COmmuNity-engaged SimULation Training for Blood Pressure Control: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for COmmuNity-engaged SimULation Training for Blood Pressure Control (NCT03375918), a na na study. The primary outcome is Blood Pressure.

2. Clinical Question

The central statistical question is whether the measured blood pressure distributions differed across these periods, and how precisely any difference was estimated. Because the comparison follows a sequence in time rather than random assignment, a period difference would not by itself establish that training caused the difference. Other changes over time could also contribute. The available registry record identifies the intended outcome and periods, but not an estimated effect or its uncertainty.

3. Trial Design

CONSULT-BP is a completed, sequential study of an educational intervention for healthcare trainees. The registry lists allocation as NA and masking as none. Its design description states that the healthcare trainees had a pre-intervention control period and a post-intervention exposure period. Thus, the intended contrast is between periods before and after training, not between randomly assigned, concurrent groups.

The intervention is named CONSULT-BP Educational Intervention. The stated training goals are to help early-stage providers mitigate bias, improve communication, and develop authentic therapeutic alliances with patients. The setting is an academic safety net hospital serving a large non-White and poor population. The registry lists enrollment of 118, which is the reported study enrollment; patient blood pressure records are described separately by period.

The control period is 6 months before the intervention, while the intervention period is up to 6 months after it. The primary outcome is systolic and diastolic blood pressure recorded in electronic medical records for patients treated by participating clinician trainees. The registry also lists the RELATE Checklist, a standardized patient rating scale for bias-mitigating and patient-centered communication, as a secondary outcome.

The registry records the study as completed, with a start date of 2018-08-13 and primary completion date of 2022-12-31. It lists phase as NA. The registry record does not describe a randomized allocation process, concurrent control group, or masking. Further details about implementation, trainee participation over time, and the analysis plan are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: To maintain clinical staff services in the hospital, we were not able to maintain the cluster randomization of trainees. We aimed to enroll cohorts of trainees over a 3-year period, with each academic year constituting an intervention cycle, with 3 cycles of implementation corresponding to 3 sequential academic years. Due to COVID, we extended enrollment to a 4-year period. Please note - only trainee participants were considered enrolled. Patient participants were not considered 'enrolled'.

Only the first 6 of 8 groups are shown.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Blood pressure data was obtained from electronic medical records of 944 patients during the control period and 1,651 period patients during the intervention period.

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: Baseline information on Healthcare Trainees, Patients--Control Arm and Patients--Intervention Arm are aggregated from all cohorts within each population.

6. Primary Endpoint

The registry lists the primary outcome as Blood Pressure, measured over Control period is 6 months before the intervention and intervention period is up to 6 months after the intervention.

The primary endpoint is blood pressure, measured as systolic and diastolic values in mm Hg and reported as means with standard deviations. The registry says these measurements came from patients’ electronic medical records and were associated with care by participating clinician trainees. Data are described for a control period before the intervention and an intervention period after it.

Blood pressure is relevant to the study’s stated question because the investigators hypothesized that patient blood pressure levels would be lower after trainees received the intervention than before training. Reporting systolic and diastolic measures allows the two components of blood pressure to be considered separately. A mean describes the central level of a continuous measure, while a standard deviation describes how spread out individual measurements are around that mean. These are general statistical interpretations, not findings about the study’s results.

The registry record does not report the period-specific means or standard deviations, a difference between periods, or an uncertainty interval. It therefore does not establish the direction, size, or precision of any observed blood pressure contrast. The methods used to select measurements from the electronic record, including any rules for repeated readings or multiple visits, are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is continuous: systolic and diastolic blood pressure in mm Hg. The registry specifies mean and standard deviation as the summary form. In general, a mean uses the observed values to describe their central level, while the standard deviation communicates their dispersion. The pair helps readers understand both the average and the variability in measurements. It does not, on its own, show the uncertainty in a comparison between periods; that requires an analysis that uses the available observations and its assumptions.

A natural descriptive contrast for this question would compare the average blood pressure in the control period with the average in the intervention period, considering systolic and diastolic measures separately. This describes what an analysis might estimate; it is not a claim that the registry reports such estimates. The record gives no period-specific means, standard deviations, mean differences, confidence intervals, or p-values. It lists 0 analyses. Consequently, neither the statistical significance nor the precision of a period contrast can be evaluated from the registry record.

In general, a confidence interval expresses uncertainty around an estimate under a chosen model and analysis. Its interpretation depends on how the data were collected and on assumptions such as independence, distributional behavior, and the way repeated observations are handled. A p-value is also model-dependent and does not measure the size or practical importance of an effect. Neither statistic is reported for the primary outcome in the registry record. General statistical concepts cannot substitute for an actual estimate and its uncertainty.

