Exercise, Brain, and Cardiovascular Health: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Exercise, Brain, and Cardiovascular Health (NCT03841669), a randomized na study. The primary outcome is Change in Hippocampal Volume.

2. Clinical Question

The study, titled “Exercise, Brain, and Cardiovascular Health,” is described as an investigation of aerobic exercise effects on brain structure and function, and of how training-related effects relate to cardiovascular function through related brain changes. The primary endpoint focuses on hippocampal volume measured by structural magnetic resonance imaging. The posted results report a least-squares mean change for each group, rather than a direct between-group treatment contrast.

The registry also identifies changes in systolic blood pressure responses to behavioral tasks and heart rate variability as secondary outcomes. These are distinct from the primary brain-volume endpoint. The main statistical question is therefore narrower than whether exercise affects cardiovascular or brain health in general: it concerns the estimated change in hippocampal volume over the stated follow-up, along with the uncertainty around each arm-specific estimate. A direct comparison of changes between the groups is not reported in the registry record.

3. Trial Design

The registry describes a randomized, double-masked, parallel-group behavioral intervention. The study is titled “Exercise, Brain, and Cardiovascular Health,” and its official title is “Behavioral Studies of Cardiovascular Disease.” The sponsor is the University of Pittsburgh. Its phase is recorded as NA, and the status is completed. Enrollment is listed as 130.

Participants were assigned to an Aerobic Exercise Group or a Physical Activity & Health Information Group. The exercise program prescribed 150 min/week. Supervision could take place remotely or in a laboratory setting because of COVID-19 restrictions. The comparator group engaged in daily life monitoring every 6 weeks. The registry does not describe the specific randomization procedure or how masking was implemented for a behavioral intervention.

The primary outcome is change in hippocampal volume, measured from baseline to 12 months. Structural MRI images were to be processed using segmentation algorithms to compute hippocampal volume. The registry describes the outcome as a continuous measure in mm^3, summarized with a least-squares mean and a 95% confidence interval. It identifies all randomized participants as included in intention-to-treat analyses.

Two secondary outcomes are identified: change in systolic blood pressure responses to behavioral tasks and change in heart rate variability. The record lists 23 outcome measures, but does not give a complete statistical analysis plan. In particular, the planned statistical model, the approach to missing measurements, the masking procedures, and any strategy for handling multiple outcomes are not reported in the registry record. The study start is listed as 2019-05-15, and primary completion is listed as 2024-02-03.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: All randomized participants were included in intention-to-treat (ITT) 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.

6. Primary Endpoint

The registry lists the primary outcome as Change in Hippocampal Volume, measured over Baseline & 12 months.

Hippocampal volume is a brain-structure measure obtained from structural magnetic resonance imaging. In this trial, segmentation algorithms applied to MRI images were used to compute the volume of the hippocampus. The primary endpoint concerns change in that volume between baseline and 12 months, expressed in mm^3.

This continuous endpoint represents a measured quantity rather than a yes-or-no event. Reporting change allows the analysis to describe whether the measured volume moved upward or downward over the follow-up period, while the least-squares mean provides an adjusted mean estimate under the analysis model. The registry does not report the model or its adjustment variables, so the details of that adjustment cannot be assessed from the record.

Hippocampal volume was selected as the primary brain-structure outcome within a study examining exercise, brain structure and function, and cardiovascular function. A structural measure can address a different aspect of the study question than behavioral-task blood pressure responses or heart rate variability, which are named as secondary outcomes. The endpoint alone does not measure all aspects of brain function or cardiovascular health, and a change in volume does not by itself establish why a change occurred.

The registry reports arm-specific least-squares mean changes and confidence intervals. It does not report a direct contrast between the randomized groups for this endpoint. As a result, the estimates describe the mean change within each group, but do not by themselves quantify the randomized treatment difference.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: it measures hippocampal-volume change in mm^3. The registry summarizes it with least-squares means and 95% confidence intervals for each randomized group. The record does not give a direct estimate of the difference between groups, a p-value, or the detailed model used to calculate the reported means.

What does a least-squares mean represent?

In general, a least-squares mean is a model-based mean estimate. Depending on the model, it can account for specified covariates or other design features, unlike a simple arithmetic average of observed values. The registry record does not state which covariates or model terms were used. It is therefore appropriate to report the values as least-squares means, but not to claim a particular adjustment or statistical model.

How should the confidence intervals be read?

In general, a confidence interval expresses uncertainty around an estimate using a method with a stated long-run coverage property under its assumptions. A narrower interval indicates greater precision than a wider interval, while an interval that includes zero is compatible with no mean change as well as changes in either direction. For the exercise group, the reported interval is -79.3 to 48.4 mm^3 around -15.4 mm^3. For the comparison group, it is -112.8 to 13.0 mm^3 around -49.9 mm^3. Both include zero, and both allow a range of possible changes. The interval alone does not establish the probability that a particular value is true.

