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 Be Fit in Residency: a Randomized Controlled Trial of an Activity Device Among Medicine Residents (NCT01287208), a randomized na study. The primary outcome is Steps Per Day.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01287208 |
| Phase | NA |
| Condition | Physical Activity |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 104 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Massachusetts General Hospital |
| Results posted | Yes |
2. Clinical Question
Population
Medicine residents were the study population. The registry describes residents as an employee population with limited time for exercise. Enrollment was 104. The primary-outcome description says that 4 participants withdrew before randomization and 5 withdrew after randomization.
Intervention
Participants assigned to the unblinded activity-monitor group could view activity data on the device and on a website where the data were uploaded. The device measured activity, with steps per day selected as the primary outcome.
Comparator
Participants assigned to the blinded activity-monitor group wore a device but could not view activity data on the device or access the website. This group served as the comparison for the visible-feedback approach.
Primary question
Over 12 weeks, did providing residents with visible activity-device feedback correspond to different steps-per-day results than wearing a blinded device? The registry reports a Wilcoxon (Mann-Whitney) comparison, with p = 0.16. The posted medians and interquartile ranges describe the observed groups; the reported test addresses the distributional comparison.
The broader study description also mentions a team competition after the randomized phase, associations between activity and weight change, and relationships involving sleep. Those aims are not the same as the primary randomized comparison. The registry record does not report results for the team competition or those associations. The clinical question addressed by the posted primary result is therefore narrower: how steps per day compared between residents who could see their activity data and residents who wore a blinded device during the randomized period.
3. Trial Design
The registry describes a randomized, parallel-group trial with single masking. Its listed enrollment was 104, and the study status is completed. The recorded start was 2010-10 and primary completion was 2011-06. Massachusetts General Hospital is listed as the sponsor. The phase is recorded as not applicable.
The two groups differed in whether participants could access activity feedback. The unblinded group could see data on the activity device and on a website; the blinded group wore a device but could not see the activity data or access the website. Thus, the comparison concerned access to feedback, rather than wearing a device versus not wearing one. The registry record does not describe who was masked in the single-masked design.
The primary outcome was steps per day over 12 weeks. The primary-outcome description states that 4 participants withdrew before randomization and 5 withdrew after randomization. The posted group results list 50 participants in the unblinded group and 49 in the blinded group. The record does not explain in detail how missing step measurements were handled or specify an analysis population such as intention-to-treat or per-protocol.
The study description identifies other aims, including a team competition after the randomized phase and associations involving activity, weight, and sleep. The registry record does not provide enough design detail to characterize those analyses or their timing beyond the stated sequence for the competition. It also does not report a sample-size justification, a detailed protocol, or interim-analysis plans. Those features should not be inferred from the randomized, parallel-group labels.
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: Subjects were randomized at the end of December 2010. Four residents withdrew prior to randomization (1 changed mind, 2 said they did not have time, 1 did not complete baseline assessment).
Phase 1: Randomized Phase
| Stage | Unblinded | Blinded |
|---|---|---|
| STARTED | 52 | 52 |
| COMPLETED | 50 | 49 |
| NOT COMPLETED | 2 | 3 |
Phase 2: Team Competition
| Stage | Unblinded | Blinded |
|---|---|---|
| STARTED | 50 | 49 |
| COMPLETED | 50 | 49 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: We excluded 4 participants who withdrew prior to randomization and 5 who withdrew after randomization.
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 | Unblinded | Blinded | Total |
|---|---|---|---|---|
| Participants analysed | 52 | 52 | 104 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 52 | 52 | 104 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Age, Continuous | years | 29 (SD 0.3) | 29 (SD 0.4) | 29 (SD 0.2) |
| Sex: Female, Male: Female | Participants | 27 | 29 | 56 |
| Sex: Female, Male: Male | Participants | 25 | 23 | 48 |
| Region of Enrollment: United States | participants | 52 | 52 | 104 |
6. Primary Endpoint
The registry lists the primary outcome as Steps Per Day, measured over 12 weeks.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | steps per day |
| Endpoint classification | Time-to-Event Analysis |
The primary endpoint was steps per day, recorded on an activity device, with a 12-week timeframe. Step counts quantify ambulatory activity in a form that can be recorded repeatedly by a wearable monitor. Using the same type of measure in both groups makes it possible to compare activity levels while varying whether participants can see their own feedback.
The registry reports the endpoint as a median with an interquartile range. The unblinded group had a median of 6369 steps per day, with an interquartile range from 1999 to 8796. The blinded group had a median of 6063 steps per day, with an interquartile range from 1299 to 8723. These summaries describe the observed distributions; the interquartile range gives the middle half of observed values, rather than a confidence interval for the median or for the group comparison.
