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 Starting a Testosterone and Exercise Program After Hip Injury (NCT02938923), a randomized phase 3 study. The primary outcome is Change in Six Minute Walk Distance.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02938923 |
| Phase | Phase 3 |
| Condition | Hip Fracture, Frailty, Sarcopenia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 129 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Washington University School of Medicine |
| Results posted | Yes |
2. Clinical Question
Population
Female participants aged 65 and older who had experienced a hip fracture. The trial summary describes a multi-center study. The primary outcome population is described as randomized participants (N=122) who completed the measure at baseline and at one other follow-up time point, either 3 Months or 6 Months after randomization.
Intervention
Supervised exercise training twice per week for six months, combined with daily topical testosterone 1% gel. The description specifies 12.5 mg per pump depression. Participants also received nutritional counseling and calcium and vitamin D supplements, according to the study summary.
Comparator
The trial included two comparison approaches: supervised exercise training with daily placebo gel for six months, and Enhanced Usual Care. Enhanced Usual Care included a home exercise program and monthly health education modules for six months. The study summary says all participants received nutritional counseling and calcium and vitamin D supplements.
Primary question
The primary outcome was change in 6-Minute Walk Distance from baseline to 6 months, measured in meters. The clinical question is whether the reported change in walking distance differed across the testosterone-plus-exercise, placebo-gel-plus-exercise, and Enhanced Usual Care groups. Numerical group estimates and between-group test results are not reported in the registry record, so the direction and precision of any differences cannot be determined.
3. Trial Design
The registry record describes a completed phase 3, randomized, parallel-group trial with enrollment of 129. It lists hip fracture, frailty, and sarcopenia as conditions. The study summary describes female hip-fracture patients aged 65 and older and characterizes the study as double-blinded and multi-center. The registry’s masking classification is quadruple. The operational meaning of these masking labels, including which roles were masked, is not reported in the registry record.
Three groups were compared. Exercise + Testosterone involved supervised exercise training twice per week and daily topical testosterone 1% gel for six months. Exercise + Placebo involved the same exercise schedule with daily placebo gel. Enhanced Usual Care involved a home exercise program three times per week and monthly health education modules, also for six months. The study summary states that all participants received nutritional counseling and calcium and vitamin D supplements.
The summary describes a randomization schedule of 4 participants assigned to Exercise + Testosterone, 4 to Exercise + Placebo, and 1 to Enhanced Usual Care for every 9 participants randomized. This planned assignment pattern differs from equal allocation. The record does not report the randomization sequence-generation method, concealment procedure, or reasons for the allocation pattern. The actual at-risk group counts reported for the adverse-event summaries are 54, 55, and 20, respectively; those counts are specific to those summaries and should not be treated as a complete account of every analysis population.
The primary outcome population is described as randomized participants (N=122) with the measure at baseline and at least one other specified follow-up. The record does not explain why this population differs from the reported enrollment or whether the same participants contributed to every analysis. Start date was 2019-03-05, primary completion was 2023-08-31, and the status is completed. Detailed recruitment, protocol deviations, and the full statistical analysis plan 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.
Before assignment: 4695 charts were reviewed. 907 were eligible by chart review and 344 agreed to complete the pre-screening questionnaire. Of those, 28 failed, another 42 were deemed ineligible after further review, 7 refused to review the consent, and 60 refused consent. Leaving 207 who consented for screening, of which 12 refused to continue to screening visit and 10 were found ineligible prior to screening. 185 screened, of whom 32 screen failed, 7 refused after screening, and 17 were ineligible to randomize.
Overall Study
| Stage | Exercise + Testosterone (EX + T) | Exercise + Placebo (EX + P) | Enhanced Usual Care (EUC) |
|---|---|---|---|
| STARTED | 55 | 54 | 20 |
| COMPLETED | 54 | 51 | 19 |
| NOT COMPLETED | 1 | 3 | 1 |
Reasons for not completing this period:
| Reason | Exercise + Testosterone (EX + T) | Exercise + Placebo (EX + P) | Enhanced Usual Care (EUC) |
|---|---|---|---|
| Death | 1 | 0 | 0 |
| Physician Decision | 0 | 2 | 0 |
| Withdrawal by Subject | 0 | 1 | 1 |
Analysis population for the primary outcome, as defined in the registry: Randomized participants (N=122) who completed the primary outcome measure at baseline and one other follow-up time point (3 Months or 6 Months post 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.