Clustering is a relevant general consideration for this design because patients were treated by participating clinician trainees. Measurements associated with the same trainee may be more alike than measurements associated with different trainees. Repeated readings for a patient could also be correlated. When observations are correlated, an analysis that assumes independence may give uncertainty estimates that are too small. Methods such as multilevel models or cluster-robust standard errors can address some forms of correlation, depending on the design and data structure. The registry record does not say whether there were repeated measurements, whether observations were grouped by trainee in the analysis, or whether any correlation-adjusted method was used.

Over-dispersion is a general issue when observed variability is greater than a statistical model expects. It is especially discussed for count outcomes, but the broader lesson is that a model’s assumed variability should be checked against the data. For a continuous endpoint, the distribution, outliers, and unequal variability between periods may also affect the suitability of a model and the reliability of standard errors. The registry does not report diagnostic checks, transformations, or variance-handling methods. No conclusion about these features can be drawn from the endpoint description alone.

The before-and-after sequence creates a distinction between a descriptive time-period contrast and a causal effect. In a randomized concurrent comparison, assignment can help balance other influences on average. Here, the registry describes a sequential model and allocation as NA. In general, changes in clinical practice, patient composition, or measurement processes over time can be mixed with the intervention contrast. Statistical adjustment may address measured differences, but it cannot automatically remove unmeasured time-related influences. The registry record does not report an adjustment strategy or covariates included in an analysis.

The record describes a sample-size planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. Such an illustration is a planning exercise, not the trial’s own design justification, and any sample-size figures are planning illustrations rather than observed power. The registry record does not report the trial’s sample-size justification, and it does not give the effect estimate needed to assess the illustration’s inputs. A planning calculation cannot establish whether the observed data produced a precise estimate.

For time-to-event outcomes, some sample-size approaches refer to total events, whereas the record’s description distinguishes those from other outcomes, which are expressed as participants per arm. The primary outcome is a continuous mean, not a time-to-event measure. In any event, the registry does not report numerical planning results for this outcome. The stated planning assumptions should not be confused with the actual analysis or with evidence about the observed blood pressure results.

Overall, the registry specifies the outcome type and period structure but leaves key analysis details unreported: the statistical model, handling of clustering or repeated records, covariate adjustment, missing-data approach, and uncertainty estimates. These details are important for judging both the validity and precision of a period comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies blood pressure as the primary outcome and lists the RELATE Checklist as a secondary outcome. It also reports 2 outcome measures. Because more than one outcome is listed, multiplicity is a general consideration: repeated testing across outcomes can increase the chance of obtaining at least one apparently unusual result by chance. Researchers may address this through a prespecified hierarchy or an adjustment to testing, but the appropriate approach depends on the study question and analysis plan. The registry record does not report a multiplicity strategy, and it reports 0 analyses, so the extent of any testing cannot be assessed.

Missing-data handling is not reported in the registry record. The outcome description identifies electronic medical records as the source of blood pressure data, but does not explain whether all eligible readings were available, how missing values were handled, or whether measurements were excluded. In general, missingness can affect both the population represented by the observed data and the estimated contrast, especially if availability is related to patient characteristics or care. The record does not provide enough information to determine whether missingness occurred or how it might affect results.

The reported enrollment is 118, while the blood pressure description separately identifies 944 patients in the control period and 1,651 period patients in the intervention period. The record does not define a complete analysis population, specify eligibility rules for measurements, or say whether the same patient could contribute observations in more than one period. It also does not explain how patients’ records were associated with particular trainees or whether analyses used trainee-level groupings. These limits make it difficult to assess the relationship between enrollment, outcome records, and the population used for any analysis.

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 lists serious adverse events, other adverse events, and deaths as zero in each of the reported groups: healthcare trainees, patients in the control arm, and patients in the intervention arm. The corresponding at-risk counts are listed as 118 for healthcare trainees, 944 for patients in the control arm, and 1,651 for patients in the intervention arm. No serious-event terms are listed.

These are registry-reported counts, not a complete description of how safety events were collected or reviewed. Zero reported events means that the record lists no events in those categories; it does not prove that no event occurred or establish that the intervention is free of risk. The registry does not report the event-ascertainment procedures, the criteria for classifying events, or how long each group was monitored. It also does not provide additional context for interpreting the patient groups’ at-risk counts. The safety figures should therefore be read as the recorded adverse-event information, not as a causal comparison between periods.

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

The study addresses an important statistical challenge: evaluating a training intervention whose intended effect is on patient outcomes, using routine clinical measurements collected before and after training. Its primary outcome is clearly identified as systolic and diastolic blood pressure in electronic medical records. The registry also states the direction of the investigators’ hypothesis and describes the periods being compared.