As general statistical context, a median can be useful when observations are skewed or include unusually high or low values, because it is less affected by extreme observations than a mean. The interquartile range complements the median by describing variability across the central portion of the data. These summaries alone do not show how precisely the difference between groups has been estimated. The registry does not report a confidence interval for the primary comparison.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry reports medians of 6369 steps per day for the unblinded group and 6063 for the blinded group over 12 weeks. The corresponding interquartile ranges are 1999 to 8796 and 1299 to 8723. These ranges overlap substantially, and the registry’s reported Wilcoxon (Mann-Whitney) comparison has p = 0.16. The test result does not provide strong evidence against the null comparison at commonly used significance thresholds, but it does not establish that the groups were equivalent.
The p-value summarizes how compatible the observed test result is with a null model for the comparison, under the test’s assumptions. It is not the probability that the null hypothesis is true, nor does it quantify the probability that visible feedback will or will not change activity. A non-small p-value can reflect a small underlying effect, variability, limited information, or a combination of these. The registry does not report a confidence interval for this primary comparison, so the precision and range of effects compatible with the data cannot be directly assessed from the reported test alone.
The trial randomized residents to access or no access to activity feedback, and the observed medians describe those groups. The result does not establish that one approach caused a particular change in activity for each participant. The registry record does not report a baseline-adjusted estimate, participant-level change, or a detailed method for addressing missing step measurements. Although the primary description identifies withdrawals before and after randomization, their handling in the reported analysis is not specified.
The endpoint and timeframe also limit the interpretation. The primary outcome concerns steps per day during 12 weeks, not every dimension of physical activity, longer-term behavior, weight change, or health outcomes. The study description lists later competition and activity, weight, and sleep aims, but their results are not reported in the registry record. The primary result should therefore be read as a comparison of recorded step-count distributions in the randomized groups over the stated period, not as a broad conclusion about all activity interventions or outcomes.
Overall, the posted primary data show numerically different group medians, alongside a reported comparison that does not indicate a clear statistical separation. Without a reported confidence interval or fuller account of missing-data handling, the magnitude and precision of the between-group contrast remain uncertain. This uncertainty is central: the findings do not show that the groups were identical, and they do not establish that visible feedback increased or decreased steps per day.
9. Statistical Methods Explained
The primary outcome is summarized using medians and interquartile ranges. As general statistical context, the median is the center of an ordered set of observations: half the values fall at or below it and half at or above it. The interquartile range runs from the lower to the upper quartile and describes the spread of the middle portion of the observations. It is a description of individual-level data dispersion, not an uncertainty interval for a treatment effect.
For the unblinded group, the posted median is 6369 steps per day and the interquartile range is 1999 to 8796. For the blinded group, the posted median is 6063 and the interquartile range is 1299 to 8723. The record reports a Wilcoxon (Mann-Whitney) test with p = 0.16. As general statistical context, this rank-based test compares the distributions of observations between two groups using their ordering, rather than requiring a comparison of means. It is often used when a continuous outcome is summarized by medians or when a distributional comparison is preferred. The test does not, by itself, estimate a difference between the two medians.
Interpreting the Wilcoxon test as a test specifically of medians requires additional assumptions about the shapes of the group distributions. More generally, a rank-based test can respond to differences in location, spread, or distributional shape. The registry provides the test name and p-value but does not report the test statistic, a confidence interval, or further distributional details. Therefore, the most direct account is that the sponsor reported a rank-based comparison with p = 0.16, alongside the group medians and interquartile ranges.
As general statistical context, a p-value is calculated under a null model and summarizes how unusual the test result would be if that model held. It is not a measure of effect size, clinical importance, or the probability that the null model is true. A p-value that is not small does not prove no difference. It may arise when the estimated contrast is modest, when the data are variable, or when the study does not provide enough precision to distinguish plausible effects. A confidence interval would help show the range of effects compatible with the data, but the registry does not report one for the primary comparison.
There is a separate posted secondary result for weight, recorded as a mean with standard deviation at 6 months. The registry reports a two-sided t-test with p = 0.59. A recalculated unadjusted comparison from the posted summary statistics gives a mean difference of 0.0 pounds, a confidence interval from -10.8 to 10.8, and p = 1.0. As general statistical context, a t-test compares group means while accounting for variability and sample size; a confidence interval expresses uncertainty in the estimated mean difference under the calculation’s assumptions. The recalculation is not a substitute for the sponsor’s analysis and does not establish why its p-value differs from the registry-reported value. The discrepancy flag is false, so the record does not identify this as a discrepancy requiring precedence to be assigned.
The record identifies enrollment as 104 and describes withdrawals before and after randomization. The primary group summaries list 50 participants in the unblinded group and 49 in the blinded group. As general statistical context, missing outcome data can affect a comparison if the likelihood of missingness is related to activity or differs between groups. The record does not report the missing-data method, the number contributing primary measurements by time point, or an analysis-population definition. These omissions make it difficult to evaluate how withdrawals influenced the estimate.