Population: Randomized participants.
| Characteristic | Unit | Exercise + Testosterone (EX + T) | Exercise + Placebo (EX + P) | Enhanced Usual Care (EUC) | Total |
|---|---|---|---|---|---|
| Participants analysed | 55 | 54 | 20 | 129 | |
| Age, Continuous | years | 79.42 (SD 9.08) | 78.28 (SD 7.52) | 82.98 (SD 8.39) | 79.49 (SD 8.43) |
| Sex: Female, Male: Female | Participants | 55 | 54 | 20 | 129 |
| Sex: Female, Male: Male | Participants | 0 | 0 | 0 | 0 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 1 | 3 | 1 | 5 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 54 | 48 | 18 | 120 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 3 | 1 | 4 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 2 | 3 | 1 | 6 |
| Race (NIH/OMB): White | Participants | 52 | 49 | 19 | 120 |
| Race (NIH/OMB): More than one race | Participants | 1 | 1 | 0 | 2 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 1 | 0 | 1 |
| Region of Enrollment: United States | participants | 55 | 54 | 20 | 129 |
| Site: Hebrew SeniorLife (HSL)/Beth Israel (BI)/Harvard Medical School | Participants | 17 | 16 | 5 | 38 |
| Site: Washington University in St. Louis (WashU) | Participants | 9 | 12 | 5 | 26 |
| Site: University of Colorado - Denver (UCD) | Participants | 1 | 0 | 2 | 3 |
| Site: University of Connecticut (UConn Health) | Participants | 11 | 11 | 4 | 26 |
| Site: University of Maryland Baltimore/Johns Hopkins University (UM/JH) | Participants | 1 | 0 | 2 | 3 |
| Site: University of Texas Medical Branch at Galveston (UTMB) | Participants | 8 | 7 | 1 | 16 |
| Site: University of Utah Health (UofU) | Participants | 5 | 4 | 0 | 9 |
| Site: University of Pittsburgh (Pitt)/University of Pittsburgh Medical Center (UPMC) | Participants | 3 | 4 | 1 | 8 |
| Time from hip surgery to baseline start date | days | 98.29 (SD 32.52) | 108.54 (SD 33.42) | 89.65 (SD 21.62) | 101.24 (SD 32.02) |
| Type of Fracture: Intertrochanteric Fracture | Participants | 23 | 23 | 13 | 59 |
| Type of Fracture: Intracapsular or Femoral Neck Fracture | Participants | 25 | 22 | 3 | 50 |
| Type of Fracture: Subtrochanteric Fracture | Participants | 7 | 9 | 4 | 20 |
| Body mass index (BMI) | Kilograms/meters ^2 (kg/m^2) | 25.61 (SD 5.49) | 26.89 (SD 5.02) | 24.43 (SD 3.36) | 25.96 (SD 5.06) |
| History of Lung Disease: No lung disease | Participants | 51 | 51 | 16 | 118 |
| History of Lung Disease: lung disease | Participants | 4 | 3 | 4 | 11 |
| Co-morbidity count | units on a scale | 5.20 (SD 3.09) | 5.13 (SD 2.52) | 5.85 (SD 3.17) | 5.27 (SD 2.86) |
| Geriatric Depression Scale (GDS) | unites on a scale | 1.84 (SD 1.75) | 2.43 (SD 2.04) | 3.15 (SD 2.62) | 2.29 (SD 2.06) |
| Six Minute Walk Distance | meters | 267.77 (SD 86.60) | 225.92 (SD 98.09) | 206.37 (SD 76.82) | 240.73 (SD 92.82) |
| Total lean body mass | Grams | 36649.42 (SD 5892.42) | 37744.01 (SD 5543.42) | 35144.55 (SD 4276.92) | 36874.31 (SD 5558.35) |
6. Primary Endpoint
The registry lists the primary outcome as Change in Six Minute Walk Distance, measured over Baseline to 6 months.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Meters |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was change in 6-Minute Walk Distance from baseline to 6 months, measured in meters. Participants walked back and forth along a measured path, turning at marked ends. They were told when each minute had passed and received verbal encouragement every 60 seconds. The total distance was recorded, with higher distances indicating greater walking endurance. Change was defined as the Month 6 distance minus the baseline distance; higher positive values therefore represent greater improvement in distance over the study period.