Why is the group contrast important?

Randomization makes the difference between assigned groups the central comparative quantity. Arm-specific changes describe what was estimated within each group, but a treatment comparison requires an estimate of how those changes differ, together with uncertainty for that contrast. In general, uncertainty for a contrast depends on the joint analysis and cannot be recovered reliably by treating two separate confidence intervals as though they were a formal comparison. The registry record does not report a between-group contrast for hippocampal-volume change, so one should not infer a treatment-effect estimate from the two arm-specific means.

What does intention-to-treat describe?

In general, an intention-to-treat analysis aims to analyze participants according to their randomized assignment, regardless of adherence or changes in treatment received. The registry states that all randomized participants were included in ITT analyses. This identifies the intended analysis population, but it does not explain how measurements that were missing at follow-up were treated. It also does not confirm that every randomized participant contributed an observed hippocampal-volume measurement to the reported estimate.

Why do assumptions matter?

Model-based estimates and confidence intervals rely on assumptions. For a continuous outcome, those can include assumptions about the form of the outcome distribution, the relationship between predictors and the outcome, and how measurements are correlated when repeated over time. The registry does not report the model diagnostics or assumptions used for this endpoint. In general, if observations have greater variability than a model allows, uncertainty estimates may be too narrow unless the analysis accounts for that extra variability. This concept is often called over-dispersion, although its relevance depends on the outcome and model; the registry does not indicate that over-dispersion was assessed for this continuous endpoint.

The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is described as a stated fraction of the observed effect. These assumptions describe planning calculations, not observed trial results. The registry record does not report the resulting planning estimates or the trial’s own sample-size justification. Such illustrations should not be described as observed power. In general, post hoc power calculations based on an observed estimate do not replace direct interpretation of the effect estimate and its confidence interval.

What can be learned from the posted estimates?

The estimates support a descriptive account of the arm-specific changes and their precision. The registry reports -15.4 mm^3 for the Aerobic Exercise Group and -49.9 mm^3 for the Physical Activity & Health Information Group, each with a 95% confidence interval. Without a reported contrast and its uncertainty, the comparative effect remains unquantified in the registry record. Statistical significance, if assessed for a between-group effect, cannot be determined from these arm-specific summaries alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 23 outcome measures and identifies two secondary outcomes: change in systolic blood pressure responses to behavioral tasks and change in heart rate variability. The primary endpoint is change in hippocampal volume. Because several outcomes are considered, multiplicity can matter: in general, conducting multiple hypothesis tests without an adjustment or prespecified hierarchy can increase the chance of at least one apparently notable result arising by chance. The registry record does not report a multiplicity adjustment, a testing hierarchy, or a statistical plan specifying how outcomes were to be interpreted together.

The record states that all randomized participants were included in intention-to-treat analyses. This clarifies the stated analysis population, but does not specify the handling of missing measurements, incomplete follow-up, or participants with no usable MRI measurement. The approach to imputation, complete-case analysis, or sensitivity analyses is not reported in the registry record. ITT wording alone does not establish that missing outcomes were imputed or that every participant contributed data to every endpoint.

The registry lists zero analysis entries, while it also contains posted primary outcome estimates. No detailed analysis plan, model specification, or p-value for the primary comparison is reported. As a result, the record does not allow readers to assess how missing data, multiple outcomes, or analysis choices were handled in the sponsor’s inferential analysis. These gaps do not change the numerical estimates, but they constrain the conclusions that can be drawn from them.

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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: tumor (1.6%).

The registry reports serious adverse event figures separately for the two groups. In the Aerobic Exercise Group, 1 of 64.0 was recorded as affected by a serious adverse event; in the Physical Activity & Health Information Group, the reported figure was 0 of 66.0. The top serious-event term listed is “Tumor,” at 1.6%. Deaths are reported as 0 in both groups.

Other adverse events are reported for 6 of 64.0 in the Aerobic Exercise Group and 0 of 66.0 in the comparison group. These figures describe what the registry records; they do not establish that an event was caused by an intervention. The registry does not provide event narratives, timing, severity details beyond the serious-event classification, or a statistical comparison of safety outcomes.

Safety counts should be read in the context of the groups and the duration and completeness of follow-up, details that may affect event ascertainment. The registry figures do not establish that no unrecorded events occurred, and zero recorded events in a group is not proof that risk is absent. The reported adverse-event information is descriptive, not a causal assessment.

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

The primary results describe estimated mean hippocampal-volume change from baseline to 12 months in each randomized group. The Aerobic Exercise Group estimate is -15.4 mm^3, with a 95% confidence interval from -79.3 to 48.4. The Physical Activity & Health Information Group estimate is -49.9 mm^3, with a 95% confidence interval from -112.8 to 13.0. Both estimates are negative, and both intervals include zero.

