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 Long-Term Effects of Torso-Weighting (NCT02743312), a randomized na study. The primary outcome is Steps Per Day.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02743312 |
| Phase | NA |
| Condition | Multiple Sclerosis |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Quadruple |
| Enrollment | 5 |
| Arms in the primary outcome | n/a |
| Lead sponsor | San Francisco State University |
| Results posted | Yes |
2. Clinical Question
Population
Five volunteers with multiple sclerosis were enrolled in this pilot study. The registry record describes the study as testing a protocol for examining longer-term effects of torso weighting, rather than reporting an analyzed comparison of participant outcomes.
Intervention
Participants wore a garment with torso weights for 2-4 hours daily during a 2-week period. They wore a Fitbit Flex throughout the study, including the period before weighting.
Comparator
The comparison was a garment with sham weights, also worn for 2-4 hours daily during a 2-week period. The sequence varied: one arm received torso weights before sham weights, and the other received sham weights before torso weights.
Primary question
The study aimed to test whether torso weighting changed daily physical activity, measured as steps per day. The stated hypotheses also concerned falls and muscle activation, while the registry lists additional balance, walking, and function measures. Because the server storing activity data failed and no data were analyzed for any participant, the registry record does not provide an estimate of the primary comparison.
3. Trial Design
This was a randomized, quadruple-masked crossover pilot study with an enrollment of 5. In a crossover design, participants receive more than one study condition in sequence, so each participant can contribute observations under different conditions. Here, the registry describes two sequences: torso weights followed by sham weights, and sham weights followed by torso weights.
Both sequences began with 4 weeks without weights. Participants then wore the assigned garment for 2 weeks, followed by the other garment for 2 weeks. The garments were worn 2-4 hours daily during those periods. Participants wore the Fitbit Flex throughout. The primary outcome timeframe was up to 8 weeks.
Random allocation to sequence can help balance factors that might otherwise differ between groups. Masking is intended to limit the influence of expectations on participants, study staff, or outcome assessment. The record identifies masking as quadruple, but does not describe which four roles were masked or how masking was maintained. That information is not reported in the registry record.
The crossover structure makes the sequence and timing important to interpretation. In general, crossover analyses need to account for within-participant comparisons and may need to consider period or carryover effects. The registry record does not report an analysis plan, a washout period, or how such effects would be assessed. It also does not give a sample-size justification. The study is identified as a pilot, and its recorded enrollment was 5; the record does not establish that this enrollment was intended to support a definitive efficacy comparison.
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: All 5 volunteers were enrolled.
Overall Study
| Stage | Torso Weights Then Sham Weights | Sham Weights Then Torso Weights |
|---|---|---|
| STARTED | 3 | 2 |
| COMPLETED | 3 | 2 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Server where data were stored failed. No data analyzed for any participant.
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.
| Characteristic | Unit | Torso Weights Then Sham Weights | Sham Weights Then Torso Weights | Total |
|---|---|---|---|---|
| Participants analysed | 3 | 2 | 5 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 3 | 1 | 4 |
| Age, Categorical: >=65 years | Participants | 0 | 1 | 1 |
| Sex: Female, Male: Female | Participants | 2 | 2 | 4 |
| Sex: Female, Male: Male | Participants | 1 | 0 | 1 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Region of Enrollment: United States | Participants | 3 | 2 | 5 |
6. Primary Endpoint
The registry lists the primary outcome as Steps Per Day, measured over up to 8 weeks.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint was steps per day, a continuous measure of physical activity. The registry describes continuous activity data collection with a commercially available remote monitoring device, with data stored on a server. Participants wore a Fitbit Flex throughout the study, and the stated timeframe was up to 8 weeks.
Step counts offer a way to quantify ambulatory activity over time in everyday settings. In general, a remotely monitored measure can capture activity outside a clinic visit, which may be relevant when the research question concerns daily behavior rather than performance during a single supervised test. Repeated measurements can also describe how activity varies across study periods, provided the device records and stores usable data.
