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 Pediatric Guideline Adherence and Outcomes- Argentina (NCT03896789), a randomized na study. The primary outcome is First 3-day Cumulative Intensive Care Unit (ICU) Traumatic Brain Injury (TBI) Guideline Adherence Rate, Overall.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03896789 |
| Phase | NA |
| Condition | TBI (Traumatic Brain Injury) |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 505 |
| Arms in the primary outcome | n/a |
| Lead sponsor | University of Washington |
| Results posted | Yes |
2. Clinical Question
Population
Children receiving intensive care for severe traumatic brain injury (TBI) in the South American clinical care context described in the registry record. The primary endpoint covers the first 3 days of ICU care, or the period until death or discharge if either occurred sooner. The record does not report further eligibility details, such as age limits or specific exclusion criteria.
Intervention
The PEGASUS Program for Care: training for sites assigned to the intervention arm, followed by delivery of the program along with usual care after randomization. The record describes the program as intended to increase adherence to pediatric TBI guidelines.
Comparator
Usual care after randomization at control sites. Those sites collected usual-care data during the baseline and post-randomization periods and were offered PEGASUS training at the end of the study.
Primary question
How did overall adherence to selected TBI guideline indicators compare between intervention and control sites after randomization? The endpoint is a mean percentage of adherence across nine indicators selected from 15 Brain Trauma Foundation indicators for their logistical relevance in the South American care context. A single numerical arm comparison is not reported in the registry record.
The trial addresses an implementation question: whether a training and care program changes how consistently selected guideline practices are followed in pediatric severe TBI care. The registry describes both guideline adherence and patient outcomes as topics of interest, but the primary endpoint is adherence, not a clinical outcome such as survival or functional status. Secondary outcomes include pathway adoption, Glasgow Outcome Scale measures, discharge survival, and mortality at 3 months.
3. Trial Design
The study is listed as a randomized, parallel-group trial with no masking. The record gives an enrollment of 505 and a completed status. The sponsor is the University of Washington. The study start date is 2019-09-09, and the primary completion date is 2023-09-26. The phase is listed as NA.
Sites, rather than individual participants, are described as allocated to the two conditions. Half of the sites were assigned to maintain usual care, while the other half were assigned to receive PEGASUS training and then deliver the program along with usual care. The number of sites and the allocation procedure are not reported in the registry record. Because adherence was clustered by site, participants cared for at the same site may share practices and outcomes; their observations cannot automatically be treated as independent.
All sites collected usual-care data during the baseline period. After randomization, intervention sites delivered PEGASUS and control sites continued usual care. The registry says the intervention arm was considered experimental with intention-to-treat after randomization. It also states that the post-randomization comparison of the primary endpoint used control and intervention arms, while baseline data contributed to other adjusted analyses. The exact timing and details of those other analyses are not reported in the registry record.
The study condition is traumatic brain injury. Its brief summary describes limited effective treatments and low adherence to existing evidence-based pediatric TBI guidelines as motivation for developing PEGASUS. The trial therefore evaluates an implementation program in the context of usual clinical care; it does not compare a drug against placebo. The registry does not report masking, consistent with its listing of NONE, or provide details on how site allocation was concealed.
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: Due to COVID-19, we reverted to a parallel-cluster design before randomization, resulting in a baseline period (9/1/19-7/13/20) with usual care data collected at all sites. Arm assignment and site notification was 7/13/20. Participants enrolled between 7/14/20-9/30/20 during intervention site training are included in the assigned arm (intention-to-treat) due to possible behavior change once assignment was known. Intervention implementation began for participants enrolled between 10/1/20-9/30/23.
