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 Learn, Engage, Act, Process (LEAP) for MOVE! (NCT02825680), a randomized na study. The primary outcome is Rate of MOVE! Visits Per 1000 Candidate Veterans.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02825680 |
| Phase | NA |
| Condition | Obesity |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | None (open label) |
| Enrollment | 137 |
| Arms in the primary outcome | n/a |
| Lead sponsor | VA Office of Research and Development |
| Results posted | Yes |
2. Clinical Question
Population
Veterans Health Administration facilities and teams involved with local MOVE! group weight management programs are the focus. The visit-rate denominator is candidate Veterans living within 40 miles of a facility. The registry lists enrollment of 137, but does not report further details about who or what the enrollment count represents.
Intervention
LEAP is a virtually delivered program for teams. It includes an improvement framework, an online learning platform for sharing ideas and progress, audit and feedback, and a barrier-buster tool. Facilities assigned to LEAP also receive enhanced NCP support.
Comparator
The comparison is enhanced support from the National Center for Health Promotion and Disease Prevention (NCP). This active support is described as being provided to VA MOVE! programs; the comparator facilities in the trial receive enhanced NCP support.
Primary question
Does assignment to LEAP, in addition to enhanced NCP support, change the facility-level rate of MOVE! group visits per 1000 candidate Veterans relative to enhanced NCP support? The outcome is measured over rolling prospective six-month periods after each stepped-wedge cohort timeframe. The registry reports a mixed-model p-value, but not the estimated rate in either group or the size and precision of the difference.
This is a question about program implementation and service use, not an individual-level comparison of weight change or another clinical outcome. The trial description frames the broader goal as bringing local MOVE! teams more in line with guidance and engaging more Veterans in the program. The primary endpoint operationalizes part of that goal by counting group visits relative to the local candidate population. It does not, by itself, indicate whether participants attended repeatedly, whether visits were clinically beneficial, or whether individual health changed.
3. Trial Design
The registry describes a randomized, unmasked trial and identifies its model as crossover. The intervention description also refers to a stepped-wedge design: cohorts of six interested facilities were to be selected each quarter to participate in LEAP. The record does not explain how the crossover model designation relates to the stepped-wedge description, so the precise sequence and timing of assignments cannot be reconstructed from the registry record.
The two conditions are LEAP and enhanced NCP support. The comparator is active rather than an absence of support. Facilities assigned to LEAP receive the same enhanced NCP support as the comparator facilities, as well as the LEAP program. The record says forty-eight sites will be randomized and describes facilities expressing interest in taking part. It lists enrollment of 137, without stating in the result entry whether that number refers to individuals, facilities, or another unit.
The study concerns local VA MOVE! teams and facilities, while the primary measure aggregates visits and candidate Veterans at facility level. This clustered structure matters: assignment and outcome measurement are not described as independent individual-level observations. The registry marks the study completed, lists the start as 2016-10-01 and primary completion as 2019-09-30, and records no masking.
The registry does not report the detailed allocation sequence, the number of facilities in each condition at each measurement period, the crossover schedule, or the exact treatment of facilities that changed status over time. Those details are not reported in the registry record. Consequently, the design can be described at the level of the stated randomization and stepped-wedge account, but not fully recreated from the posted entry.
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
| Stage | Enhanced NCP Support | LEAP Intervention |
|---|---|---|
| STARTED | 82 | 55 |
| COMPLETED | 82 | 39 |
| NOT COMPLETED | 0 | 16 |
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: This study did not collect data about individual patients. The unit of randomization for this study was VA Medical Centers. The unit of measurement is facility aggregated data.
| Characteristic | Unit | Enhanced NCP Support | LEAP Intervention | Total |
|---|---|---|---|---|
| Participants analysed | 0 | 0 | 0 | |
| Sex: Female, Male: Female | — | — | — | |
| Sex: Female, Male: Male | — | — | — |
6. Primary Endpoint
The registry lists the primary outcome as Rate of MOVE! Visits Per 1000 Candidate Veterans, measured over 6 months.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Rate of MOVE! Visits per 1000 Candidate |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the rate of MOVE! visits per 1000 candidate Veterans, with a six-month timeframe. The registry defines the rate as the aggregate number of MOVE! group visits at a facility divided by the aggregate count of candidate Veterans living within 40 miles, multiplied by 1000. Counts are drawn from the Corporate Data Warehouse. The measurement periods are rolling prospective six-month periods after the end of each stepped-wedge cohort timeframe.