A walking-distance measure captures performance during a standardized walking task. In general, such a measure can offer a functional outcome that is directly expressed as distance rather than as a rating alone. Its interpretation depends on how the task is administered and how complete the measurements are. The registry record gives the task description and direction of scoring, but it does not report measurement reliability, a clinically meaningful change threshold, or how the endpoint’s clinical importance was established.
The record identifies the primary timeframe as baseline to 6 months, while the primary outcome population is described as having a baseline measure and a measure at either 3 Months or 6 Months. It also mentions analyses of change through 3 months. How the 3-month data contributed to the stated primary analysis is not reported in the registry record. Numerical changes, between-group estimates, and confidence intervals are also not reported, so the endpoint definition is clear but the size and precision of the observed effects cannot be evaluated.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record defines a clinically interpretable primary measure: change in 6-Minute Walk Distance from baseline to 6 months, in meters, with higher positive values representing greater improvement. However, it does not report the numerical change for any group, the between-group contrasts, their confidence intervals, or the corresponding statistical test results. Consequently, the posted results do not establish the direction, magnitude, or precision of the primary comparison. It is not possible to determine from this record whether the groups differed, or whether any estimated difference was compatible with a range of potentially important effects.
The outcome description says that the primary analysis was an unadjusted baseline-to-6-month comparison. It also says that statistical contrasts were evaluated for change through 3 months and that additional analyses of covariance adjusted for baseline values and covariates that differed across groups at baseline. The estimates and results from these analyses are not reported in the registry record. The record also does not explain how the stated least-squares-mean parameter relates to the description of an unadjusted comparison. Without model details and numerical estimates, it is not possible to reconcile these descriptions or assess the impact of adjustment.
The primary outcome population is reported as randomized participants (N=122) who completed the measure at baseline and one other follow-up time point, either 3 Months or 6 Months. This definition does not by itself tell the reader how many participants contributed to the 6-month analysis, how incomplete measurements were handled, or whether all randomized participants were included in the primary comparison. Those details are not reported in the registry record. Enrollment of 129 and the primary outcome population definition are different pieces of information; the record does not explain the difference.
The record lists 11 outcome measures and 138 analyses. That breadth makes it important to know which comparisons were prespecified and how the risk of false-positive findings across outcomes and analyses was addressed. A multiplicity adjustment, hierarchy, or other strategy is not reported in the registry record. The absence of those details does not show that the analysis was inappropriate, but it limits interpretation of any isolated result. In this case, the numerical outcome results themselves are also absent, so no conclusion about statistical evidence can be drawn.
The planning illustration is described as using a two-sided significance level of 0.05, equal allocation, and a normal approximation, with effects expressed as fractions of the observed effect. These are stated assumptions for planning illustrations, not evidence about the trial’s own design choices or the analysis threshold used for its results. They do not establish observed power. Overall, the registry record describes the comparison and endpoint but does not report enough outcome statistics to estimate the treatment-group effects or judge their precision.
9. Statistical Methods Explained
The registry record describes a continuous primary outcome: change in walking distance, measured in meters. It identifies the reported parameter as a least-squares mean and the dispersion measure as a 95% confidence interval. The analysis description calls the primary comparison an unadjusted baseline-to-6-month comparison, while also noting additional analyses of covariance. The model specification and the numerical estimates are not reported in the registry record. The distinction between those analysis descriptions therefore cannot be resolved from the record.
What does a least-squares mean describe?
In general, a least-squares mean is an estimated group mean obtained from a statistical model. Depending on the model, it may represent a mean adjusted for other variables. The estimate is usually interpreted in the units of the outcome, so a walking-distance estimate is expressed in meters. The precise interpretation depends on the model terms, data included, and any adjustment. Although the registry identifies the parameter as a least-squares mean, the model details needed to understand its exact construction are not reported in the registry record.