As general statistical context, testing multiple outcomes or questions raises the chance of obtaining at least one apparently noteworthy result by chance if every test is interpreted separately. The registry lists five outcome measures and two analyses, but does not report a multiplicity adjustment or explain how secondary analyses were prioritized. The sample-size assumptions describe an illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such an illustration is not a description of the trial’s original design justification, which is not reported in the registry record. It is not observed power and should not be interpreted as evidence that the trial had a particular probability of detecting an effect.
Finally, the randomized parallel-group design supports a comparison of the assigned groups, but details needed to assess implementation are limited. The record labels masking as single without identifying who was masked. It does not report interim analyses, a complete statistical analysis plan, or detailed procedures for handling protocol deviations. As general statistical context, such information helps readers understand which analyses were planned, how missing observations were handled, and whether reported comparisons followed the original plan.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Wilcoxon (Mann-Whitney) | not reported | 0.160 |
The sponsor’s reported primary analysis is a Wilcoxon (Mann-Whitney) comparison, with p = 0.16. The registry does not report a confidence interval or an estimated between-group effect for that analysis. No recalculated primary comparison is available because the posted summaries are medians and interquartile ranges, and a simple unadjusted comparison cannot be reconstructed from those summaries alone. The sponsor’s reported primary test is therefore the available inferential result; the group medians and ranges add descriptive context but do not replace it.
For weight, the sponsor reports a two-sided t-test with p = 0.59. A simple unadjusted recalculation from the posted means, standard deviations, and group sizes gives a mean difference of 0.0 pounds, a confidence interval from -10.8 to 10.8, and p = 1.0. These figures do not match the sponsor’s reported p-value. The recalculation uses only the posted summary statistics and is not necessarily the same analysis as the sponsor’s test. The registry does not explain the difference, and the record does not flag a discrepancy. The sponsor’s own analysis should not be replaced by the summary-statistic recalculation.
11. Multiplicity, Missing Data and Analysis Populations
The record identifies multiple outcomes and aims: steps per day, distance per day, calories burned per day, weight, sleep, a later team competition, and associations among activity, weight, and sleep. It lists five outcome measures and two analyses. However, it does not report a multiplicity adjustment, a hierarchy of hypotheses, or a detailed analysis plan explaining which questions were primary, secondary, or exploratory beyond the stated primary endpoint. The results should therefore not be read as if every analysis were a separately protected confirmatory test.
As general statistical context, multiplicity arises when many outcomes, time points, or comparisons are tested. If each is judged using a conventional threshold without adjustment, the chance of at least one false-positive result across the collection of tests can be greater than for a single test. Adjustments can control that error rate, but they may also change the precision or power of individual comparisons. Because the registry does not describe an adjustment, it is not possible to determine whether one was used outside the reported record.
The primary-outcome description says that 4 participants withdrew before randomization and 5 withdrew after randomization. The reported primary group summaries list 50 participants in the unblinded group and 49 in the blinded group. The record does not state whether all randomized participants were included in the primary analysis, how withdrawals were treated, whether measurements were missing at particular time points, or whether any imputation or sensitivity analyses were used. It also does not identify an intention-to-treat or per-protocol analysis population.
As general statistical context, these details matter because estimates can be affected if outcome availability is related to activity, treatment assignment, or other participant characteristics. Excluding participants after randomization can also weaken the protection randomization provides if the exclusions differ meaningfully between groups. The available record is insufficient to assess these possibilities. It also does not report interim-analysis plans or a detailed protocol-based account of the analysis decisions. These omissions constrain evaluation of both multiplicity and missing-data assumptions.
12. 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 |
|---|---|---|---|
| Unblinded | 0 | 52 | 0.0% |
| Blinded | 0 | 52 | 0.0% |
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 |
|---|---|
| Unblinded | 0 of 52 |
| Blinded | 0 of 52 |
The registry reports 0 serious adverse events among 52 participants at risk in the unblinded group and 0 among 52 at risk in the blinded group. It also reports 0 other adverse events among 52 at risk in each group. No serious adverse-event terms are listed. Death counts are not reported in the registry record.
These figures describe the safety counts recorded for the groups; they do not show that no adverse events could occur. As general statistical context, a count of zero observed events in a study does not establish that the underlying event risk is zero. The number at risk, follow-up and event-collection procedures, and completeness of reporting all affect what can be learned from a zero count. The registry record does not provide further detail about adverse-event definitions, monitoring procedures, or follow-up for safety.
