An Evaluation of an Integrated Approach to Prevention and Early Intervention in the Elementary School Years: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for An Evaluation of an Integrated Approach to Prevention and Early Intervention in the Elementary School Years (NCT03132805), a randomized na study. The primary outcome is Change in Direct Classroom Observations of Student Behavior Between Pre-test and Post-test, 6 Months After the Pre-test..

2. Clinical Question

The trial also includes the no-intervention control condition, so the design encompasses more than the central comparison stated in the study summary. The registry describes behavior as including both on-task behavior and physical aggression. The primary endpoint records the percentage of observed intervals in which the behaviors occurred, using direct observation rather than relying solely on reports from students or teachers.

The registry record does not provide numerical outcome results, estimates of change, or a reported comparison between the combined and universal-only approaches. It therefore does not establish the direction or size of any observed difference. The question can be identified from the registered aims and intervention descriptions, but its answer cannot be inferred from the trial description alone.

3. Trial Design

NCT03132805 is recorded as a randomized, single-masked, factorial trial with an enrollment of 5233. Its phase is listed as not applicable, and its status is completed. The sponsor is Johns Hopkins Bloomberg School of Public Health. The recorded start date is 2016-06-01, and the primary completion date is 2021-04-30.

The intervention descriptions are school-based. Schools in the control condition receive no intervention. Schools in the PATHS to PAX condition receive a universal classroom-based preventive intervention designed to reduce aggression. The combined condition adds the Incredible Years child and parent groups to PATHS to PAX. The registry lists both interventions as behavioral.

Randomization means assignment to study conditions is determined by a chance process rather than chosen solely by participants or investigators. As a general statistical principle, random assignment can reduce systematic differences between groups at the start of a trial, although it does not guarantee that groups will be identical. The record does not describe the allocation procedure, allocation ratio, or whether assignment occurred at the school or another level. Those details are not reported in the registry record.

The trial is labeled factorial, but the registry description does not explain the factors, their levels, or the statistical contrasts planned for them. It is therefore not possible to reconstruct the intended factorial analysis from the label alone. Single masking is recorded, but who was masked and how masking was maintained are not reported in the registry record.

The endpoint is based on observations at pre-test and at post-test six months later. Independent observers conducted observations on two occasions, one week apart, at each assessment period. The record also names four secondary outcomes concerning peer assessment, perceived competence and social acceptance, teacher self-efficacy, and burnout. The total number of outcome measures recorded is 6.

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: In accordance with our original protocol, we successfully recruited 48 schools over the four cohorts, which were subsequently randomized to one of three conditions. Randomization occurred at the school level and not at the teacher or student level within school. A total of 4,923 students with parental consent to participate were enrolled over the four cohorts from the 48 participating schools. A total of 310 teachers were also enrolled over the four cohorts from the 48 participating schools.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Students enrolled within participating schools with parental consent to participate in the trial who were observed at baseline and at post-test.

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: Students and teachers enrolled within participating schools who were assessed at baseline.

6. Primary Endpoint

The registry lists the primary outcome as Change in Direct Classroom Observations of Student Behavior Between Pre-test and Post-test, 6 Months After the Pre-test., measured over The observations are carried out at pre-test and at post-test, 6 months after the pre-test..

The primary endpoint is change in direct classroom observations of student behavior between pre-test and post-test, six months after pre-test. The registry identifies the unit as the percentage of intervals observed and the summary parameter as a mean, with standard error as the stated dispersion measure. The behaviors of interest are on-task behavior and physical aggression.

Independent observers watched students in 10-second intervals. A behavior was recorded as present if it occurred at least once during an interval. For each behavior, the interval score could range from 0 to 1: 0 means the behavior was not observed, and 1 means it was observed. The scores used in analysis were each student’s average across the observed intervals. Observations were conducted on two occasions, one week apart, at both pre-test and post-test.

Direct observation offers a way to measure behavior in the classroom setting in which the intervention is intended to operate. Repeated observations can capture behavior across more than a single instant, while the pre-test and post-test assessments define the change being studied. These are general measurement considerations, not evidence that the endpoint is free of measurement error. The registry record does not report observer training, agreement between observers, or the exact method used to combine observations across occasions.

The endpoint description names both on-task behavior and physical aggression. The registry record does not report a single numerical result for either behavior or a combined primary estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the trial as randomized, single-masked, and factorial, and describes the primary endpoint as a continuous measure summarized by a mean. It does not report the statistical model, the planned primary contrast, or an analysis result. The explanations below describe general statistical concepts relevant to this design; they are not claims about methods used in the trial.

What does a mean represent?

A mean summarizes the central level of a continuous measure across observations. In this study, each student’s score was based on the average of interval scores for observed behavior. A group mean would summarize those student-level scores within a condition. Comparing means can describe an average difference, but the mean alone does not show how variable individual observations were or how precisely the group contrast was estimated.

