants with negative results as controls. The sample-receipt period is described as July through December 2018. The record lists no intervention, states that the result-defined groups are not a randomized comparison, and reports zero analyses.

The registry does not report the primary endpoint counts, the endpoint denominator, or results of the planned follow-up and modelling work. It lists 14 deaths in CDI Cases and 24 in CDI Negative Controls without details that support a comparative interpretation. The record therefore describes the study’s aims and outcome definitions but does not quantify the reported burden of CDI.