How should a confidence interval be read?
In general, a confidence interval describes the statistical uncertainty around an estimated quantity under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval indicates less precision. The interval should be considered alongside the estimate and the clinical meaning of the outcome; it does not by itself establish whether a difference matters to patients. The registry identifies a 95% confidence interval for the primary outcome but does not report its limits or the estimate it accompanies. Its actual width and implications therefore cannot be assessed.
What do the baseline and follow-up comparisons mean?
The outcome is defined as each participant’s Month 6 distance minus their baseline distance. Comparing change across groups asks whether the amount of change differed between randomized groups. In general, a between-group comparison of change is not the same question as asking whether each group changed from its own baseline. Improvement within a group alone does not establish a difference between groups, because changes may also occur under the comparison approaches or over time. The registry describes a baseline-to-6-month primary comparison and mentions contrasts for change from baseline to 3 months. The actual contrasts and results are not reported in the registry record.
The record also mentions analysis of covariance to adjust for baseline values and covariates that differed across groups at baseline. In general, adjustment can account for measured differences included in a model and may improve precision when model assumptions are appropriate. It does not guarantee that all imbalances or sources of bias are addressed. Which covariates were used, how they were selected, how the models were specified, and whether the adjusted analyses were prespecified are not reported in the registry record. Because the record calls the primary analysis unadjusted while identifying a least-squares mean, further detail would be needed to explain the relation between those terms.
What can randomization contribute?
In general, random assignment helps make comparison groups similar in expectation with respect to both measured and unmeasured characteristics at the start of a trial. It does not ensure identical groups in a particular trial, and it does not resolve problems caused by missing outcomes, deviations from assigned interventions, or selective reporting. The record describes randomized allocation and a planned 4:4:1 assignment pattern. The sequence-generation and concealment methods, as well as the analysis approach to deviations from assigned groups, are not reported in the registry record.
What can the planning assumptions tell readers?
The planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It considers effects expressed as a fraction of the observed effect. These assumptions describe an illustrative planning calculation, not the trial’s own sample-size justification or its actual allocation pattern. The registry record does not report the trial’s original sample-size rationale or statistical plan. Planning sample sizes are not observed power, and they cannot substitute for the effect estimates and uncertainty intervals from the trial. No sample-size values or recalculated effects are reported in the registry record.
In general, a normal approximation can be useful for planning under specified assumptions, but its usefulness depends on the outcome distribution and the design. The record identifies the outcome as continuous and gives its unit, but does not report diagnostic checks, distributional summaries, or model-assumption assessments. It also does not report whether any adjustment accounted for the parallel-group structure beyond the stated analyses of covariance. Those omissions limit independent assessment of the statistical methods, rather than demonstrating that any particular assumption failed.
Why do analysis populations matter?
The primary outcome population is defined as randomized participants (N=122) with a baseline measurement and a measurement at either 3 Months or 6 Months. In general, analysis populations matter because they determine which participants contribute to an estimate and what question the estimate addresses. A population based on having observed measurements can differ in meaning from an analysis retaining all randomized participants. The registry record does not report the number included in the 6-month comparison, the extent or pattern of missingness, or a method for handling missing outcome data. It also does not report sensitivity analyses that might examine the effect of missing data assumptions.
Finally, the record lists 11 outcome measures and 138 analyses. In general, examining multiple outcomes, time points, and contrasts raises questions about multiplicity and selective emphasis. A prespecified primary comparison can help focus interpretation, while an appropriate multiplicity strategy can address the chance of false-positive findings across a family of tests. The registry record does not report a multiplicity strategy, an analysis hierarchy, or a complete account of which analyses were prespecified. Since the numerical primary result is also not reported, the methods information supports understanding the planned measurement and broad analytic approach but not an independent evaluation of the trial’s findings.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists 11 outcome measures and 138 analyses. The outcomes include changes in body composition, strength, physical performance, and functional-status measures, alongside the primary walking-distance measure. Multiple outcomes and contrasts can create a multiplicity issue: in general, the more statistical questions examined, the greater the chance that some findings may appear notable by chance. A prespecified hierarchy, adjustment method, or distinction between confirmatory and exploratory analyses can help readers interpret that breadth. Such a strategy is not reported in the registry record.