What does a standard error describe?

A standard error describes the uncertainty of an estimated statistic under a statistical model and its assumptions. It is not the same as the spread of individual student scores. A confidence interval is often constructed using an estimate and its standard error; in general, it expresses a range of values compatible with the data and the method used. Neither standard errors nor confidence intervals for the primary outcome are reported in the registry record.

Why does the design matter?

Randomization is intended to make comparisons less vulnerable to systematic differences between study conditions. In general, it supports causal interpretation when assignment is implemented as intended and when the analysis respects the assignment process and relevant features of the data. The record does not specify the randomization unit or allocation procedure. It describes conditions at the school level, but the exact assignment structure is not reported in the registry record.

A factorial design generally studies multiple intervention components or factors in a structured set of conditions. Depending on the factor structure and the questions planned, analyses may estimate effects associated with individual factors, combinations, or interactions. An interaction asks whether the effect associated with one component differs according to the level of another component. These are general features of factorial designs; the registered label alone does not reveal which factors or contrasts were planned in this trial. The factor definitions and analysis plan are not reported in the registry record.

What issues arise with classroom observations?

The endpoint uses repeated observations of students within schools. In general, observations from students in the same school may be more alike than observations from students in different schools. Such dependence can affect standard errors if it is not addressed appropriately. The intervention is described at the school level, but the registry record does not state the assignment unit or whether the analysis accounted for grouping by school, classroom, observer, or repeated observations.

The endpoint also summarizes behavior recorded in 10-second intervals. Interval-level scoring is a way to operationalize the observation, but an average score still depends on what was observed and how the observations were combined. General measurement concerns include variability in behavior across occasions and consistency in how observers classify it. Observer agreement and the analysis treatment of the two observation occasions are not reported in the registry record.

What can be inferred from the planning assumptions?

The registry’s sample-size assumptions describe a planning illustration based on an observed effect, not the trial’s own design justification. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. For time-to-event calculations, the illustration concerns the total number of events; other illustrations concern participants per arm.

These assumptions are not evidence that the trial was designed using those values, and they do not establish the analysis plan. A sample-size illustration is a planning exercise, not observed power and not a substitute for an estimate with uncertainty. The illustration’s numerical results are not repeated. The trial’s original sample-size justification and statistical plan are not reported in the registry record.

What is missing for an inferential reading?

To interpret an inferential result, readers would usually need an estimate for a defined contrast, an uncertainty measure, and an account of the analysis model and population. A p-value, when used, describes how unusual data at least as incompatible with a specified null model would be under that model; it does not quantify the size or practical importance of an effect. No p-value or effect estimate for the primary endpoint is reported in the registry record, so inferential conclusions cannot be drawn from the design description alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry records 6 outcome measures and names four secondary outcomes: change in the Peer Assessment Inventory, the Pictorial Scale of Perceived Competence and Social Acceptance for Young Children, the Teacher Sense of Self Efficacy Scale, and the Maslach Burnout Inventory. The primary endpoint concerns direct classroom observations. The record does not describe how the outcomes were ordered for testing or whether any adjustment for multiple comparisons was planned.

As a general statistical issue, testing several outcomes or several contrasts can increase the chance of finding at least one apparently unusual result even when there is no underlying effect. Multiplicity methods can address this issue, but the choice depends on the study questions and analysis plan. No correction method or multiplicity strategy is reported in the registry record. The number of analyses recorded is 0, and there are no numerical outcome analyses to assess.

Missing-data information is also not reported in the registry record. The primary endpoint population is described as students enrolled in participating schools with parental consent who were observed at baseline and post-test. The record does not state how many enrolled students had observations at both time points, how missing observations were handled, or whether the analysis included all randomized participants or a subset.

These details matter because an analysis limited to participants with observations at both assessments may not represent all randomized participants if the availability of observations is related to behavior or intervention condition. This is a general concern, not a finding that such a pattern occurred. The registry record does not describe an analysis population, missing-data assumptions, imputation, or sensitivity analyses.

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 registry record reports zero serious adverse events and zero other adverse events in each of the three groups. The number at risk for serious and other adverse events is 1629 in Control, 1792 in PATHS to PAX, and 1812 in PATHS to PAX and the Incredible Years. Deaths are recorded as 0 in each group. No serious adverse event terms are listed.

These figures describe what the registry record reports; they are not evidence that no adverse events could occur. A zero count means that no events were recorded in the posted results for the specified category and group. It does not by itself establish that the groups had identical safety experience or that all possible events were captured.

The counts at risk differ across the groups, and the registry record does not provide additional detail about event collection, reporting thresholds, follow-up for safety, or the reasons for the group counts. The figures therefore support only a narrow reading of reported events. They should not be interpreted as an assessment of comparative safety beyond the recorded counts.

