Paediatric Rapid Sepsis Trigger (PRST) Tool: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Paediatric Rapid Sepsis Trigger (PRST) Tool (NCT04304235), a non-randomized na study. The primary outcome is Time to Administration of an Appropriate Antimicrobial.

2. Clinical Question

The registry’s stated rationale is that a rapid trigger tool could help identify children at high risk and reduce delays in treatment. That is the rationale for evaluating the tool, not evidence of an effect. The primary outcome focuses on treatment timing rather than whether the tool changed diagnosis, clinical outcomes, or mortality. Other outcomes are registered, but their detailed results are not reported in the registry record.

The comparison is not randomized. The registry explicitly warns that the arms are not a like-for-like control comparison. Thus, the central question is descriptive as well as comparative: what treatment times were recorded in the groups and settings, and how did those records differ? Without reported endpoint summaries and a clearly described analysis, the registry record does not establish the size, precision, or cause of any difference.

3. Trial Design

The registry lists this study as completed, with nonrandomized allocation, no masking, and a sequential model. It gives an enrollment of 18693. The recorded start date is 2020-04-27 and the primary completion date is 2023-07-01. The phase is listed as NA, and the sponsor is the University of British Columbia.

The study includes intervention and control groups in Kenya, as well as intervention groups in Uganda. At baseline, nurses in the listed Kenya intervention and control groups and the Uganda intervention group used ETAT guidelines. During implementation, nurses in the intervention groups used the digital triage tool, while nurses in the listed control implementation groups continued to use ETAT. A Uganda control baseline group is not listed in the registry descriptions.

Because allocation is nonrandomized, group membership cannot be treated as if it arose from random assignment. The registry also identifies the arms as not being a like-for-like control comparison. General statistical context: nonrandomized comparisons can reflect differences between groups or settings as well as an intervention, and sequential evaluations can be affected by changes over time. Whether and how those concerns were addressed is not reported in the registry record.

Masking is listed as none. The registry does not report a detailed sequence of implementation, the dates for each group or phase, or the criteria used to assign participants to groups. It also does not report the statistical model, covariate adjustments, or the rationale for the enrollment target. The study should therefore be read through the information recorded for its design and results, without inferring unreported protocol details.

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: Non infectious illness, routine clinic visit

Only the first 6 of 7 groups are shown.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Children presenting to the outpatient department at facility

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: same as Participant Flow information

Only the first 6 of 8 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Time to Administration of an Appropriate Antimicrobial, measured over 1 day.

The primary endpoint is time, in minutes, from a child’s first registration on arrival at the facility until the start of administration of an appropriate antimicrobial. The registry defines an appropriate antimicrobial as including at least one antibiotic or antimalarial, with treatment determined and administered by hospital staff. The timeframe is one day, and the stated population is children presenting to the outpatient department at a facility.

The registry specifies the mean as the summary parameter and the inter-quartile range as the measure of dispersion. The mean describes the average recorded time; the inter-quartile range describes the spread of the middle portion of observations. These summaries answer different questions, so considering both can help readers understand not only the average time but also variability in treatment timing.

Time to treatment is used because the evaluation concerns whether triage with a rapid trigger tool is associated with earlier administration of an appropriate antimicrobial. It is a process measure: it records timing in the care pathway, rather than directly measuring a clinical outcome such as recovery or survival. The endpoint definition does not by itself establish that a shorter time occurred or that any change in timing caused a change in health outcomes. Group-level endpoint values and comparative estimates are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the endpoint summary represent?

The primary endpoint is a duration measured in minutes. The registry specifies a mean and an inter-quartile range. In general, the mean is calculated from the recorded observations and represents their arithmetic average. The inter-quartile range describes the spread between the lower and upper quartiles, giving information about the middle of the distribution rather than its extremes. These two summaries complement one another: the mean gives an average, while the inter-quartile range gives a sense of how concentrated or dispersed the central observations are.

Time measurements can have distributions that are asymmetric or include unusually long delays. General statistical context: extreme observations can influence a mean, while the inter-quartile range is less influenced by extremes. That is one reason to consider both summaries. The registry does not report the actual mean or inter-quartile range for any group, nor does it report a distribution plot or the handling of unusually long or incomplete times.

What would a group comparison require?

A comparison could describe how treatment times differ between groups, but the meaning of that contrast depends on the design and analysis. Here, allocation is recorded as nonrandomized, and the registry warns that the groups are not a like-for-like control comparison. General statistical context: random assignment can help balance known and unknown characteristics on average, whereas nonrandomized groups may differ before treatment. Statistical adjustment can account for measured factors under specified assumptions, but cannot automatically remove bias from unmeasured differences.