This facility-level rate puts service volume in relation to the size of a nearby candidate population. It can describe the level of group-visit activity without simply comparing facilities that may serve different numbers of candidate Veterans. Because the measure is based on aggregate counts, it is not an individual-level measure of whether a particular Veteran enrolled, attended, or benefited. Nor does it identify whether changes in visit activity reflect new participants, repeat visits, or other patterns.
For interpreting a rate comparison, the estimated group rates and their difference would indicate the magnitude and direction of the observed contrast; a confidence interval would express uncertainty around an estimate. The registry describes a 95% confidence interval as the dispersion measure, but the result entry does not report either group mean or confidence limits. Therefore, the endpoint definition is clear, while the amount and precision of the observed rate contrast are not available in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry reports a mixed models analysis for the primary comparison and gives p = .011. This is evidence, under the model used by the study team, that the observed data are not well described by a null hypothesis of no difference between the conditions. The test is reported as a superiority analysis. The result does not state the size or direction of the contrast, because the group means and the model estimate are not reported.
A p-value is not an estimate of practical importance, and it is not the probability that the null hypothesis is true. It does not indicate how many more visits occurred, whether a rate difference was large or small, or whether the result is precise enough to distinguish among plausible effect sizes. Those judgments require an estimate and an interval, neither of which is available in the result entry. Although the endpoint record specifies a 95% confidence interval, the reported analysis contains no confidence limits.
The analysis is described as a mixed model, which is relevant to the facility-level, clustered structure of the trial. The registry does not provide the model specification, the variables included, the covariance structure, or the method used to account for the stepped-wedge timing. Those features are not reported in the registry record. It is therefore not possible to assess from the entry whether particular modeling choices materially affected the estimate or its uncertainty.
The entry does not provide the observed rate in either condition, the number of facilities contributing to each period, or missing-outcome details. It also does not state whether the reported test was adjusted for other outcomes or analyses. The p-value should thus be read narrowly as the reported result of the sponsor's mixed-model comparison, rather than as a complete account of the evidence.
There is no independent unadjusted recalculation to compare with the sponsor's result. A simple individual-level recalculation would not be appropriate for this cluster-randomized setting, because it could treat observations from the same facility as independent and misrepresent uncertainty. The sponsor's own analysis is therefore the relevant reported analysis. No conclusion about the direction, magnitude, or operational importance of the rate contrast can be drawn from the p-value alone.
9. Statistical Methods Explained
The registry names a mixed models analysis for the primary outcome. It does not give the model equation or identify the fixed and random effects, so the particular implementation cannot be checked from the record. In general, mixed models can represent related observations, such as repeated measurements over time or observations grouped within sites, by allowing for dependence rather than assuming every measurement is independent. This is potentially important when facilities, rather than individual Veterans, are assigned to an intervention and when outcomes are observed repeatedly.
Why does clustering matter?
In general, people or measurements within the same facility can share conditions, workflows, staffing, and local context. Their outcomes may consequently be more alike than outcomes from different facilities. If a statistical analysis ignores this dependence, it may overstate the information in the data and produce uncertainty estimates that are too small. A method that accounts for clustering should reflect the unit of assignment and the pattern of repeated observations. The registry identifies the study as cluster-randomized in its analysis note and says an individual-level reanalysis would be misleading. It does not report how the mixed model represented within-facility dependence.
What does the p-value say?
The reported p-value is .011. In general, a p-value describes how incompatible the observed data, or data more extreme, are with a specified null model, assuming that model and its analysis conditions are correct. It is not the probability that the null hypothesis is true, nor does it quantify the size of an effect. Interpretation requires knowing what contrast was tested and how the model was specified. The registry reports a superiority test but does not give the estimated contrast, so the direction and magnitude of the modeled difference cannot be read from the result.
Why are estimates and confidence intervals useful?
