Mindfulness-Based Stress Reduction, Health Education and Exercise: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Mindfulness-Based Stress Reduction, Health Education and Exercise (NCT02665481), a randomized na study. The primary outcome is Memory Composite Score.

2. Clinical Question

The central statistical question concerns both intervention assignment and the trajectory of memory scores over the specified follow-up. The record labels the study model factorial and lists four groups, but the precise primary contrasts and whether the analysis tested interaction between interventions are not reported in the registry record. Consequently, the comparison cannot be reduced to a single treatment-versus-control estimate from the record. It is also not possible to determine from the posted results whether the groups differed at particular assessments or in their patterns of change.

3. Trial Design

This was a randomized, single-masked, factorial-model trial with an enrollment of 585. It is listed as completed, with a start date of 2015-04 and primary completion in 2020-06. Washington University School of Medicine is identified as the sponsor. The phase is listed as NA.

The four named groups were MBSR, exercise, MBSR plus exercise, and health education. MBSR involved an introductory meeting, eight weekly 2.5-hour classes, a retreat, and monthly booster sessions for approximately 15 months. Exercise involved classes twice weekly for 6 months, building up to 1.5 hr, under direct supervision. The combined group was described as receiving both programs, with exercise and mindfulness activities also undertaken at home. Health education involved ten weekly 2.5-hour classes followed by monthly classes for approximately 15 months.

Random allocation means assignment is intended to create comparable groups at the start, on average. The record does not report the allocation ratio, sequence-generation procedure, concealment method, or which people were masked. “Single” identifies the masking category but the identity of the masked party is not reported in the registry record. The detailed factorial structure, including the specific factors and planned contrasts, is also not reported. The group descriptions alone do not establish which statistical model was used to represent the factorial design.

The primary outcome was a memory composite score measured at Month 0, Month 3, Month 6, and Month 18. The primary outcome population is described as modified intent-to-treat, including participants who received at least one dose of the intervention. This differs from an analysis that necessarily retains every randomized participant, so the details of participant inclusion and exclusions matter for interpreting the results. Those details 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: modified Intent-to-Treat, which includes all participants that received at least one dose of the intervention.

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 Memory Composite Score, measured over Month 0, Month 3, Month 6, and Month 18.

The primary endpoint is the Memory Composite Score, expressed as a Z-score. It combines list recall, paragraph recall, and picture sequence memory. For each task, the participant’s score is standardized using the stated formula of participant score minus mean, divided by standard deviation; the task-level Z-scores are then averaged to form the composite.

The record states that a Z-score of 0 represents the population mean and that a higher score indicates a better outcome. Combining several memory tasks can provide a broader summary of memory performance than any one task alone, while standardization puts the task scores onto a common scale before averaging. These are general measurement principles; the record does not explain how the reference mean and standard deviation were chosen or provide evidence about the composite’s measurement properties.

The outcome is listed as a least-squares mean with standard error at Month 0, Month 3, Month 6, and Month 18. In general, a least-squares mean is a model-based mean; the exact model and any covariates are not reported in the registry record. A standard error describes uncertainty in an estimated mean, rather than the variability among individual participants. No primary endpoint estimates, standard errors, confidence intervals, or group comparisons are reported in the registry record. The outcome’s direction and definition are therefore clear, but its observed results cannot be assessed from the posted record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does randomization contribute?

In general, random assignment helps make groups comparable with respect to both measured and unmeasured characteristics at baseline. It supports causal interpretation when the assigned groups are analyzed appropriately and when losses, nonadherence, and other post-randomization events are handled transparently. Randomization does not guarantee exact balance in a particular trial, nor does it remove uncertainty. The record states that allocation was randomized, but the sequence-generation and concealment procedures are not reported in the registry record.

What does the factorial label tell the reader?

The study model is labelled factorial and the record names MBSR, exercise, their combination, and health education. In general, factorial designs can be used to examine intervention effects individually and, when planned, whether effects depend on combining interventions. The precise factors, contrast definitions, and interaction analysis are not reported in the registry record. The label alone is not enough to determine how the analysis represented the four groups. It is therefore important not to interpret the group names as evidence that any particular main-effect or interaction estimate was reported.

How should least-squares means be understood?

In general, a least-squares mean is an estimate from a statistical model rather than necessarily the raw arithmetic average of the observed scores. Depending on the model, it may account for other variables or repeated measurements. The registry record does not report the model specification, covariates, or exact method used to handle repeated outcome measurements. As a result, the meaning of any adjustment cannot be evaluated from the record. The outcome is specified as a least-squares mean with a standard error, but the corresponding numerical estimates are not reported in the registry record.

What does the standard error describe?

In general, a standard error describes the precision of an estimated quantity across repeated samples under the assumptions of the analysis. It is not the same as the standard deviation, which describes variation among individual observations. An interval estimate constructed using a standard error can convey a range of plausible values, subject to the method and assumptions. The record does not report the primary standard errors or confidence intervals, so the precision of the group means and comparisons cannot be assessed.

Why does the timing of measurement matter?