The group descriptions include baseline and implementation observations in Kenya and Uganda, with the intervention changing from ETAT to the digital PRST tool for the intervention groups during implementation. Control implementation groups are described as continuing ETAT. The registry labels the model sequential, but does not explain the analysis structure or how phase and site were represented. It does not report whether the primary estimate was an intervention-versus-control contrast, a baseline-versus-implementation contrast, or another comparison.

In general, comparisons across sequential phases can be affected by calendar-time changes, and comparisons across settings can be affected by differences in context. A statistical model may account for such structure if its specification and assumptions are appropriate. No model specification, adjustment variables, or estimated effects are reported in the registry record. These are important omissions for interpreting the causal meaning and precision of a contrast.

How should uncertainty be assessed?

General statistical context: an estimate describes the observed contrast, while a confidence interval communicates uncertainty around that estimate under the analysis assumptions. A narrow interval generally indicates more precision than a wide interval, though precision does not establish that a comparison is unbiased. A p-value addresses compatibility with a specified null hypothesis under a statistical model; it does not measure the size or importance of an effect. No primary endpoint estimate, confidence interval, or p-value is reported in the registry record, so neither the estimated contrast nor its uncertainty can be evaluated.

The registry lists the mean and inter-quartile range, but does not state how uncertainty around the mean was calculated. It does not report whether clustering by facility or repeated observations over time were considered. General statistical context: when observations are correlated within a site or period, treating them as independent can make uncertainty estimates too small. Whether this issue applies to the analysis and how it was addressed are not reported in the registry record.

What do the planning assumptions mean?

The record describes sample-size planning illustrations using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumed effect to detect is a stated fraction of the observed effect. General statistical context: such calculations explore how much information might be needed under chosen assumptions; they do not describe the precision actually achieved in a completed study. The record explicitly distinguishes planning illustrations from the trial’s own design. They should not be interpreted as observed power or as proof that the study was adequately powered for its actual analyses.

The registry says time-to-event illustrations use the total number of events, while other illustrations use participants per arm. The primary endpoint itself is a time duration, but it is specified as a mean rather than a time-to-event parameter. No computed planning values or trial-specific sample-size justification are reported in the registry record. The statistical analysis plan, including model choice and handling of missing endpoint data, is also not reported.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists seven outcome measures and one analysis. It identifies time to administration of an appropriate antimicrobial as the primary outcome. The details of the other outcome measures and the analysis are not reported in the registry record. No multiplicity adjustment or prespecified hierarchy for testing outcomes is reported.

Multiplicity arises when several outcomes, comparisons, or statistical tests are considered. General statistical context: the more tests that are performed, the greater the chance that at least one may appear noteworthy by chance if each is interpreted separately. Methods such as adjusting significance thresholds or specifying a primary testing hierarchy can address this concern, but the registry record does not state whether any such method was used. The listed count of analyses alone does not explain which comparisons or tests were included.

Missing-data methods are not reported in the registry record. It is not stated how missing registration times, antimicrobial administration times, or other outcome information were handled, or whether any observations were excluded. General statistical context: missing data can affect both the estimate and its precision, particularly if the likelihood of missingness is related to the outcome or group. The extent and pattern of missingness are not reported, so their possible influence cannot be assessed.

The analysis populations are also not reported in the registry record. It is therefore unclear which participants contributed to each outcome analysis, whether any eligibility rules were applied after enrollment, or how observations were assigned to analysis groups. These details matter for understanding the denominator behind an estimate and whether the analysed population matches the intended study population. No group-level primary endpoint summaries are reported, preventing assessment of the analysis population through the outcome results.

11. 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.

The adverse-event record lists zero serious events affected and zero other events affected in each listed group. It also records the number at risk for those categories and separate death counts. The listed deaths are 29 for Kenya Intervention Baseline, 10 for Kenya Intervention Implementation, 11 for Kenya Control Baseline, 10 for Kenya Control Implementation, 11 for Uganda Intervention Baseline, 4 for Uganda Intervention Implementation, and 2 for Uganda Control Implementation. These are registry entries, not a causal comparison.

Deaths are reported as a separate field from serious and other adverse events. The registry record does not establish whether any listed death was classified as an adverse event, nor does it describe the timing, cause, or relationship of deaths to the intervention. It also does not report serious adverse-event terms; no top serious terms are listed. The zero affected values should therefore be read as the event counts recorded in those fields, not as proof that no adverse experiences of any kind occurred.

General statistical context: safety counts need their definitions, observation periods, and ascertainment procedures to be interpreted. Those details are not reported in the registry record. Because allocation was nonrandomized and the groups are not described as like-for-like, the group counts alone cannot support a treatment-related safety conclusion.

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

The registry record establishes that a completed, nonrandomized, sequential evaluation of the PRST tool was registered, and that its primary outcome was time from first facility registration to the start of an appropriate antimicrobial. It describes use of ETAT for baseline and control implementation groups, and use of the digital tool for intervention implementation groups in Kenya and Uganda. It also reports an enrollment of 18693.