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

The registry record identifies a randomized trial of three school conditions and defines a primary behavioral endpoint based on direct observation before and after intervention. It also gives the central study question: whether adding the Incredible Years groups to PATHS to PAX would yield greater impact on aggression than PATHS to PAX alone. Those details clarify what the study was designed to examine.

They do not establish what the trial found. The registry record does not report numerical primary outcome results or analyses. Without estimates and measures of uncertainty, the relative change in observed on-task behavior or physical aggression cannot be described. Nor can the evidence be classified as indicating a statistically distinguishable difference, an imprecise estimate, or a result compatible with several possible effects.

Even when numerical results are available, a statistical contrast is specific to its endpoint, population, time point, and analysis. Here, the described endpoint applies to students with parental consent who were observed at baseline and post-test, and the post-test is six months after pre-test. It should not be generalized from the registry description alone to other populations, settings, or time periods.

The study’s completed status and enrollment of 5233 are trial-record details, not substitutes for outcome estimates. Likewise, the safety counts report no recorded events in the listed categories, but do not answer the behavioral effectiveness question. The appropriate interpretation is therefore limited: the record defines an evaluable research question and its measurement approach, while leaving the comparative result and its precision unreported.

13. Why This Trial Matters Statistically

This trial is statistically notable for linking an intervention question to direct observations of behavior in classrooms. The primary endpoint is not merely a general label such as aggression; it specifies observed behaviors, 10-second intervals, a scoring rule, two observation occasions one week apart at each assessment, and a post-test six months after pre-test. Clear operational definitions are important because statistical analyses are only as interpretable as the measures they summarize.

The randomized, factorial designation also raises useful design questions. Randomization can support comparisons between conditions, while factorial structures can address questions about combinations of components. However, the label alone does not identify the factors, assignment process, or planned contrasts. Those design details are necessary to understand which comparisons would answer the study’s central question and how uncertainty should be calculated.

The school setting makes the structure of the observations relevant. Students and classrooms are situated within schools, and general statistical methods may need to account for related observations when estimating uncertainty. The registry record does not explain whether or how this structure was handled. That omission illustrates why a trial’s design label and endpoint description do not fully determine the statistical analysis.

Finally, the record’s lack of numerical outcome analyses is an important distinction between a registered research question and an interpretable result. The endpoint and groups can be described, but without estimates and uncertainty there is no basis to quantify the comparison. This matters statistically because transparent reporting of both the design and analysis is needed to assess what the data show and how precisely they support a conclusion.

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 interpreting this trial from the registry record is the absence of numerical primary outcome results and analyses. No group means, changes, contrasts, standard errors, confidence intervals, or p-values are reported. The record lists 0 analyses, so the comparative question cannot be answered from the posted information.

The primary endpoint includes observations of on-task behavior and physical aggression, but the registry record does not present a single numerical comparison for either measure. It also does not report how the several categories or time points were reduced to a single analysis. This limits the ability to identify a unique primary contrast or evaluate the endpoint’s statistical interpretation.

Important design and analysis details are not reported in the registry record, including the randomization unit, allocation procedure, masking details, factorial structure, planned contrasts, statistical model, and handling of school-level or repeated-observation dependence. Missing-data methods and the analysis population are also not described. These are unknowns in the available record, not evidence that a particular method was or was not used.

The endpoint population is restricted in its description to students in participating schools with parental consent and observations at both baseline and post-test. The registry record does not report how this group compares with all enrolled students or how many students contributed endpoint data. The endpoint’s specified post-test is six months after pre-test; longer-term measurement is not described for this primary outcome.

Finally, the record lists four secondary outcomes but does not report their findings or a strategy for handling multiple comparisons. Together, these limitations prevent assessment of effect size, precision, robustness, and the range of outcomes considered in the analysis.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: An Evaluation of an Integrated Approach to Prevention and Early Intervention in the Elementary School Years — NCT03132805.
• Registry record last updated 2022-08-02; status last verified 2022-07.
• Results first posted 2022-07-29.
• 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.

18. Record Summary

NCT03132805 evaluates school-based approaches to preventing and addressing aggression. It is recorded as randomized, single-masked, and factorial, with enrollment of 5233 and completed status. The study compares a no-intervention control, PATHS to PAX alone, and PATHS to PAX combined with Incredible Years child and parent groups.

The primary outcome is change in directly observed classroom behavior from pre-test to post-test six months later. Independent observers recorded on-task behavior and physical aggression in 10-second intervals. The registry also names four secondary outcomes concerning peer assessment, perceived competence and social acceptance, teacher self-efficacy, and burnout.

The registry record does not report numerical primary outcome results, statistical analyses, or the detailed analysis plan. It reports zero serious adverse events, zero other adverse events, and zero deaths in each group. Consequently, the record describes the trial question and measurement approach but does not provide an estimate of the comparative behavioral outcome.