Baseline
| Stage | Usual Care, Then Usual Care (Control) | Usual Care, Then PEGASUS Program (Intervention) |
|---|---|---|
| STARTED | 55 | 61 |
| 3-month Post-admission Follow-up | 52 | 55 |
| COMPLETED | 51 | 53 |
| NOT COMPLETED | 4 | 8 |
Reasons for not completing this period:
| Reason | Usual Care, Then Usual Care (Control) | Usual Care, Then PEGASUS Program (Intervention) |
|---|---|---|
| Lost to Follow-up | 1 | 2 |
| Death | 3 | 6 |
Post-Randomization
| Stage | Usual Care, Then Usual Care (Control) | Usual Care, Then PEGASUS Program (Intervention) |
|---|---|---|
| STARTED | 208 | 181 |
| 3-month Post-admission Follow-up | 196 | 168 |
| COMPLETED | 184 | 157 |
| NOT COMPLETED | 24 | 24 |
Reasons for not completing this period:
| Reason | Usual Care, Then Usual Care (Control) | Usual Care, Then PEGASUS Program (Intervention) |
|---|---|---|
| Lost to Follow-up | 12 | 11 |
| Death | 12 | 13 |
Analysis population for the primary outcome, as defined in the registry: We compared control and intervention arms post-randomization (07/13/20) in statistical analysis. Baseline period data contributed to other adjusted analyses.
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: 16 sites contributed participants during baseline period. After randomization, 8 of those sites were assigned Control and 8 were assigned Intervention. Participants were allocated to the study arms based on their admission to either the control study hospitals or intervention study hospitals.
| Characteristic | Unit | Usual Care, Then Usual Care (Control) | Usual Care, Then PEGASUS Program (Intervention) | Total |
|---|---|---|---|---|
| Participants analysed | 263 | 242 | 505 | |
| Age, Continuous: Baseline | years | 6.2 (SD 4.7) | 7.7 (SD 5.7) | 7.0 (SD 5.3) |
| Age, Continuous: Post-Randomization | years | 6.9 (SD 4.5) | 7.8 (SD 4.8) | 7.3 (SD 4.7) |
| Sex: Female, Male: Baseline – Female | Participants | 19 | 23 | 42 |
| Sex: Female, Male: Baseline – Male | Participants | 36 | 38 | 74 |
| Sex: Female, Male: Post-Randomization – Female | Participants | 77 | 58 | 135 |
| Sex: Female, Male: Post-Randomization – Male | Participants | 131 | 123 | 254 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Total Glasgow Coma Scale (GCS) score <=8: Baseline – At admission <=8 | Participants | 35 | 48 | 83 |
| Total Glasgow Coma Scale (GCS) score <=8: Baseline – Deteriorated <=8 | Participants | 18 | 13 | 31 |
| Total Glasgow Coma Scale (GCS) score <=8: Baseline – Total GCS >8 (but motor<=5) | Participants | 2 | 0 | 2 |
| Total Glasgow Coma Scale (GCS) score <=8: Post-Randomization – At admission <=8 | Participants | 146 | 138 | 284 |
| Total Glasgow Coma Scale (GCS) score <=8: Post-Randomization – Deteriorated <=8 | Participants | 62 | 43 | 105 |
| Total Glasgow Coma Scale (GCS) score <=8: Post-Randomization – Total GCS >8 (but motor<=5) | Participants | 0 | 0 | 0 |
| Any extracranial injury: Baseline – No | Participants | 20 | 19 | 39 |
| Any extracranial injury: Baseline – Yes | Participants | 35 | 42 | 77 |
| Any extracranial injury: Post-Randomization – No | Participants | 61 | 31 | 92 |
| Any extracranial injury: Post-Randomization – Yes | Participants | 147 | 150 | 297 |
| Injury Severity Scale (ISS): Baseline | units on a scale | 26.9 (SD 20.8) | 28.8 (SD 18.7) | 27.3 (SD 20) |
| Injury Severity Scale (ISS): Post-Randomization | units on a scale | 26.3 (SD 18.8) | 29.6 (SD 16.7) | 27.8 (SD 17.9) |
6. Primary Endpoint
The registry lists the primary outcome as First 3-day Cumulative Intensive Care Unit (ICU) Traumatic Brain Injury (TBI) Guideline Adherence Rate, Overall, measured over ICU Stay for severe TBI care, first 3 days from admission (or until death or discharge, if before 3 days)..
