Audit-and-feedback to Improve Antimicrobial-prescribing Among Urologists: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Audit-and-feedback to Improve Antimicrobial-prescribing Among Urologists (NCT04196777), a na na study. The primary outcome is Percentage of Cases Who Received Excessive Post-procedural Antimicrobials.

2. Clinical Question

The study addresses prescribing after common urologic procedures in Veterans Health Administration hospitals. The registry background describes opportunities to reduce unnecessary antimicrobial use and notes that guidance recommends antibiotics for no more than 24-hours after most urologic procedures. It also cites a national cohort in which excessive post-procedural antimicrobials were common. Those background statements motivate the intervention; they are not results of this pilot study.

The intervention was feedback to providers at selected sites. The registry says three intervention sites would be randomly selected from sites in the top quartile for frequency of excessive prescribing after the procedures of interest. This describes site selection, not random assignment of participants or sites to intervention and control groups. The primary question is therefore whether the recorded prescribing outcome changed in the intervention setting, not whether a randomized comparison established a treatment effect.

The registry record defines the outcome and identifies the study’s intended focus, but does not report the primary outcome counts or percentages. Consequently, the clinical question can be stated from the registered aims, while the size and precision of any observed change cannot be assessed from the posted results described in the record.

3. Trial Design

This is a completed pilot intervention study of audit-and-feedback, listed as a behavioral intervention. The registry classifies the model as single-group, allocation as not applicable, and masking as none. The study enrolled 525 participants. Its registered start date is 2022-08-01, its primary completion date is 2023-07-31, and its status is completed.

The intervention description says that three sites would be selected from the top quartile of Veterans Health Administration sites, ranked by frequency of excessive post-procedural antimicrobial use after the procedures of interest. Providers at those intervention sites were to receive feedback at baseline and at regular intervals. The record does not describe a concurrent randomized control group. It lists a Baseline group as well as an Intervention group in the adverse-event information, but does not establish that these groups are exchangeable or directly comparable.

The registry’s enrollment figure and the group sizes appearing in the serious adverse event information differ. The record does not explain the relationship between those counts, how the baseline cohort was selected, or whether the same eligibility and follow-up procedures applied across groups. It also does not report the number of sites contributing to each result beyond the intervention-site description.

The study is identified as a pilot, and its phase is listed as not applicable. The registry record does not report a detailed protocol, a sample-size justification for the study itself, or a prespecified statistical analysis plan. The record’s planning assumptions for sample-size illustrations are not evidence that the trial was designed or powered according to those assumptions. The single-group structure is central to interpretation: any contrast between intervention and baseline observations is vulnerable to changes over time and other differences that the registry does not describe.

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: All patients who underwent a qualifying urologic procedure during 8/1/2022-7/31/2023 at a participating site were included in the intervention phase. These procedures were ureteroscopy (URS), transurethral resection of the prostate (TURP), and transurethral resection of a bladder tumor (TURBT). Outcomes in these enrolled patients were compared to outcomes in patients who underwent these qualifying procedures during a baseline period (n=1272); the baseline period spanned from 7/1/2020-6/30/2022.

Overall Study

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.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Cases Who Received Excessive Post-procedural Antimicrobials, measured over Within 1 day of the urologic procedure.

The primary endpoint is the percentage of cases who received excessive post-procedural antimicrobials. The registry defines excessive use for this measure as a prescription for an antimicrobial agent on post-procedural day one. The numerator is the number of patients who received such a prescription, and the denominator is all patients at the site who underwent a qualifying urologic procedure. Its timeframe is within one day of the urologic procedure.

This endpoint focuses on prescribing behavior that the intervention is intended to address. Expressing the outcome as a percentage places the number of affected cases in relation to the eligible procedure population, rather than presenting an event count alone. It can therefore describe how common the specified prescribing practice was among qualifying cases at a site or during a defined observation period.

The definition is specific to a prescription on post-procedural day one. It should not be conflated with the separate secondary measure of mean excessive post-procedural antimicrobial duration, or with late prescriptions and clinical outcomes such as readmission, emergency or urgent care presentation, death, or Clostridioides difficile outcomes. Those are separately listed outcomes.