The record describes analyses of baseline-to-6-month change, contrasts involving the 3-month time point, and additional analyses of covariance. It does not report which of the 138 analyses were prespecified, how the analyses were grouped for multiplicity control, or whether results were selected for reporting based on their findings. The missing numerical outcome results also prevent assessment of how any statistical evidence relates to the range of analyses.
The primary outcome population includes randomized participants (N=122) who completed the measure at baseline and one other follow-up time point, either 3 Months or 6 Months. The registry record does not report the number contributing to the 6-month comparison, the amount or reasons for missing data, patterns of missingness, or the method used to handle incomplete outcomes. It also does not report sensitivity analyses under alternative missing-data assumptions.
In general, restricting an analysis to participants with observed measurements can affect interpretation if availability of those measurements is related to outcome or group assignment. This possibility cannot be evaluated from the registry information. The record does not report an intention-to-treat analysis population or explain how participants who did not complete assigned activities were treated in analyses. These are limitations of the available reporting, not evidence that a particular bias occurred.
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 |
|---|---|---|---|
| Exercise + Testosterone (EX + T) | 7 | 54 | 13.0% |
| Exercise + Placebo (EX + P) | 3 | 55 | 5.5% |
| Enhanced Usual Care (EUC) | 1 | 20 | 5.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Exercise + Testosterone (EX + T) | 1 of 55 |
| Exercise + Placebo (EX + P) | 0 of 54 |
| Enhanced Usual Care (EUC) | 0 of 20 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Exercise + Testosterone (EX + T) | 36 of 54 |
| Exercise + Placebo (EX + P) | 24 of 55 |
| Enhanced Usual Care (EUC) | 11 of 20 |
Most frequent serious adverse event terms
| Event term | Organ system | Exercise + Testosterone (EX + T) | Exercise + Placebo (EX + P) | Enhanced Usual Care (EUC) |
|---|---|---|---|---|
| Pneumonia | Infections and infestations | 0 of 54 | 0 of 55 | 1 of 20 |
| Hip Fracture | Injury, poisoning and procedural complications | 1 of 54 | 0 of 55 | 0 of 20 |
| Ruptured diverticulum of colon | Gastrointestinal disorders | 1 of 54 | 0 of 55 | 0 of 20 |
| Chest pain | General disorders | 1 of 54 | 0 of 55 | 0 of 20 |
| Kidney fibrosis | Renal and urinary disorders | 1 of 54 | 0 of 55 | 0 of 20 |
| Medical device complication | General disorders | 0 of 54 | 1 of 55 | 0 of 20 |
| Musculoskeletal pain | Musculoskeletal and connective tissue disorders | 1 of 54 | 0 of 55 | 0 of 20 |
| Diverticulum intestinal | Gastrointestinal disorders | 1 of 54 | 0 of 55 | 0 of 20 |
| Fractured sacrum | Injury, poisoning and procedural complications | 1 of 54 | 0 of 55 | 0 of 20 |
| Lumbar vertebral fracture | Injury, poisoning and procedural complications | 1 of 54 | 0 of 55 | 0 of 20 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Exercise + Testosterone (EX + T) | Exercise + Placebo (EX + P) | Enhanced Usual Care (EUC) |
|---|---|---|---|---|
| Arthralgia | Musculoskeletal and connective tissue disorders | 9 of 54 | 3 of 55 | 2 of 20 |
| Contusion | Injury, poisoning and procedural complications | 7 of 54 | 2 of 55 | 0 of 20 |
| Urinary tract infection | Infections and infestations | 4 of 54 | 4 of 55 | 0 of 20 |
| Back pain | Musculoskeletal and connective tissue disorders | 4 of 54 | 2 of 55 | 2 of 20 |
| Pain in extremity | Musculoskeletal and connective tissue disorders | 3 of 54 | 5 of 55 | 0 of 20 |
| Diarrhea | Gastrointestinal disorders | 5 of 54 | 2 of 55 | 0 of 20 |
| Upper respiratory tract infection | Infections and infestations | 3 of 54 | 3 of 55 | 1 of 20 |
| Joint swelling | Musculoskeletal and connective tissue disorders | 4 of 54 | 3 of 55 | 0 of 20 |
| Musculoskeletal pain | Musculoskeletal and connective tissue disorders | 6 of 54 | 1 of 55 | 0 of 20 |
| Nausea | Gastrointestinal disorders | 5 of 54 | 1 of 55 | 0 of 20 |
Serious events with the highest arm-level rate: pneumonia (5.0%), hip fracture (1.9%), ruptured diverticulum of colon (1.9%), chest pain (1.9%), kidney fibrosis (1.9%).