Here, however, the registry says that the server where the data were stored failed and that no data were analyzed for any participant. Thus, the intended endpoint is clear, but the posted results do not include an analyzed step-count value, a comparison between torso and sham weights, or an estimate of uncertainty. The record does not report whether any activity data were recoverable by another means.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive fact is that the primary outcome was not analyzed. The registry specifically attributes this to failure of the server where the data were stored and states that no data were analyzed for any participant. There is therefore no posted estimate of steps per day under torso weighting, no corresponding estimate under sham weighting, and no treatment-effect estimate to interpret.
This is different from a result showing little or no difference. A null or small estimated effect would still require analyzed observations and would carry uncertainty. In this record, the primary comparison is unavailable; the evidence does not distinguish between a possible increase, decrease, or little change in activity. The study’s hypotheses about increased activity, fewer falls, and improved muscle activation are hypotheses, not findings.
The registry lists an enrollment of 5 and identifies the study as a pilot. That small enrollment is important context, but it is not the reason the primary result is absent: the stated reason is the server failure and lack of analyzed data. Nor can the enrollment be used to infer an effect. The posted record lists no analyses, and the primary outcome has not been reduced to a comparison.
Thirteen outcome measures are recorded, including the primary measure and outcomes relating to balance, gait, movement ability, confidence, multiple sclerosis impact and walking, six-minute walking, and stride length. The registry does not provide analyzed results for these measures. The availability of outcome names should not be mistaken for availability of outcome findings.
The adverse-event section lists zero serious events in each of the torso-weight, sham-weight, and pre-weighting groups, with 5 at risk in each. These are descriptive figures in the registry record, not an estimate of comparative safety. The record does not describe event-monitoring methods or provide further detail that would establish how completely events were captured.
Any tables, plots, or planning displays on this page should be read in light of this missing primary analysis. A display cannot replace the underlying participant data or create an observed treatment effect. Planning illustrations, if shown, are not observed power and do not change the fact that no primary outcome data were analyzed. Overall, the posted record documents a study design and a data-storage failure, but does not establish the direction, magnitude, or precision of an intervention effect.
9. Statistical Methods Explained
The record identifies randomization, quadruple masking, and a crossover model. It does not report a statistical analysis plan. The methods below explain general statistical considerations for this type of design; they should not be read as a description of analyses that were actually conducted in this study.
What does randomization contribute?
In general, random assignment to treatment sequence aims to make the sequences comparable with respect to both measured and unmeasured characteristics, apart from chance variation. In a crossover study, the assignment is to an order of conditions. Randomization does not itself ensure that a comparison is unbiased if outcome data are missing, if the assigned sequence is not followed, or if the analysis does not respect the design. The registry record does not report the randomization method or allocation concealment.
Why does crossover structure matter?
In general, a crossover design compares conditions within the same participant, which can account for stable differences between people. Observations from one participant are related, however, and should not be treated as if they came from unrelated individuals. A suitable analysis usually represents that within-participant relationship and considers the period in which each condition was received. A carryover effect occurs when an earlier condition continues to influence outcomes during a later period. The record does not report whether carryover was considered, whether a washout was planned, or what model would have been used.
Sequence can matter as well: participants received torso weights then sham weights or the reverse. In general, an analysis may need to distinguish treatment condition from period and sequence, depending on the design and available observations. The registry record does not report a model specification, an estimand, or how deviations from the planned sequence would be handled.
What would an effect estimate and confidence interval tell readers?
In general, an effect estimate summarizes the contrast between conditions on the outcome scale. A confidence interval describes the range of values compatible with the data and the stated statistical procedure. It communicates precision as well as direction and magnitude; it is not a guarantee that the true effect lies within the interval. A p-value, when used, summarizes how incompatible the observed data are with a specified null model, under the analysis assumptions. It does not measure the probability that a treatment works or that the null hypothesis is true.
None of these inferential quantities is available for the primary outcome in the posted record because no data were analyzed. The record does not report a p-value, a confidence interval, or an effect estimate for steps per day. It also does not report a planned significance threshold or whether the analysis would have been adjusted for other factors.
Why can variability affect analysis?
In general, the variability of an outcome affects how precisely a comparison can be estimated. Step counts may vary across days and participants, so repeated observations can require methods that represent both within-person and between-person variation. For count-like data, over-dispersion means that variability exceeds what a simple count model assumes; if present, ignoring it can make uncertainty appear too small. These are general considerations, not claims about the distribution of the unobserved data in this study. The record gives no analyzed step counts with which to assess variability or over-dispersion.