These findings do not establish that either group experienced a true decline, because the uncertainty intervals also include values on the other side of zero. They do not establish that the interventions are equivalent, because equivalence requires an appropriate comparison and a prespecified margin, neither of which is reported in the registry record. Nor do the arm-specific values establish a difference between groups. A direct between-group estimate and its uncertainty are not reported.

The results address one measured aspect of brain structure. They do not, on their own, answer every question about brain function or cardiovascular function. The registry names changes in systolic blood pressure responses to behavioral tasks and heart rate variability as secondary outcomes, but those outcomes are not detailed in the primary estimates. An interpretation of those outcomes would require their own posted estimates and methods.

The registry describes an intention-to-treat population, but does not report how missing measurements were handled. The statistical model, adjustment variables, multiplicity strategy, and detailed analysis plan are also not reported in the registry record. These omissions limit assessment of the precision and robustness of the posted estimates beyond the confidence intervals shown.

Accordingly, the results are best read as arm-specific estimates with substantial uncertainty, rather than a conclusive comparative finding. They do not show that exercise caused a particular brain-volume change, and they do not show that the comparison group’s activities caused its estimated change. They also cannot establish a clinical benefit or harm. Any broader interpretation would need to consider the comparative analysis, the secondary outcomes, safety information, and the reporting gaps in the registry record.

13. Why This Trial Matters Statistically

This trial illustrates why an arm-specific mean and a randomized treatment comparison answer different questions. The posted primary results give a mean change for each group, but the central randomized comparison is the difference between the groups. Without that contrast and its confidence interval, readers cannot judge the estimated comparative effect directly. The two within-group estimates should not be treated as a substitute.

It also demonstrates the importance of uncertainty. Both reported 95% confidence intervals include zero and span a broad range of possible changes. A point estimate alone can appear more definite than the evidence warrants; the interval makes the imprecision visible. General statistical interpretation requires attention to both the size and direction of an estimate and to how precisely it is measured.

The record identifies a continuous MRI-based primary endpoint, two secondary outcomes, and 23 outcome measures. When a study examines multiple outcomes, transparent reporting of the prespecified testing strategy matters for interpreting isolated results. The multiplicity strategy is not reported in the registry record, limiting evaluation of how findings across outcomes were meant to be weighed.

Finally, an ITT population statement is useful but does not answer every question about analysis. It does not specify missing-data handling, the statistical model, or adjustments. Clear reporting of those elements would help readers assess how the randomized design connects to the estimates. The posted record offers meaningful descriptive information, while leaving important parts of the comparative analysis unreported.

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 primary reporting limitation is that the registry gives arm-specific least-squares mean changes but no direct between-group estimate, confidence interval, or p-value for hippocampal-volume change. Since the trial is randomized, the treatment contrast is central to interpreting comparative effects. The arm-specific intervals cannot be used as a formal test of that contrast.

The record does not report the model specification, covariate adjustments, or model diagnostics. It also does not explain how missing MRI measurements or other incomplete outcome data were handled. Although all randomized participants are described as included in ITT analyses, the record does not state how that aim was implemented when an outcome measurement was unavailable.

Multiplicity handling is not reported, despite the registry listing 23 outcome measures and identifying secondary outcomes. The analysis plan and the rationale for the sample size are also not reported. A planning illustration is described, but its resulting estimates are not reported in the registry record; it should not be confused with a justification for the trial’s enrollment or with observed power.

The intervention description notes that exercise supervision could occur remotely or in a laboratory setting because of COVID-19 restrictions. The record does not report how many participants received supervision in either setting or whether setting affected adherence or outcomes. Nor does it provide details about adherence to the prescribed exercise program or the comparator’s monitoring beyond the stated schedule.

Finally, the registry’s safety data are descriptive and do not include a statistical comparison or event narratives. Together, these reporting gaps mean the posted results offer a limited view of the analysis and its sensitivity to assumptions. They do not negate the reported estimates, but they restrict the strength and scope of interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Exercise, Brain, and Cardiovascular Health — NCT03841669.
• Registry record last updated 2025-02-24; status last verified 2025-02.
• Results first posted 2025-02-24.
• 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

The completed eBACH study was a randomized, double-masked, parallel-group behavioral trial sponsored by the University of Pittsburgh. It compared an aerobic exercise program prescribed for 150 min/week with a physical activity and health information group. The registry lists enrollment as 130.

The primary endpoint was change in hippocampal volume from baseline to 12 months, measured using structural MRI and summarized with arm-specific least-squares means and 95% confidence intervals. The exercise-group estimate was -15.4 mm^3, with a 95% confidence interval from -79.3 to 48.4; the comparison-group estimate was -49.9 mm^3, with an interval from -112.8 to 13.0. A direct between-group estimate is not reported.

The registry identifies changes in systolic blood pressure responses to behavioral tasks and heart rate variability as secondary outcomes. The analysis plan, missing-data approach, and multiplicity strategy are not reported in the registry record. Safety figures are descriptive and include 0 deaths in both groups.