The registry record reports serious adverse events, other adverse events, and deaths by group. For serious adverse events, the reported affected and at-risk counts are 7 and 54 for Exercise + Testosterone, 3 and 55 for Exercise + Placebo, and 1 and 20 for Enhanced Usual Care. The death counts are 1, 0, and 0, respectively. These are descriptive counts; the record does not provide adjusted comparisons or statistical tests for these group summaries.
For other adverse events, the affected and at-risk counts are 36 and 54 for Exercise + Testosterone, 24 and 55 for Exercise + Placebo, and 11 and 20 for Enhanced Usual Care. The registry record does not explain in this summary how events were defined, whether participants could contribute more than one event, or how follow-up exposure was handled. Counts should therefore be read in the context of the reported at-risk group, not as a complete account of event timing, duration, severity, or relatedness.
The listed serious-event terms include Pneumonia (5.0%), Hip Fracture (1.9%), Ruptured diverticulum of colon (1.9%), Chest pain (1.9%), Kidney fibrosis (1.9%), and Musculoskeletal pain (1.9%). The registry record does not specify the denominator or group assignment for these term percentages in the information available. They should not be interpreted as comparative risks without that context. Sparse event counts and the absence of formal comparisons limit conclusions about between-group safety differences.
12. What the Result Does, and Does Not, Mean
The registry record establishes the trial’s question and describes how walking endurance was measured, but it does not report the primary outcome estimates, between-group differences, confidence intervals, or statistical test results. As a result, readers cannot determine the observed direction or size of the effect for Exercise + Testosterone relative to Exercise + Placebo or Enhanced Usual Care. The absence of numerical results is not evidence that the groups had the same outcomes; it means that the primary comparison cannot be evaluated from the registry record.
Even if numerical estimates were available, interpretation would require more than a significance test. In general, the estimate describes the observed difference, while a confidence interval conveys its precision and the range of values compatible with the data under the analysis assumptions. Clinical importance also depends on the relevance of the difference, not solely on whether a statistical threshold is crossed. The registry record does not report a threshold for a clinically meaningful change in walking distance, so that aspect of interpretation is unavailable as well.
The study compared a combined approach of supervised exercise and testosterone with supervised exercise and placebo gel, as well as with Enhanced Usual Care. The comparisons could address different questions: the exercise groups differed by gel assignment, whereas comparison with Enhanced Usual Care also involved differences in the exercise and education programs. The record does not report results that would allow the relative contributions of these components to be assessed. No conclusion about the effect of testosterone alone can be drawn from the described group structure.
The primary outcome population, missing-data details, multiplicity strategy, and full statistical plan are not reported sufficiently to assess how robust any result would be. Safety summaries are descriptive and do not include reported comparative statistical analyses. These limitations concern what can be concluded from the registry record; they do not show that a treatment was effective or ineffective, or that an adverse event was caused by an assigned intervention.
Accordingly, the available information supports a description of the population, interventions, outcome definition, and some safety counts, but not a judgment about treatment-group effects. Results from other sources are not included in the registry record, so they cannot fill these reporting gaps for this analysis.
13. Why This Trial Matters Statistically
This trial is statistically notable for comparing a combined intervention with both an active exercise comparison and an enhanced-care comparison. That structure matters because the two contrasts do not isolate the same component. Exercise + Testosterone versus Exercise + Placebo is the comparison most directly focused on gel assignment within the supervised-exercise setting. Comparisons with Enhanced Usual Care also differ in exercise format and health education. Without numerical results, however, the estimated effects of any of these contrasts cannot be evaluated.