How should planning displays be understood?
In general, sample-size calculations are planning tools: they explore how many participants might be needed under stated assumptions about the target effect, variability, significance level, and allocation. They are not evidence that a study had a particular probability of detecting an effect after it was conducted. Sample-size illustrations are not observed power. The registry record does not report the study’s own sample-size justification or planning assumptions. Any planning section on this page should therefore be treated as an illustration rather than as a reported feature of the trial design.
Finally, the registry records 0 analyses. This means readers should not infer that a particular statistical procedure was used merely because the design could support one. The planned handling of missing observations, outcome transformations, repeated measurements, covariates, and sensitivity analyses is not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 13 outcome measures, including steps per day and several secondary measures. When a study examines multiple outcomes or time points, the chance of finding at least one apparently notable result by chance can increase. In general, multiplicity can be addressed through a prespecified hierarchy, adjustment procedures, or clear separation of primary and exploratory analyses. The registry record does not report a multiplicity strategy, and it reports no analyses. The record therefore gives no basis for determining how multiple comparisons would have been handled.
The primary outcome is explicitly unavailable because the server where the activity data were stored failed. The registry states that no data were analyzed for any participant. It does not report whether the missing activity data could be recovered, whether any participants had partial records, or whether the secondary outcomes were collected and analyzed. Those details are not reported in the registry record.
In general, missing-data methods depend on why observations are missing and on what information remains available. A failed storage server is a data-availability problem; without analyzable observations, statistical methods cannot reconstruct the missing outcome values from the registry description alone. The record does not report an imputation method, a missing-data assumption, or a sensitivity analysis.
The registry also does not define an analysis population, such as whether results would have been based on assignment to sequence, treatment received, or participants with complete data. No analyses are listed. Accordingly, the posted record cannot show whether the analysis would have followed the randomized assignment or how exclusions and protocol deviations would have been treated.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Torso Weights | 0 | 5 | 0.0% |
| Sham Weights | 0 | 5 | 0.0% |
| Pre-weighting (no Weights) | 0 | 5 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Torso Weights | 0 of 5 |
| Sham Weights | 0 of 5 |
| Pre-weighting (no Weights) | 0 of 5 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Torso Weights | 0 of 5 |
| Sham Weights | 0 of 5 |
| Pre-weighting (no Weights) | 0 of 5 |
The adverse-event section lists three groups: Torso Weights, Sham Weights, and Pre-weighting (no Weights). For each listed group, 5 were at risk, 0 serious adverse events were reported, 0 other adverse events were reported, and 0 deaths were reported. These figures describe what is recorded in the registry; they are not a comparative estimate of safety.
The study used a crossover design, so the listed groups correspond to study conditions or periods rather than necessarily representing separate sets of participants. The record does not explain how the at-risk counts were assigned across periods, how events were collected, or whether events were attributed to a particular condition. The registry also lists no serious adverse-event terms.
In general, a zero count in a small study does not establish that an event cannot occur or that two conditions have equivalent safety. The counts need to be read alongside the enrollment, study duration, and completeness of event ascertainment. The registry record does not report additional detail on surveillance or follow-up for adverse events, so these figures should be interpreted as limited descriptive reporting rather than proof of absence of harm.
12. What the Result Does, and Does Not, Mean
This record documents a randomized, quadruple-masked crossover pilot study designed to examine torso weighting in volunteers with multiple sclerosis. Its primary measure was steps per day, collected through a remote activity-monitoring device. The planned sequence involved a period without weights followed by periods with torso weights and sham weights, with the order of the two weighted conditions varying between sequences.
What the posted record does establish is that the server storing the primary activity data failed and that no data were analyzed for any participant. It also records the outcome measures and the adverse-event counts. These registry entries describe what happened to the data and what appears in the registry; they do not provide evidence about the size or direction of a treatment effect.
The record does not show that torso weighting increased physical activity, reduced falls, or changed muscle activation. Those were stated hypotheses. Nor does the record show that torso weighting had no effect. Without analyzed outcome data, neither conclusion follows. The same applies to the listed secondary measures: their presence in the record does not mean results were obtained or reported.