The primary outcome is assessed at Month 0, Month 3, Month 6, and Month 18. Repeated assessments can distinguish an initial score from later measurements and can describe whether groups follow different trajectories. In general, analyses of repeated measurements need to account for the relationship between measurements from the same participant. The model, correlation structure, handling of missed assessments, and planned time-specific contrasts are not reported in the registry record. Without those details and the estimated results, it is not possible to determine how change over time was modeled or to reproduce the analysis.

What can planning calculations tell us?

In general, a sample-size planning illustration describes the number of participants that might be needed under specified assumptions to detect a chosen effect with a target probability of finding statistical evidence at a chosen significance level. Such an illustration is conditional on its assumptions and is not a measure of the probability that the completed trial detected an effect. It is also not observed power. The record describes its planning illustrations as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. These calculations are explicitly planning illustrations, not the trial’s own design justification. The sample-size justification used when designing the trial is not reported in the registry record.

In general, an observed-effect-based planning calculation is sensitive to the estimate used as its starting point. If an observed effect is uncertain, a planning result based on it is uncertain as well. The calculation should not be read as proof that the trial was adequately powered for every outcome or contrast. The record does not report a design-stage target or the assumptions used to select the enrollment of 585, so those should not be inferred from a planning illustration.

Why might over-dispersion matter?

In general, over-dispersion means that observed variability is greater than a statistical model’s assumptions allow. When a model underestimates variability, its standard errors can be too small and its intervals too narrow. Appropriate modeling can address this, but the need and method depend on the outcome and model. The primary endpoint is a continuous memory composite, and the record does not report an over-dispersion assessment or a specific correction. This general concept should not be taken to mean that over-dispersion was observed in this trial.

What is missing for independent appraisal?

A full statistical appraisal would ordinarily require the analysis model, prespecified contrasts, missing-data approach, treatment of protocol deviations, and numerical effect estimates with uncertainty. The registry record does not report these details for the primary endpoint. The information available therefore permits a clear description of the endpoint and the broad design, but not an independent reconstruction or verification of the primary analysis. No interim-analysis plan or stopping rule is reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The trial includes four intervention groups, repeated primary outcome assessments at Month 0, Month 3, Month 6, and Month 18, 7 outcome measures, and 28 analyses. When multiple outcomes, groups, time points, or contrasts are examined, the chance of at least one apparently notable result can increase. In general, multiplicity procedures can control that risk by adjusting the interpretation or testing strategy. The registry record does not report whether any adjustment was used, which analyses were included in any adjustment, or which contrasts were designated as primary. These omissions prevent a clear assessment of how the analysis set was managed.

The primary outcome population is modified intent-to-treat and includes participants who received at least one dose of the intervention. This definition does not include, by definition, randomized participants who received no dose. The number and characteristics of any excluded participants are not reported in the registry record. The handling of participants who started treatment but missed later assessments is also not reported.

In general, missing outcome data can reduce precision and can bias estimates if the probability of missingness is related to outcome values or other relevant factors. Statistical methods such as imputation or likelihood-based approaches can rely on assumptions about the missingness process; sensitivity analyses can examine how conclusions change under alternative assumptions. The registry record does not report the amount or pattern of missing primary outcome data, the method used to address missingness, or sensitivity analyses.

The record also does not report whether the analysis followed an intention-to-treat principle beyond the stated modified intent-to-treat definition, how participants were analyzed if their treatment differed from assignment, or how protocol deviations were handled. These details are relevant to interpreting randomized comparisons and cannot be inferred from the population label alone.

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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: cardiovascular event (2.9%), tumor/cancer (2.7%), injury (2.2%), transient ischemic attack (tia) (1.4%), laminectomy (1.4%).

The registry record reports serious adverse event counts by group. For MBSR, 18 participants were affected among 150 at risk, and 1 death is listed. For exercise, 16 were affected among 138 at risk, with 0 deaths. For MBSR plus exercise, 14 were affected among 144 at risk, with 0 deaths. For health education, 5 were affected among 153 at risk, with 0 deaths. These are counts as reported; no rate comparisons have been calculated.

The record lists 0 other adverse events in each group. It also lists the serious adverse event terms Cardiovascular Event at 2.9%, Tumor/Cancer at 2.7%, Injury at 2.2%, Transient Ischemic Attack (TIA) at 1.4%, Laminectomy at 1.4%, and Bowel Obstruction at 1.4%. These term percentages are reported as such, without additional denominator or attribution details in the record.

In general, an adverse event occurring during a trial is not necessarily caused by the assigned intervention. Counts can also be influenced by group size, duration of observation, event definitions, and how events are collected. The record does not describe the event-collection procedures, severity grading beyond the serious-event classification, relatedness assessment, or comparative safety analysis. The figures are therefore descriptive and should not be interpreted as establishing a causal difference between interventions.

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

The registry record establishes what the trial set out to compare: MBSR, exercise, their combination, and health education in healthy older adults studied for age-related cognitive decline. It also defines the primary outcome as an average of standardized scores from list recall, paragraph recall, and picture sequence memory. The outcome’s stated direction is clear: a higher Z-score indicates better performance, and 0 represents the population mean.