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | mean percentage of guideline adherence |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the first 3-day cumulative ICU TBI guideline adherence rate, overall, expressed as a mean percentage of guideline adherence with a standard deviation. It covers severe TBI ICU care during the first 3 days from admission, or until death or discharge if either occurred before 3 days.
The registry describes adherence for nine of the 15 Brain Trauma Foundation TBI guideline indicators. The indicators were selected as logistically relevant to the South American clinical care context. For each participant, adherence to each indicator was coded daily as yes, no, or not applicable. For continuous indicators, the daily adherence rate for an indicator was averaged over the first three days. For non-continuous indicators, adherence was counted as 100% if achieved on any of those days. The overall cumulative rate was then formed by averaging adherence rates across all nine indicators.
This measure is used to assess whether the program changes the delivery of recommended care practices. It combines adherence information across several indicators into one overall measure, rather than representing a single treatment or patient outcome. A mean describes the average adherence rate, while the standard deviation describes variation in the measured rates. The registry record does not report the arm-specific means, standard deviations, or a single numerical comparison for this endpoint.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies the primary question and defines how the adherence endpoint was constructed, but it does not provide a single numerical result comparing the randomized arms for that endpoint. It is therefore not possible to describe the direction or size of an arm difference, its precision, or whether the posted primary analysis supports a statistical distinction between groups. A result estimate, confidence interval, and p-value for one primary comparison are not reported in the registry record.
The reason a single comparison is unavailable is reflected in the record’s note that the primary outcome has several categories or time points and was not reduced to one comparison. The endpoint description specifies a cumulative measure over the first 3 days, but the record does not give the relevant arm-specific values or explain which multiple categories or time points underlie the analyses. The record lists seven outcome measures and two analyses, but that alone does not identify the exact analysis structure or resolve which estimate should be treated as the principal result.
The adherence score is a composite across nine indicators. A change in the overall mean could reflect changes in different indicators or in the mix of indicators that were applicable. The record describes the indicator-level coding rules but does not report indicator-specific results or the frequency of not-applicable classifications. The overall score should therefore be read as a summary of the specified adherence process, not as evidence that every guideline indicator changed in the same way.
Site clustering is important for interpretation because care practices are grouped within sites, and the registry explicitly says adherence was clustered by site and grouped by site allocation. Analyses that account for clustering generally reflect that participants within a site may be more alike than participants at different sites. The specific model, uncertainty estimation method, and adjustment variables are not reported in the registry record, so the degree to which these features were incorporated into any posted estimate cannot be assessed from this record.
The trial also lists clinical and patient-centered secondary outcomes, including Glasgow Outcome Scale measures, discharge survival, and mortality at 3 months. Their presence does not establish their results or their relationship to the adherence endpoint. Numerical results for those secondary outcomes are not reported in the registry record. The registry’s adverse-event entries are described separately and should not be treated as a substitute for efficacy or endpoint results.
Accordingly, interpretation should remain at the level of design and measurement: a randomized site-level implementation comparison was planned and conducted, and the primary endpoint measured average adherence across selected practices. Without a reported primary arm estimate and its uncertainty, the posted registry record does not establish the magnitude, precision, or statistical significance of an adherence difference.
9. Statistical Methods Explained
The statistical details available in the registry record clarify how the outcome was defined and which design features matter, but they do not specify a complete analysis plan. General explanations below describe the concepts involved; they should not be read as claims about methods beyond those explicitly recorded.
What does the mean adherence score represent?
The primary endpoint is a continuous measure expressed as a mean percentage. A mean summarizes the central level of measured adherence across the observations included in an analysis. Here, the endpoint definition first assigns adherence values for each participant and indicator, applies different rules for continuous and non-continuous indicators, and then averages across nine indicators. This produces an overall score that summarizes multiple care practices. The score is not a count of patients who received every recommended practice, nor does the description say that all indicators carry different weights.