The serious and other adverse-event counts are all recorded as zero in the listed groups, but those counts do not establish comparative safety. The enrollment was 5, and the registry does not provide further detail on event ascertainment or the handling of crossover periods.
For readers considering the calculated displays elsewhere on this page, the key distinction is between a design description, a result, and a planning exercise. A design specifies how a study was intended to compare conditions. A result requires analyzed observations. A planning illustration explores assumptions about future study size and is not an observed trial finding. This registry record contains the design and the failure explanation, but no analyzed primary comparison from which to judge effect or precision.
13. Why This Trial Matters Statistically
This study is statistically informative as an example of the dependence of inference on data capture. A carefully described randomized and masked design can support a comparison only if the outcomes are available for analysis. Here, the registry says that the activity-data server failed and that no data were analyzed for any participant. The practical consequence is that the primary statistical question remains unanswered in the posted record.
The crossover design also highlights why study structure must be reflected in analysis. In general, repeated observations from the same person are related, and treatment order and period may matter. A statistical method that ignored those features could misrepresent uncertainty. The registry does not report a model or an analysis plan, so readers cannot assess how those design features would have been addressed.
The record lists several outcomes, which raises general questions about prioritization and multiplicity. A prespecified primary outcome can help distinguish the main question from secondary or exploratory questions. However, even a clearly identified primary endpoint cannot support an inference when its data are unavailable. Here, the step-count endpoint is named, but no estimate or uncertainty is reported.
The study also illustrates the difference between a pilot’s role and a definitive comparison. A pilot can test procedures and expose operational issues, including failures in data storage. The registry describes the project as a pilot study, but does not report a sample-size justification or analysis plan. Readers should not use planning displays as evidence of observed power or as a substitute for study results.
The most defensible statistical reading is therefore limited but useful: the record describes a randomized crossover protocol, identifies a primary activity measure, and reports a data-storage failure that prevented analysis. It does not support a conclusion about efficacy, no effect, or comparative safety.
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 limitation reported in this record is the failure of the server where the primary activity data were stored. The registry states that no data were analyzed for any participant. As a result, there is no posted estimate of steps per day, no treatment contrast, and no confidence interval or other measure of precision for the primary outcome.
Enrollment was 5, and the study is described as a pilot. The registry does not report a sample-size justification, so the intended precision or detectable effect cannot be assessed from the record. The small enrollment also limits how much the study could describe variation across participants, but the data failure is the direct reason the primary result is unavailable.
The registry identifies a randomized, quadruple-masked crossover design and describes the two treatment sequences. It does not report the randomization procedure, which roles were masked, whether masking was maintained, or how period and possible carryover effects would have been handled analytically. A washout period and statistical analysis plan are not reported in the registry record.
Although 13 outcome measures are listed, the record reports 0 analyses. It does not establish which secondary outcomes were successfully collected, whether they were affected by the storage failure, or whether any results were analyzed elsewhere. It also does not report how missing data, protocol deviations, or analysis populations would have been handled.
The safety figures are limited descriptive counts. The record lists 5 at risk and zero reported events for each of three groups, but does not explain event ascertainment or attribution across crossover periods. These limitations mean the adverse-event counts should not be treated as a precise comparison between conditions.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Long-Term Effects of Torso-Weighting — NCT02743312.
- Registry record last updated 2018-11-06; status last verified 2018-10.
- Results first posted 2018-11-06.
- 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
The registry describes a completed randomized, quadruple-masked crossover pilot study of torso weighting in multiple sclerosis, with an enrollment of 5. Participants followed one of two sequences involving torso weights and sham weights after 4 weeks without weights. Each garment period lasted 2 weeks, with wear for 2-4 hours daily; participants wore a Fitbit Flex throughout.
The primary outcome was steps per day over up to 8 weeks. The registry says the server storing activity data failed and no data were analyzed for any participant. It lists 13 outcome measures and 0 analyses. The adverse-event section records 0 serious events, 0 other events, and 0 deaths in each of the torso-weight, sham-weight, and pre-weighting groups, with 5 at risk in each. The record does not provide an analyzed estimate of the primary treatment comparison.