In general, a point estimate conveys the observed magnitude of a contrast, while a confidence interval conveys its statistical uncertainty under the chosen method. A narrow interval indicates greater precision than a wide interval, though neither alone determines practical importance. The outcome description identifies a 95% confidence interval as the dispersion measure, but the analysis result gives no group means or interval bounds. The p-value alone cannot substitute for those missing quantities.
What can be said about reanalysis?
The registry's analysis note excludes a simple individual-level recalculation because of the cluster-randomized design. This is a reasonable caution in general: aggregate summaries or analyses that fail to respect the assignment structure may not reproduce the trial's valid uncertainty estimate. No unadjusted recalculated result is reported. Where the record supplies no estimates or model details, an outside reader cannot reconstruct the sponsor's model-based result from the entry.
What do planning calculations represent?
In general, a sample-size calculation asks how much information a future study might need to detect a specified effect under stated assumptions. The planning illustration described for this analysis uses a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects stated as fractions of the observed effect. These are planning assumptions, not a description of the original trial's design or evidence that the trial had a particular probability of detecting an effect. The registry does not report the study's own sample-size justification. Such illustrations must not be interpreted as observed power or as proof that the trial was adequately powered.
Why do repeated periods need careful handling?
The endpoint uses rolling six-month periods, and the study description refers to stepped-wedge cohorts. In general, when intervention exposure changes over time across clusters, outcome trends and calendar time can be intertwined with intervention status. A suitable analysis may need to represent time and the assignment sequence. The record does not report the exact timing variables, model structure, or assumptions used for this purpose. It therefore supports reporting the sponsor's stated mixed-model result, but not independently evaluating whether every feature of the design was represented correctly.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | not reported | 0.011 |
The posted result is the sponsor's mixed-model analysis, with p = .011 for the superiority comparison. No group means, estimated contrast, or confidence interval are reported alongside that test. The registry's analysis note states that a simple individual-level reanalysis would be misleading because the study is cluster-randomized, and that only the sponsor's own analysis is shown.
Accordingly, there is no unadjusted recalculation to compare with the sponsor's result. A recalculation based on individual-level summaries could fail to account for the facility-level assignment and dependence among observations, giving a misleading impression of precision. The sponsor's own analysis takes precedence in this setting. That does not make the result fully assessable: without the estimate, interval, model specification, or facility-level information, the size and precision of the contrast cannot be independently examined from the registry record.
The reported p-value supports only a statement about the sponsor's test under its model. It cannot establish the direction or practical magnitude of a difference. The registry does not report enough detail to reproduce the model-based analysis.
11. Multiplicity, Missing Data and Analysis Populations
The registry reports one outcome measure and one analysis entry. It does not state whether other outcomes or analyses were considered, whether the primary test was selected before results were examined, or whether any adjustment for multiple testing was used. These matters are not reported in the registry record. A single listed analysis does not establish that only one analysis was planned or conducted.
Missing-data handling is also not described. The endpoint is constructed from facility-level aggregate visit counts and candidate-Veteran counts, but the record does not report whether any facilities or periods lacked data, how incomplete counts were addressed, or whether any imputation or exclusion rules were applied. These details are not reported in the registry record. Without them, it is not possible to assess the impact of incomplete outcome data on the comparison.
The registry does not describe the analysis population, such as whether all assigned facilities were analyzed according to assignment, whether facilities could be excluded, or how changes in participation were handled. It also does not explain how the stepped-wedge timing, repeated six-month periods, and cluster structure were incorporated in the mixed model. Those details are not reported in the registry record.
In general, multiplicity can increase the chance of finding at least one apparently noteworthy result when many hypotheses are tested. Missingness can affect both the estimate and its precision, depending on why observations are absent and how they are handled. These are general statistical considerations, not claims that a particular problem occurred in this trial. The available registry entry does not permit a detailed appraisal of multiplicity, missingness, or analysis-population choices.
12. What the Result Does, and Does Not, Mean
The reported finding is a mixed-model superiority test with p = .011 for the facility-level MOVE! visit-rate comparison. It indicates that the sponsor's model found evidence against its null comparison, conditional on the assumptions and methods used. The posted result does not provide the group rates, estimated contrast, direction, or confidence interval, so it cannot show how large or precise the observed difference was.