The registry also names the standard deviation as the measure of dispersion. In general, a standard deviation describes how spread out measurements are around their mean. It is not the same as the uncertainty around the mean: a confidence interval is generally used to express uncertainty in an estimated quantity, taking account of the analysis and data structure. No standard deviation or confidence interval for the primary arm comparison is reported in the registry record.
Why does site clustering matter?
Randomization is described at the site level, and the registry says adherence was clustered by site. General statistical reasoning is that measurements from people treated at the same site may be correlated because they share staff, routines, and local implementation conditions. If this dependence is ignored, an analysis can overstate how much independent information the data contain and may produce uncertainty estimates that are too narrow. Methods that account for clustering adjust the analysis to reflect the grouping structure. The specific clustering method, any assumptions about within-site similarity, and the model used are not reported in the registry record.
Site-level allocation also means that the intervention contrast is tied to differences between allocated sites, not simply to individual assignment within each site. The number of sites and any details about balancing or stratification are not reported in the registry record. These details can matter for judging how well the randomized groups were balanced and how precisely a site-level effect can be estimated.
How should baseline and post-randomization data be distinguished?
The primary population description says that control and intervention arms were compared post-randomization, while baseline-period data contributed to other adjusted analyses. This distinction matters because baseline observations were collected before the intervention phase. General statistical practice may use baseline information to describe groups or improve precision, but the registry does not state the particular adjustment variables or models used for other analyses. It would be inappropriate to infer that baseline observations were included in the primary endpoint comparison.
The intervention-arm description uses the phrase intention-to-treat after randomization. In general, an intention-to-treat approach analyzes participants or clusters according to their randomized assignment, even if actual implementation differs. The record does not give enough detail to establish how this principle was applied to every endpoint or how departures from assigned intervention delivery were handled.
What can be inferred from the listed analysis information?
The record lists two analyses and says the primary outcome has several categories or time points and was not reduced to one comparison. Those statements indicate that a single summary comparison is not available in the registry record. They do not disclose the model terms, tests, contrasts, multiplicity handling, or how repeated measurements were treated. Those statistical analysis details are not reported in the registry record.
The sample-size assumptions describe planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They state that the effect to detect is a specified fraction of the observed effect, and distinguish participants per arm for outcomes other than time-to-event outcomes from total events for time-to-event calculations. Such calculations are planning illustrations, not observed power and not evidence about the result. The registry record does not report a sample-size justification for the trial’s original design or provide an analysis showing the probability of detecting an effect after the study was completed.
Over-dispersion is a general concern when observed variation is greater than a chosen statistical model expects. Accounting for it can improve uncertainty estimates when the model would otherwise understate variability. The registry record does not identify an over-dispersion assessment or adjustment. The same caution applies to assumptions about missing measurements, outcome distributions, and model fit: unless the record states that a method was used, its use should not be assumed.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists seven outcome measures and two analyses. The primary adherence endpoint is described as involving several categories or time points, and the record notes that it was not reduced to one comparison. These features raise questions about multiplicity: when several outcomes, comparisons, or time points are tested, the chance of a false-positive finding across the full set can differ from the chance associated with one prespecified test. The approach to multiplicity, including whether any adjustment or hierarchy was used, is not reported in the registry record.
The record does not provide a complete statistical analysis plan, nor does it specify which outcomes or time points were designated as confirmatory beyond naming the primary endpoint. It also does not report how the two listed analyses relate to the seven outcome measures. Results for the secondary outcomes are not reported in the registry record, so their estimates, uncertainty, and multiplicity context cannot be assessed.
Information on missing data is also limited. The endpoint coding allows a value of not applicable for an indicator, but the registry record does not report how often this occurred or how missing observations were distinguished from not-applicable observations. It does not state whether missing data were excluded, imputed, or handled by another method. The number of participants with endpoint data in each arm and the reasons for missingness are not reported in the registry record.