The registry record identifies the endpoint and its calculation, but does not report its group-specific counts, percentages, or uncertainty estimates. Therefore, the registered definition explains what would be measured; it does not establish the observed frequency of excessive prescribing or the magnitude of any change after feedback.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation depends on distinguishing the outcome definition, the study structure, and the quantities actually reported. The primary endpoint is a binary case-level outcome: a qualifying patient either received an antimicrobial prescription on post-procedural day one or did not. The registry defines the percentage using affected cases as the numerator and all qualifying procedure patients at the site as the denominator. The resulting percentage is a measure of frequency, not by itself a causal estimate.

What does a confidence interval describe?

General statistical explanation: a confidence interval expresses uncertainty around an estimated quantity under the assumptions of the analysis. A narrow interval indicates greater precision than a wide interval, but neither interval width nor the estimate’s proximity to a threshold establishes clinical importance on its own. The registry record does not report confidence intervals for the primary endpoint or the serious adverse event counts. Precision for those results therefore cannot be evaluated from the record.

Why does the single-group design matter?

General statistical explanation: a before-and-after or baseline-versus-intervention comparison can reflect both the intervention and changes that occur over time. Without a randomized concurrent comparison, changes in patient mix, prescribing practice, staffing, documentation, or other aspects of care may influence the observed outcome. The registry identifies the study as single-group and does not describe a randomized control comparison. It also does not report an adjustment strategy that would separate these possible influences from the effect of feedback.

The description says intervention sites were selected from the top quartile of sites ranked by excessive prescribing frequency. Selecting sites with high recorded frequencies can make interpretation particularly dependent on how later measurements are compared with baseline. General statistical explanation: when a group is selected because its initial measurement is high, later observations may differ partly because of ordinary variation or regression toward a more typical level. The record does not report whether or how this issue was addressed, so it cannot be assumed that a change would be attributable to feedback.

What does over-dispersion mean for event counts?

General statistical explanation: over-dispersion occurs when event counts vary more than a simple statistical model expects. If it is ignored, uncertainty can be understated and statistical tests can appear more decisive than warranted. The registry identifies the primary outcome as a count of participants with an event and lists a binary outcome kind, but does not report a model for clustering or over-dispersion. The study involves sites, so methods for accounting for site-level structure would be relevant to a detailed analysis; whether such methods were used is not reported in the registry record.

How should sample-size illustrations be read?

The registry’s sample-size assumptions describe planning illustrations based on an observed effect, not the trial’s own design. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect set as a stated fraction of the observed effect. The record says that non-time-to-event illustrations are expressed as participants per arm, while time-to-event illustrations are expressed as total events. These assumptions are not a report of the study’s actual enrollment rationale, and the resulting sample sizes are not observed power. The record does not report a trial-specific sample-size justification.

General statistical explanation: power is a design-stage probability tied to assumptions about effect size, variability, analysis, and sample size. Retrospective calculations based on an observed effect cannot show that a study had adequate power or that an effect is real. Planning illustrations can instead show how sample-size requirements would vary under stated assumptions. No numerical illustration values are reported in the registry record, and the primary endpoint estimate needed to assess its precision is also absent.

What can the event counts establish?

The serious adverse event figures are counts of affected participants and participants at risk, with deaths also listed. Counts provide a description of recorded events, but comparison requires clarity about follow-up, outcome definitions, overlap between categories, and cohort comparability. The registry record does not provide all of that context. General statistical explanation: two event counts from groups of different sizes should not be compared as though the group sizes were the same. A formal comparison would need appropriate denominators and a method that reflects the study design and uncertainty. No such comparison is reported in the registry record.

Finally, statistical significance and practical importance are distinct. A small difference can be estimated precisely in some settings, while a potentially important difference can remain uncertain in a small or variable dataset. Since primary estimates and uncertainty measures are not reported, neither question can be evaluated for the endpoint of interest.