What the registry record does not establish is the observed blood pressure result. It reports no period means, standard deviations, estimated contrast, confidence interval, or p-value, and it lists no analyses. Without these, the direction and size of any observed difference, as well as its precision, cannot be evaluated. The absence of reported results should not be treated as evidence of no change or as evidence of a benefit.

Even if a period difference were reported, a sequential before-and-after design would require cautious interpretation. The timing of the intervention overlaps with any other changes between periods. A measured difference could be consistent with an intervention effect, but it could also reflect changes in patient characteristics, clinical care, or record capture. These are general alternative explanations, not findings about this study. The registry’s allocation and design descriptions do not support interpreting the comparison as randomized.

The registry’s stated planning assumptions are not a substitute for outcome results. Planning illustrations describe assumptions for a hypothetical design calculation; they are not observed power and do not reveal how precisely the completed study estimated an effect. The reported enrollment and period-specific patient-record counts likewise do not, on their own, establish the effective information available for a comparison, particularly when the handling of repeated records and trainee-level clustering is not described.

Accordingly, the defensible conclusion from the record is about study intent and structure: the investigators planned a sequential comparison of blood pressure before and after CONSULT-BP training. Whether the posted data show a difference, and how much uncertainty surrounds it, is not reported in the registry record.

13. Why This Trial Matters Statistically

This study illustrates why the unit of intervention and the unit of outcome matter in clinical research. Training is directed at healthcare trainees, while the primary outcomes are blood pressure measurements among patients treated by those trainees. The relationship between trainees, patients, and repeated electronic-record measurements affects how an analysis should represent the data and calculate uncertainty. The registry describes these elements but does not report whether analyses accounted for their dependence.

The design also highlights the distinction between measuring change and attributing change. Comparing a pre-intervention period with a later post-intervention period can reveal whether recorded outcomes differ over time. Because the periods are sequential, however, the comparison is open to other time-related explanations. That is a design consideration, not a judgment about what happened at this hospital.

Routine electronic medical records offer a way to examine outcomes in the context of clinical care, but they also make measurement and data-selection rules consequential. The registry does not describe how readings were selected or how missing or repeated measurements were handled. Such details are important when interpreting averages and their uncertainty.

Finally, a clear hypothesis and a specified primary endpoint do not replace a reported estimate. The registry identifies what investigators intended to test, but does not report the analysis results needed to judge the size or precision of a blood pressure contrast. This distinction is central to statistical reading: study rationale and design describe the question, while estimates and uncertainty are needed to assess what the collected data show.

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.

A key limitation for interpretation is the sequential, nonrandomized comparison. The intervention and calendar period are linked, so changes over time could coincide with training. The registry does not describe a concurrent randomized control group, and it lists allocation as NA. This limits causal interpretation of any before-and-after difference.

The registry record does not report primary outcome results or analyses. Period-specific means and standard deviations, an estimated contrast, confidence intervals, and p-values are absent. This prevents assessment of the direction, magnitude, or precision of the blood pressure findings.

Important data and analysis details are also not reported: how electronic-record readings were selected, whether patients contributed repeated measurements, how missing values were handled, and whether analyses accounted for possible clustering by trainee. These details affect the interpretation of both the estimated mean and its uncertainty. The record also does not report adjustment for changes in patient composition or other measured factors across periods.

The patient data description gives 944 patients during the control period and 1,651 period patients during the intervention period, while overall enrollment is reported as 118. The registry does not define the analysis population or explain how the patient records relate to trainee enrollment. The duration after intervention is described as up to 6 months, so the observation window is not stated as identical in length to the control period. The record does not explain how this difference in stated period duration affected data collection or analysis.

Finally, the registry lists a secondary communication measure but does not report its results or a multiplicity strategy. These omissions limit evaluation of the full outcome set as well as the primary blood pressure question.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: COmmuNity-engaged SimULation Training for Blood Pressure Control — NCT03375918.
• Registry record last updated 2025-09-19; status last verified 2025-08.
• Results first posted 2025-09-19.
• 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.

18. Record Summary

CONSULT-BP (NCT03375918) is a completed sequential study of an educational intervention for healthcare trainees, sponsored by University of Massachusetts, Worcester. The registry lists enrollment of 118, phase NA, allocation NA, and masking as none. The study start date is 2018-08-13, and the primary completion date is 2022-12-31.

The primary outcome is systolic and diastolic blood pressure in mm Hg, summarized as means and standard deviations from electronic medical records. The control period is 6 months before the intervention; the intervention period is up to 6 months after it. The registry also lists the RELATE Checklist as a secondary outcome. It states the hypothesis that patient blood pressure levels would be lower after training, but reports 0 analyses and no primary outcome estimates. The registry record therefore describes the intended question and design without reporting the blood pressure result.