What the record does not establish is whether any group had higher or lower memory scores, whether scores changed across assessments, or whether any observed group differences were compatible with chance. The primary outcome estimates, uncertainty measures, and between-group comparisons are not reported in the registry record. No conclusion about the magnitude, precision, or statistical significance of the primary outcome can therefore be drawn from the posted information.

This distinction matters because a study’s question and design do not themselves answer the question. Randomization supports a comparative design, but the results and analysis details are needed to understand what was observed. Likewise, a completed status confirms the trial is listed as completed, not that a particular result was obtained or that an intervention was effective or ineffective.

The available adverse-event counts provide descriptive safety information, but do not show that an event was caused by an intervention or establish comparative safety. The registry also lists secondary outcomes, but their estimates are not reported in the record. Further interpretation would require the outcome results, their uncertainty, details of the analysis and missing-data handling, and information about how multiple comparisons were addressed.

Accordingly, the most defensible reading is limited: the study’s groups, primary measurement, assessment schedule, analysis population definition, and some serious adverse event counts are described; the primary comparative findings are not reported in the registry record. The absence of those findings should not be interpreted as evidence of either benefit or lack of benefit.

13. Why This Trial Matters Statistically

This trial is statistically informative as a design because it combines randomized assignment with multiple behavioral intervention groups and repeated measurement of a cognitive outcome. The factorial label raises questions about whether investigators sought to separate the effects of MBSR and exercise or examine their combination, but the planned contrasts and model are not reported in the registry record. This makes the trial a useful example of why design labels and intervention descriptions do not replace a clear account of the estimand: the precise effect or comparison the analysis is intended to estimate.

The memory composite also illustrates a measurement choice. It combines standardized scores from three tasks, placing them on a shared scale before averaging. That approach can make a multi-task outcome easier to interpret as a single measure, while still depending on the tasks, standardization reference, and assumptions behind the composite. The record describes the construction and direction of the score, but not the reference values or measurement properties.

Repeated assessment creates another important statistical issue. Comparisons may concern scores at specific times or differences in trajectories, and repeated measurements from a participant are generally related. The record supplies the assessment schedule, but not the repeated-measures model, time contrasts, or missing-data methods. The registry also lists 7 outcome measures and 28 analyses, making transparency about multiplicity particularly relevant; the adjustment strategy is not reported.

Finally, the distinction between an observed result and a planning illustration is central. The record’s sample-size illustrations are based on the observed effect and specified assumptions, rather than serving as the trial’s design justification or a measure of observed power. Overall, the trial highlights how interpretation depends not only on randomization and a defined endpoint, but also on reporting the estimates, uncertainty, model, analysis population, and multiplicity strategy.

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 limitation for statistical interpretation is that the registry record does not report the primary memory outcome estimates, group comparisons, confidence intervals, standard errors, or p-values. The outcome is described at multiple assessment times, but the record does not state which time-specific or trajectory contrasts were primary. Consequently, neither the direction nor the precision of the primary findings can be evaluated.

The record labels the design factorial but does not describe the factors, planned contrasts, or interaction model. It reports randomized allocation and single masking, but not the allocation procedure or who was masked. The statistical model, covariates, and method for accounting for repeated measurements are also not reported in the registry record. These gaps limit assessment of how the analysis corresponds to the intended design.

The analysis population is defined as modified intent-to-treat participants who received at least one dose. The number excluded under that definition, reasons for exclusion, and handling of later missing measurements are not reported. Missingness can affect both precision and bias, depending on its pattern and the assumptions used to handle it. No imputation method, sensitivity analysis, or protocol-deviation approach is described.

Multiplicity is a further concern because the record lists 7 outcome measures and 28 analyses, but does not report a testing hierarchy or adjustment strategy. The planning illustrations are based on the observed effect, and should not be mistaken for the trial’s original sample-size justification or observed power. The design-stage justification is not reported in the registry record.

Finally, the serious adverse event counts are descriptive, and the record does not report collection procedures, attribution assessments, or comparative safety analyses. These limitations do not demonstrate that the trial’s analyses were inappropriate; they identify information not reported in the registry record that is needed for independent statistical interpretation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Mindfulness-Based Stress Reduction, Health Education and Exercise — NCT02665481.
• Registry record last updated 2022-09-28; status last verified 2022-09.
• Results first posted 2022-09-28.
• 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

NCT02665481, titled “Mindfulness-Based Stress Reduction, Health Education and Exercise,” is a completed, randomized, single-masked trial with a factorial model and enrollment of 585. Washington University School of Medicine is listed as sponsor. The study concerns age-related cognitive decline in healthy older adults and compares MBSR, exercise, their combination, and health education.

The primary outcome is the Memory Composite Score, a Z-score combining list recall, paragraph recall, and picture sequence memory. It is assessed at Month 0, Month 3, Month 6, and Month 18. The stated population is modified intent-to-treat, including participants who received at least one dose.

The registry record reports serious adverse event counts by group but does not report the primary outcome estimates or comparisons. The statistical model, missing-data methods, multiplicity approach, and design-stage sample-size justification are not reported in the registry record.