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("Intervention", "Baseline Period")
n      <- c(525, 1272)
events <- c(216, 644)
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  Intervention  525  216  41.1  36.9  45.5  37.0  45.4  37.0  45.4
ARM  Baseline Period  1272  644  50.6  47.8  53.4  47.9  53.4  47.9  53.4

11. Multiplicity, Missing Data and Analysis Populations

The registry lists seven outcome measures and eight analyses. It names the primary endpoint and several secondary outcomes, but does not report how the analyses were organized, which analysis corresponded to each outcome, or whether any outcomes were tested more than once. A multiplicity adjustment or a hierarchy for interpreting multiple tests is not reported in the registry record. Without that information, the risk of chance findings across multiple analyses cannot be assessed.

The primary endpoint definition describes a numerator and denominator, but the registry record does not report the outcome counts, percentages, or analysis population used for the posted primary result. It also does not describe missing outcome data, exclusions from analysis, or how incomplete records were handled. The number of participants enrolled does not, by itself, establish how many contributed to each endpoint.

The record does not state whether analyses were conducted according to an intention-to-treat principle, a modified intention-to-treat approach, or a per-protocol population. Because the study is described as single-group and its intervention is delivered at sites, the record would also need to clarify which units and time periods contributed to each analysis to support a full interpretation. Those details are not reported in the registry record.

General statistical explanation: missing-data methods and analysis-population definitions matter because exclusions or incomplete follow-up can change who is represented in an estimate. If missingness is related to the outcome or to intervention exposure, a complete-case analysis may not represent all eligible participants. No assumptions about the amount, pattern, or treatment of missing data can be made from the record.

The absence of these details does not establish that the analyses were inappropriate; it means the registry record does not allow readers to evaluate multiplicity, missingness, or the population underlying the endpoint estimates.

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

Serious events with the highest arm-level rate: emergency department, urgent care clinic, or hospital re-admission within 30 days of the procedure (21.5%).

The registry lists serious adverse event information for an Intervention group and a Baseline group. The Intervention group has 525 participants at risk and 113 affected; the Baseline group has 1272 participants at risk and 186 affected. The record also lists 4 deaths in each group. These figures describe what the registry records for the named groups; they do not establish that the intervention caused or prevented any event.

The registry identifies Emergency Department, Urgent Care Clinic, or Hospital Re-admission within 30 days of the procedure as the top serious-event term and lists 21.5%. It does not give further detail about the denominator or the way this term relates to the affected-participant counts. The term should therefore be read as a recorded event category, not assumed to be a separate, mutually exclusive outcome.

The groups should not be treated as a randomized safety comparison. The study is listed as single-group, and the record does not explain the relationship between baseline and intervention cohorts, their follow-up, or their comparability. It also does not report a statistical comparison or uncertainty estimate for the serious adverse events. The counts alone cannot determine whether event risk differed between groups or whether any difference was related to audit-and-feedback.

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

The registry record describes a pilot study intended to assess whether audit-and-feedback could reduce excessive antimicrobial prescribing after urologic procedures. It specifies the primary outcome as a prescription on post-procedural day one and describes feedback at baseline and regular intervals to intervention sites. These details establish the study question and the outcome definition.

The primary endpoint results are not reported in the registry record. There is no primary outcome count or percentage with which to assess the level of excessive prescribing, the change over time, or the precision of an estimate. The named secondary outcomes likewise lack reported results in the record. As a result, the available information does not establish whether prescribing changed after feedback.

The serious adverse event figures provide descriptive counts for the Intervention and Baseline groups. They should not be interpreted as evidence that feedback improved or worsened safety. The registry does not report random assignment, a like-for-like control comparison, or analyses accounting for differences between the groups. It also does not provide uncertainty estimates for these figures.

Even if a change in prescribing were reported elsewhere, a single-group design would limit causal interpretation. A change could reflect the intervention, but could also reflect other changes over time or differences in the patients and procedures observed. The registry record does not report methods that would distinguish these explanations.

The study therefore contributes a registered intervention question and specified outcome framework, but the record does not provide enough primary endpoint results to judge the observed change. It does not establish that the intervention did or did not reduce prescribing, and it does not establish a safety effect. Those conclusions would require results and analysis details not reported in the registry record.