The record does not report the primary outcome values or an estimated comparison. It therefore does not show whether the intervention groups had shorter or longer treatment times, how large any contrast was, or how precise an estimate might be. It does not report a confidence interval or p-value for the primary endpoint. The completion status is administrative study information, not an outcome finding.

The stated aim was to support rapid identification and timely treatment. The registry’s rationale is not itself evidence that the tool improved recognition, shortened delays, reduced organ damage, or changed mortality. The listed primary endpoint measures a process in care; it does not directly establish patient benefit. The record’s death counts also do not establish a causal effect on mortality.

General statistical context: a nonrandomized comparison can be affected by differences between groups and by changes over time. Since the registry explicitly warns that the arms are not a like-for-like control comparison, any unadjusted difference should not automatically be attributed to the PRST tool. The analysis approach and adjustments are not reported in the registry record, so the strength of any causal interpretation cannot be assessed.

Accordingly, the registry record is informative about the study question, implementation structure, and endpoint definition, but it does not provide enough outcome detail to determine what happened to treatment timing or to quantify the uncertainty of a result. No conclusion that the tool does or does not work follows from the information recorded.

13. Why This Trial Matters Statistically

This study addresses a practical measurement question: whether a triage tool can be evaluated through the timing of treatment in routine facility care. Its primary endpoint has a clear operational anchor, beginning at first registration and ending when administration of an appropriate antimicrobial starts. Defining both points is important because a time measure depends on consistent identification of its start and finish.

Statistically, the record also illustrates why design labels matter. The registry classifies allocation as nonrandomized and warns that the arms are not a like-for-like control comparison. A numerical contrast, even if reported, would need to be interpreted in light of group composition, setting, and sequence. Those considerations are central to distinguishing an observed association from an intervention effect.

The evaluation spans implementation and baseline groups in Kenya and Uganda, but the registry does not provide the results or analysis details needed to assess cross-setting consistency, temporal changes, or uncertainty. This makes transparent reporting of group-specific summaries, comparison methods, missing data, and analysis populations especially important. Without them, readers cannot judge whether a difference is large, precise, or robust to plausible sources of bias.

The registry’s primary outcome is a treatment-process measure. That can be relevant to the rationale for a triage intervention, but it is not interchangeable with diagnostic performance or patient outcomes. Statistical interpretation depends on keeping the measured endpoint separate from the broader aims described in the study rationale. The record highlights the need for outcome reporting that connects a defined operational measure to a clearly specified analysis, while avoiding claims beyond what the data establish.

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 statistical interpretation is that group-level values for the primary endpoint are not reported in the registry record. There is no reported mean, inter-quartile range, comparison estimate, confidence interval, or p-value. Without those results, the observed treatment timing and its precision cannot be evaluated.

The design is nonrandomized, and the registry cautions that the arms are not a like-for-like control comparison. The record does not report how participants entered groups or whether analyses accounted for differences among groups, settings, or phases. Although the model is classified as sequential, the analysis structure and timing of implementation are not described in detail.

The registry does not report the statistical model, covariates, clustering approach, missing-data handling, or analysis populations. It lists seven outcome measures and one analysis, but does not report a multiplicity strategy or the detailed results for other outcomes. The registry does not provide a sample-size justification for the study; its stated planning assumptions are illustrations based on an observed effect and are not observed power.

The safety fields list zero serious and other events affected in the groups, alongside separate death counts, but event definitions, ascertainment procedures, and relationships to the intervention are not reported. The listed counts therefore cannot establish comparative safety. More generally, the registry information does not support conclusions about diagnostic accuracy, clinical benefit, or the cause of any group differences.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Paediatric Rapid Sepsis Trigger (PRST) Tool — NCT04304235.
• Registry record last updated 2025-05-13; status last verified 2025-04.
• Results first posted 2025-05-13.
• 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

NCT04304235 is the registry record for the Paediatric Rapid Sepsis Trigger (PRST) Tool study, sponsored by the University of British Columbia. It is marked completed, with nonrandomized allocation, no masking, a sequential model, and enrollment of 18693. The recorded start date is 2020-04-27, and the primary completion date is 2023-07-01.

The primary outcome is mean time in minutes from first facility registration to the start of an appropriate antimicrobial, with inter-quartile range as the dispersion measure and a timeframe of one day. The registry describes ETAT use in baseline and control implementation groups, and PRST use in intervention implementation groups in Kenya and Uganda. It cautions that the arms are not a like-for-like control comparison.

Although the record lists seven outcome measures, primary endpoint results and the statistical analysis details are not reported in the registry record. The design and endpoint can be described, but the magnitude, precision, and causal meaning of any outcome difference cannot be assessed from the registry information.