The record describes a post-randomization comparison for the primary endpoint and says baseline data contributed to other adjusted analyses. It does not report the detailed analysis population, exclusions after randomization, or how incomplete follow-up was handled. Although the intervention-arm description mentions intention-to-treat, the registry record does not provide enough information to verify the implementation of that principle for each outcome analysis. These omissions limit assessment of potential bias arising from missingness, analysis-population choices, or selective emphasis among multiple results.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Usual Care at Baseline (Site That Was Later Assigned to the Control) | 6 | 55 | 10.9% |
| Usual Care at Post-Randomization Control Site | 34 | 208 | 16.3% |
| Usual Care at Baseline (Site That Was Later Assigned to the Intervention) | 7 | 61 | 11.5% |
| PEGASUS Program at Post-Randomization Intervention Site | 17 | 181 | 9.4% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Usual Care at Baseline (Site That Was Later Assigned to the Control) | 3 of 55 |
| Usual Care at Post-Randomization Control Site | 12 of 208 |
| Usual Care at Baseline (Site That Was Later Assigned to the Intervention) | 6 of 61 |
| PEGASUS Program at Post-Randomization Intervention Site | 13 of 181 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Usual Care at Baseline (Site That Was Later Assigned to the Control) | 21 of 55 |
| Usual Care at Post-Randomization Control Site | 72 of 208 |
| Usual Care at Baseline (Site That Was Later Assigned to the Intervention) | 16 of 61 |
| PEGASUS Program at Post-Randomization Intervention Site | 78 of 181 |
Most frequent serious adverse event terms
| Event term | Organ system | Usual Care at Baseline (Site That Was Later Assigned to the Control) | Usual Care at Post-Randomization Control Site | Usual Care at Baseline (Site That Was Later Assigned to the Intervention) | PEGASUS Program at Post-Randomization Intervention Site |
|---|---|---|---|---|---|
| Shock | General disorders | 4 of 55 | 17 of 208 | 3 of 61 | 8 of 181 |
| Other: Neurological | Nervous system disorders | 0 of 55 | 12 of 208 | 1 of 61 | 4 of 181 |
| Pneumonia | Infections and infestations | 1 of 55 | 9 of 208 | 1 of 61 | 5 of 181 |
| Intraparenchymal hemorrhage | Nervous system disorders | 2 of 55 | 3 of 208 | 1 of 61 | 2 of 181 |
| Other: Respiratory | Respiratory, thoracic and mediastinal disorders | 2 of 55 | 3 of 208 | 2 of 61 | 0 of 181 |
| Other: Electrolyte/Metabolic | Metabolism and nutrition disorders | 0 of 55 | 4 of 208 | 0 of 61 | 1 of 181 |
| Arrhythmia | Cardiac disorders | 0 of 55 | 2 of 208 | 0 of 61 | 1 of 181 |
| Other: Infectious | Infections and infestations | 0 of 55 | 2 of 208 | 0 of 61 | 1 of 181 |
| Other: Cardiac | Cardiac disorders | 0 of 55 | 1 of 208 | 0 of 61 | 0 of 181 |
| Peritonitis | Infections and infestations | 0 of 55 | 0 of 208 | 0 of 61 | 0 of 181 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Usual Care at Baseline (Site That Was Later Assigned to the Control) | Usual Care at Post-Randomization Control Site | Usual Care at Baseline (Site That Was Later Assigned to the Intervention) | PEGASUS Program at Post-Randomization Intervention Site |
|---|---|---|---|---|---|
| Pneumonia | Infections and infestations | 6 of 55 | 49 of 208 | 10 of 61 | 40 of 181 |
| Other: Neurological | Nervous system disorders | 5 of 55 | 20 of 208 | 2 of 61 | 21 of 181 |
| Other: Infectious | Infections and infestations | 4 of 55 | 4 of 208 | 4 of 61 | 19 of 181 |
| Other: Respiratory | Respiratory, thoracic and mediastinal disorders | 5 of 55 | 4 of 208 | 0 of 61 | 3 of 181 |
| Shock | General disorders | 2 of 55 | 3 of 208 | 1 of 61 | 2 of 181 |
| Intraparenchymal hemorrhage | Nervous system disorders | 2 of 55 | 2 of 208 | 1 of 61 | 1 of 181 |
| Other: Hematology | Blood and lymphatic system disorders | 0 of 55 | 2 of 208 | 1 of 61 | 3 of 181 |
| Other: Electrolyte/Metabolic | Metabolism and nutrition disorders | 1 of 55 | 2 of 208 | 1 of 61 | 2 of 181 |
| Other: Gastrointestinal | Gastrointestinal disorders | 0 of 55 | 1 of 208 | 1 of 61 | 2 of 181 |
| Other: Ophthalmalogical | Eye disorders | 0 of 55 | 2 of 208 | 0 of 61 | 1 of 181 |
Serious events with the highest arm-level rate: shock (8.2%), other: neurological (5.8%), pneumonia (4.3%), intraparenchymal hemorrhage (3.6%), other: respiratory (3.6%).