14. Why This Trial Matters Statistically

This study is statistically instructive because its central question concerns a measurable prescribing behavior, while its design is single-group. The endpoint has a clear operational definition: a prescription for an antimicrobial agent on post-procedural day one. Clear definitions are important because they determine which cases count as events and which cases contribute to the denominator. They also make it possible to distinguish the primary prescribing measure from other listed outcomes, such as antimicrobial duration or acute-care use.

The site-based intervention also raises questions about the level at which data are generated and analyzed. Feedback is delivered to providers at sites, while the primary outcome is defined among patients undergoing procedures. General statistical explanation: when an intervention operates at a group level but outcomes are measured at an individual level, observations within the same group may be more alike than observations from different groups. Analyses that treat all observations as independent can understate uncertainty if they do not account for that structure. The registry record does not describe whether or how site-level clustering was handled.

Site selection is another important feature. Intervention sites were to be chosen from the top quartile for excessive prescribing frequency. That choice may make the intervention relevant to sites with high measured use, but it complicates interpretation of later measurements if baseline values are unusually high. A transparent analysis would need to show baseline and follow-up outcomes and describe how natural variation and time trends were considered. Those primary results and methods are not reported in the registry record.

Finally, the registered outcome framework includes prescribing measures and patient outcomes. Reporting both can help readers distinguish a change in a targeted process from changes in clinical events. However, multiple outcomes and analyses require clear reporting of which tests were planned and how results should be interpreted. The record lists seven outcome measures and eight analyses but does not report a multiplicity approach or the primary endpoint results. These omissions limit statistical interpretation, not the importance of the question being studied.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most direct limitation of the registry record is that it does not report the primary endpoint results. Without the intervention and baseline counts or percentages, readers cannot assess the observed frequency of excessive prescriptions, any change, or the precision of the estimate. Results for the listed secondary outcomes are also not reported in the record.

The study is classified as single-group, with allocation not applicable and masking none. The record does not describe a randomized comparison, and the Baseline group is not documented as a like-for-like control group. Differences between groups or changes over time may therefore complicate attribution of any observed change to feedback. The registry does not report an adjustment strategy for these concerns.

The intervention is described at the site level, but the record does not report how many sites contributed to each analysis, how patient observations were clustered, or whether analysis methods accounted for site-level dependence. The intervention sites were selected from the top quartile of prescribing frequency, but the record does not describe how selection based on high frequency was handled analytically.

The registry record does not report a trial-specific sample-size justification, detailed statistical analysis plan, missing-data methods, analysis populations, or multiplicity approach. It also does not fully explain the relationship between the enrollment figure and the different at-risk group sizes listed for adverse events. Follow-up details and the denominator for the reported 21.5% serious-event term are not reported in the registry record.

These limitations concern what can be evaluated from the registry record; they do not establish that unreported analyses or procedures were absent from the full study. They do mean that the intervention’s effect and the precision of the primary outcome cannot be determined from the recorded information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Audit-and-feedback to Improve Antimicrobial-prescribing Among Urologists — NCT04196777.
• Registry record last updated 2024-09-19; status last verified 2024-08.
• Results first posted 2024-09-19.
• 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

The registry describes a completed pilot study of audit-and-feedback for excessive post-procedural antimicrobial prescribing among patients undergoing qualifying urologic procedures at Veterans Health Administration hospitals. The study is listed as single-group, with no masking and allocation not applicable. The registered enrollment is 525, and the intervention description specifies feedback at baseline and regular intervals to selected intervention sites.

The primary outcome is the percentage of cases receiving a prescription on post-procedural day one. The registry record defines its numerator and denominator but does not report primary endpoint results. Several secondary outcomes are named, but their results are not reported in the record.

Serious adverse event information lists 113 affected among 525 at risk in the Intervention group and 186 affected among 1272 at risk in the Baseline group, with 4 deaths in each group. These are descriptive figures, not a randomized comparison. The record does not report the analysis methods, uncertainty estimates, missing-data handling, multiplicity approach, or a trial-specific sample-size justification needed for a fuller statistical assessment.