The outcome is a change score in meters, with a defined direction: higher positive values indicate greater improvement in walking distance. Clear units and scoring direction help readers interpret estimates when reported. The registry record also identifies 95% confidence intervals as the dispersion measure, which would allow assessment of precision alongside the group comparisons. Neither the estimates nor interval limits are reported, leaving the central statistical evidence unavailable.
The record identifies many outcomes and analyses, making transparent reporting of prespecification and multiplicity important. It also describes both an unadjusted primary comparison and additional adjusted analyses, while the stated parameter is a least-squares mean. Full model details would help readers understand which estimates answer the primary question and how adjusted results relate to the unadjusted comparison. Those details are not reported in the registry record.
The primary outcome population is conditional on measurement availability at baseline and a follow-up, and the record does not explain missing-data handling or the number included in the 6-month comparison. These details affect the population to which an estimate applies and how secure its interpretation is. The listed safety counts provide descriptive information, but without event definitions, comparative analyses, or fuller context, they cannot independently establish differences in safety.
The trial therefore illustrates a general statistical point: a clearly described design and endpoint are not a substitute for reporting estimates and uncertainty. The registry record offers useful design and measurement details, but its missing outcome statistics prevent readers from judging the primary comparison’s magnitude, precision, or clinical relevance.
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 main limitation for interpreting this trial from the registry record is that numerical primary outcome results are not reported. Group changes, between-group contrasts, confidence intervals, and statistical test results are unavailable. Without these, neither the magnitude nor the precision of the primary comparison can be assessed. The record also does not report a clinically meaningful change threshold for 6-Minute Walk Distance.
The analysis description leaves details unresolved. The primary analysis is described as an unadjusted baseline-to-6-month comparison, while the reported parameter is a least-squares mean and additional analyses of covariance are mentioned. Model specifications, included covariates, and the relation between these descriptions are not reported. The record also mentions analyses involving 3 months, but does not fully explain how these fit with the stated primary timeframe.
The primary outcome population is described as randomized participants (N=122) with baseline and one additional measurement at either 3 Months or 6 Months. The record does not give the number with complete 6-month measurements, reasons for missing measurements, or the method used to address missing data. Enrollment is reported as 129, but the reasons for the difference between that figure and the described primary outcome population are not reported in the registry record.
Multiplicity is another uncertainty: the record lists 11 outcomes and 138 analyses but does not describe a testing hierarchy or adjustment strategy. It also does not specify which analyses were prespecified. The registry classifies masking as quadruple, while the study summary describes the trial as double-blinded; who was masked and how masking was implemented are not reported. Randomization sequence generation and allocation concealment are also not reported.
Safety information is summarized with counts and selected event terms, but definitions, event timing, relatedness, and details of the term-percentage denominators are not reported. These gaps limit comparisons and interpretation. They are limitations of the available registry information and do not establish that bias, missing-data problems, or safety differences occurred.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Starting a Testosterone and Exercise Program After Hip Injury — NCT02938923.
- Registry record last updated 2025-02-04; status last verified 2025-01.
- Results first posted 2025-02-04.
- 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 record describes NCT02938923, a completed phase 3 randomized parallel-group trial sponsored by Washington University School of Medicine. It reports enrollment of 129 and lists hip fracture, frailty, and sarcopenia as conditions. The study summary describes female hip-fracture patients aged 65 and older and a multi-center trial. The trial compared supervised exercise with testosterone gel, supervised exercise with placebo gel, and Enhanced Usual Care.
The primary outcome was change in 6-Minute Walk Distance from baseline to 6 months, measured in meters. Higher positive values represent greater improvement. The primary outcome population is described as randomized participants (N=122) with a baseline measure and one additional measure at 3 Months or 6 Months. The record reports 11 outcome measures and 138 analyses, but numerical primary outcome results and key analysis details are not reported in the registry record.
The record reports serious and other adverse-event counts by group, along with death counts and selected serious-event terms. These summaries are descriptive and lack reported comparative statistical analyses. Start date was 2019-03-05, primary completion was 2023-08-31, and the trial status is completed.