The adverse-event record gives separate entries for baseline periods and post-randomization periods, with labels distinguishing sites later assigned to control or intervention and the PEGASUS post-randomization period. It reports numbers at risk and numbers affected for serious and other adverse events, along with deaths. These are descriptive counts within the named groups; they are not a single randomized comparison and should not be interpreted as one.
Among the listed serious-event terms, the registry reports Shock at 8.2%, Other: Neurological at 5.8%, Pneumonia at 4.3%, Intraparenchymal hemorrhage at 3.6%, Other: Respiratory at 3.6%, and Other: Electrolyte/Metabolic at 1.9%. The record does not explain the method used to calculate these percentages or provide a comparative estimate with uncertainty for these terms. The entries do not establish whether a difference between periods or assigned groups is statistically distinguishable.
Baseline and post-randomization entries refer to different periods and, in some cases, different care conditions. Directly comparing their counts without accounting for group size, timing, and study design would be misleading. Serious adverse events, other adverse events, and deaths are listed as separate measures; the record does not supply a causal attribution assessment or additional detail on event definitions. These safety entries do not replace the missing primary endpoint comparison.
12. What the Result Does, and Does Not, Mean
This trial is relevant to the statistical study of implementation because its primary endpoint measures adherence to guideline indicators, rather than a direct biological response to a treatment. The randomized parallel-site design asks whether introducing PEGASUS at assigned sites is associated with a difference in the defined adherence measure compared with continuing usual care. The outcome construction and site-level grouping are central to understanding what the comparison represents.
The record establishes that the primary endpoint was defined as a mean percentage across nine selected indicators and that the comparison was intended for the post-randomization period. It also identifies several secondary outcomes and gives some adverse-event counts. However, a numerical primary comparison, its uncertainty, and the results for the listed secondary outcomes are not reported in the registry record. The record therefore does not establish the direction or magnitude of an effect on adherence, nor does it establish a difference in survival, function, or other clinical outcomes.
A mean adherence score is not interchangeable with patient benefit. It summarizes recorded adherence to the selected indicators under the stated coding rules. Even if an adherence difference were reported, its clinical meaning would depend on the indicator-specific pattern, the applicability of practices, and the relationship between those practices and patient outcomes. Those details are not reported in the registry record. Conversely, the lack of a single numerical comparison in this record is not evidence that the program had no effect; it means the registry record does not provide the result needed to characterize that effect.
The results should also not be generalized beyond the population and care context described in the record without further evidence. The endpoint was tailored to indicators considered logistically relevant in the South American clinical care context, and allocation occurred at the site level. The registry record does not provide the number of participating sites or detailed information about their characteristics. Thus, the record alone cannot show how consistent outcomes were across settings.