13. What the Result Does, and Does Not, Mean
The primary result describes steps per day among residents assigned to an unblinded activity monitor and residents assigned to a blinded monitor during a 12-week period. The posted medians were 6369 and 6063, respectively, and the sponsor reported a Wilcoxon (Mann-Whitney) p-value of 0.16. This is evidence about the comparison recorded in this trial, not a general estimate for every resident, workplace, or activity-feedback approach.
The reported p-value does not establish that feedback had no effect, that the groups were equivalent, or that any difference was important to participants. The registry does not report a confidence interval for the primary comparison, so readers cannot use the posted record to assess a range of plausible primary effects. Nor does the endpoint measure every form of physical activity. It records steps per day, and the stated timeframe is 12 weeks.
The study description names other questions about weight, sleep, distance, calories, and a later competition. Results for most of these questions are not reported in the registry record. The available weight result is not a substitute for the primary activity endpoint, and it does not establish the relationships described in the study aims. The record also does not provide enough detail to evaluate missing-data handling, analysis populations, multiplicity, or the masking procedures beyond the single-masked label.
As general statistical context, trial results are best interpreted by considering both the estimated outcome and its uncertainty, the study design, and the completeness of the analysis record. Here, the group summaries and reported test offer a limited account of the primary comparison. They do not support a categorical conclusion about whether visible feedback changes activity, nor do they establish what would happen in settings or populations beyond the residents studied.
14. Why This Trial Matters Statistically
This trial illustrates a common design question in behavioral measurement: whether participants’ access to feedback from a device is associated with different recorded behavior than wearing a device without feedback. Because both groups wore activity devices, the randomized contrast focuses on visibility of activity information rather than device use alone. That distinction matters when translating a study question into an estimand: the result concerns the comparison between these two assigned approaches.
The choice of a median and interquartile range for steps per day is also statistically informative. These summaries describe the center and spread of the observed activity distributions without relying on means. The reported Wilcoxon (Mann-Whitney) test is consistent with a rank-based comparison, but it does not directly supply a median difference or its confidence interval. This is a useful example of why readers should distinguish descriptive summaries from the parameter a test evaluates.
The record also demonstrates the importance of reporting precision and analysis details. A p-value alone does not show the range of effects compatible with the data, and the primary comparison has no reported confidence interval. The withdrawals are identified, but their handling and the analysis population are not. The multiple aims are described, while results for most secondary questions are not reported. These gaps limit how far the findings can be interpreted, independently of the numerical test result.
More broadly, this study highlights that randomized allocation does not remove the need for transparent reporting of outcome definitions, missing data, multiplicity, and analysis decisions. Those details help distinguish a narrow comparison supported by the posted results from broader claims about behavior, weight, sleep, or later interventions.
15. 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 registry record leaves several trial-specific questions unresolved. It reports single masking but does not identify who was masked. It does not provide a detailed account of randomization procedures, allocation concealment, or implementation of the intervention beyond whether participants could see device and website data. These details are not reported in the registry record and should not be inferred from the design labels.
The primary outcome is specified as steps per day over 12 weeks, but the record does not state the exact summary period or how repeated readings were combined into the reported participant-level values. It identifies withdrawals before and after randomization without explaining how missing measurements were handled. The analysis population and any sensitivity analyses are not reported. These omissions make it difficult to determine how representative the posted group summaries are of all randomized participants.
The primary comparison has no reported confidence interval or effect estimate beyond the separate group medians. The reported rank-based test therefore gives limited information about the magnitude and precision of the comparison. The record lists additional study aims, but numerical results for most secondary outcomes and the later competition are not reported. It also does not report a multiplicity procedure, interim-analysis plan, full statistical analysis plan, or sample-size justification.
Finally, the recorded enrollment and the group counts in the primary results do not on their own explain participant flow in detail. The record does not describe the residents’ characteristics, recruitment setting beyond the population description, adherence to device use, or completeness of device recordings. Without those details, it is difficult to assess generalizability or the influence of implementation and data availability on the observed results.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Be Fit in Residency: a Randomized Controlled Trial of an Activity Device Among Medicine Residents — NCT01287208.
- Registry record last updated 2017-12-07; status last verified 2017-12.
- Results first posted 2012-11-29.
- 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.
19. Record Summary
This completed randomized, parallel-group trial examined visible versus blinded activity-monitor feedback among medicine residents. Enrollment was 104. The primary endpoint was steps per day over 12 weeks, summarized by median and interquartile range. The registry reports a Wilcoxon (Mann-Whitney) comparison with p = 0.16 and does not report a confidence interval for the primary comparison.
The primary-outcome description notes withdrawals before and after randomization, while the posted group summaries list 50 participants in the unblinded group and 49 in the blinded group. The record does not report detailed missing-data methods or an analysis-population definition. A weight outcome is reported at 6 months; results for several other stated aims are not reported in the registry record. Safety counts list no serious or other adverse events in either group, while death counts are not reported.