The endpoint measures aggregate group visits relative to the nearby candidate-Veteran population. It does not measure individual weight change, health outcomes, program completion, or the quality of each visit. Nor does a facility-level rate by itself show whether more unique Veterans engaged, because the registry definition counts visits and does not separate first visits from repeat attendance.
The intervention contrast is LEAP plus enhanced NCP support versus enhanced NCP support. Any interpretation therefore concerns the incremental comparison between those conditions, not LEAP versus no support. The trial's broader stated aim includes helping teams align with guidance and engaging more Veterans, but the primary outcome directly captures visit activity rather than every aspect of that aim.
These findings do not establish that the intervention did or did not produce a clinically important benefit. The registry does not report enough numerical detail to judge practical importance, and the service-use endpoint does not directly measure individual clinical benefit. The p-value alone is not a measure of effect size, value, or certainty about an intervention's consequences.
13. Why This Trial Matters Statistically
This trial raises a useful statistical issue: the unit of intervention is the facility or team, while the outcome is expressed as a facility-level rate. Analyses must respect that structure. Treating individual-level observations as independent can make an analysis appear more precise than it is when outcomes within the same facility are related. The registry explicitly cautions against a simple individual-level reanalysis and reports the sponsor's mixed-model result instead.
The design description also emphasizes the importance of time. Facilities were to participate in cohorts under a stepped-wedge account, and the endpoint uses rolling six-month periods. In general, intervention exposure and calendar time can change together in such designs, making the handling of time an important part of the analysis. The registry identifies the model family but does not provide enough detail to assess how these features were represented.
The primary measure has a clear service-delivery interpretation: aggregate MOVE! group visits relative to the nearby candidate population. At the same time, the measure does not reveal the number of individual Veterans reached or their clinical outcomes. That distinction is important when interpreting program activity as evidence about engagement or health.
Finally, the result illustrates why a p-value should be reported with an estimate and interval. The registry's p-value conveys the test result, but the absence of group rates and confidence limits prevents readers from judging direction, magnitude, and precision. A statistically informative test is not a complete account of the size or practical meaning of an outcome.
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 most consequential reporting limitation is that the primary result contains a p-value but no group means, estimated contrast, or confidence limits. Although the endpoint description specifies a 95% confidence interval, the result does not give its bounds. The direction and magnitude of the rate comparison therefore cannot be determined from the registry result.
The design is described both as crossover in the model field and as stepped-wedge in the intervention and endpoint descriptions. The record does not explain the relationship between these labels. It also does not report the detailed assignment sequence, the timing of facility transitions, or the exact number of facilities contributing to each period. This restricts independent assessment of the design and analysis.
Although a mixed models analysis is named, the statistical model, covariates, assumptions, and treatment of repeated periods are not specified. The registry does not describe missing data, exclusions, analysis populations, or multiplicity handling. The enrollment count is listed without explaining its unit, and cannot be assumed to identify the number of facilities or individual Veterans in the analysis.
The outcome is an aggregate visit rate. It does not distinguish unique Veterans from repeat visits and is not a direct measure of individual clinical outcomes. The record also does not report adverse-event terms; the listed event counts are zero for the two groups, but details of ascertainment are not provided. These reporting limits do not establish that an unreported issue occurred; they limit what can be concluded from the registry entry.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Learn, Engage, Act, Process (LEAP) for MOVE! — NCT02825680.
- Registry record last updated 2023-07-27; status last verified 2023-07.
- Results first posted 2020-10-12.
- 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
NCT02825680 evaluates LEAP, a virtually delivered support program for VA MOVE! teams, against enhanced NCP support. The registry describes randomization and a stepped-wedge cohort approach, while also labeling the model crossover. The primary endpoint is the facility-level rate of MOVE! group visits per 1000 candidate Veterans over rolling six-month periods.
The sponsor reports a mixed-model superiority test with p = .011. Group rates, the estimated contrast, and the stated 95% confidence interval are not reported in the result entry. A simple individual-level recalculation is not presented because the cluster-randomized design makes it potentially misleading. The result therefore documents the sponsor's test, but does not establish the magnitude, direction, or precision of the rate contrast.