The appropriate statistical conclusion is narrow: the registry describes a completed randomized evaluation and a defined adherence endpoint, but it does not provide the single arm comparison necessary to quantify the primary result. No treatment recommendation follows from these registry entries alone.
13. Why This Trial Matters Statistically
Implementation trials pose statistical questions that differ from those in many individually randomized treatment studies. In this study, sites were allocated to usual care or to PEGASUS after a baseline period, and the primary measure concerns whether selected care practices were followed. The unit of allocation and the unit of measurement therefore need to be considered together. The explicit statement that adherence was clustered by site signals that within-site dependence is part of the analysis context.
The endpoint also illustrates the trade-off involved in composite measures. Combining adherence across nine indicators can provide a broad summary of practice, but a single average may conceal differences among its components. The registry specifies different scoring rules for continuous and non-continuous indicators and permits a not-applicable classification. These choices define what the mean represents and affect how it can be interpreted. Without indicator-level results and details about applicability, readers cannot determine which components drive the overall measure.
The trial’s two-period structure makes clear why baseline and post-randomization observations should not be casually combined. Baseline data were collected at all sites, while the primary comparison is described as post-randomization; baseline data contributed to other adjusted analyses. Keeping those roles distinct helps avoid confusing a pre-intervention description with an intervention effect.
Finally, this record demonstrates the importance of complete results reporting. It identifies the trial design, endpoint construction, secondary outcomes, and safety entries, but a single primary arm comparison and its precision are not reported in the registry record. A well-defined endpoint is necessary for interpretation, yet it cannot substitute for an estimate and an uncertainty measure. For statisticians and clinical researchers, the combination of site-level allocation, composite adherence scoring, multiple listed outcomes, and an unavailable primary comparison makes transparent analysis reporting especially important.
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 a single numerical comparison for the primary endpoint. Arm-specific means, standard deviations, an effect estimate, and its uncertainty are not available in the record. This prevents assessment of the size and precision of any observed difference.
Several design and analysis details are also not reported in the registry record: the number and characteristics of sites, the randomization procedure, allocation concealment, detailed eligibility criteria, the statistical model, adjustment variables, and the method used to account for site clustering. The record states that adherence was clustered by site, but does not specify the implementation or assumptions of that analysis.
The composite endpoint combines nine indicators, with separate coding rules for continuous and non-continuous indicators. Indicator-specific results and the frequency of not-applicable classifications are not reported. Missing-data amounts, reasons for missingness, and handling methods are also not reported in the registry record. These omissions limit evaluation of how the observed endpoint would reflect incomplete or inapplicable measurements.
The registry lists several outcome measures and analyses, but the analysis hierarchy and multiplicity approach are not reported. The available record also does not provide numerical results for the listed clinical secondary outcomes. The safety entries include counts and selected serious-event terms, but do not provide a comparative causal analysis or uncertainty estimates. Finally, the intervention is described as site training followed by PEGASUS delivery along with usual care; the extent and fidelity of implementation across sites are not reported in the registry record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Pediatric Guideline Adherence and Outcomes- Argentina — NCT03896789.
- Registry record last updated 2025-01-29; status last verified 2025-01.
- Results first posted 2025-01-29.
- 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.
18. Record Summary
NCT03896789 is a completed randomized, parallel-group study of the PEGASUS program for pediatric traumatic brain injury care in Argentina. The registry lists no masking, an enrollment of 505, the University of Washington as sponsor, a start date of 2019-09-09, and a primary completion date of 2023-09-26.
Sites assigned to the intervention received PEGASUS training and delivered the program with usual care after randomization; control sites maintained usual care and were offered training at the end of the study. The primary endpoint was a mean percentage of adherence across nine selected TBI guideline indicators during the first 3 days of severe TBI ICU care, or until death or discharge if sooner.
The record lists seven outcome measures and two analyses. It does not report a single numerical primary arm comparison or results for the listed secondary outcomes. Adverse-event entries describe baseline and post-randomization groups, but do not provide a comparative analysis that resolves the primary endpoint question.