A Physical Therapy Program Versus Surgery for Femoroacetabular Impingement: Randomized Clinical Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Physical Therapy Program Versus Surgery for Femoroacetabular Impingement: Randomized Clinical Trial (NCT01993615), a randomized na study. The primary outcome is Hip Outcome Score - ADL Subscale at 24 Months.

2. Clinical Question

The question concerns outcomes under two treatment strategies in a selected population: people already considered candidates for surgery. It is not a comparison in an unselected population with hip symptoms, nor does the record establish that participants could be masked to the treatment received. Masking is listed as single, but who was masked is not reported in the registry record. The trial followed participants for a two-year period, and the posted primary endpoint is measured at 24 months. The registry also lists secondary outcomes concerning the International Hip Outcome Score and global change. Their presence broadens the outcome context, but does not replace the primary question or supply a formal primary comparison.

3. Trial Design

The registry identifies a randomized, parallel-group trial comparing arthroscopic surgery with physical therapy for femoroacetabular impingement. Enrollment was 80. The trial is listed as completed, with Madigan Army Medical Center as sponsor. The phase is listed as NA. The study summary says enrollment was limited to patients already determined to be surgical candidates, which defines the population to which the results most directly apply.

The surgery group received arthroscopy at the femoroacetabular joint followed by a standardized postoperative physical therapy rehabilitation protocol. The physical therapy group received an impairment-based supervised in-clinic program. The summary describes the latter as a six-week program. Because the surgical group also received rehabilitation, the comparison is between treatment pathways, not surgery versus no physical therapy.

The primary endpoint was assessed at 24 months. The registry describes its analysis population as an intention-to-treat analysis including participants who received any treatment; six enrolled participants who never returned for treatment were excluded. Thus, the label does not mean every randomized participant contributed to the posted primary analysis. The endpoint estimates use 37 participants in each group, while the safety figures use 40 participants at risk in each group. Those populations should not be conflated.

Masking is recorded as single, but the masked party or parties are not reported in the registry record. The allocation ratio, random-sequence generation, concealment procedures, and detailed conduct of follow-up are not reported in the registry record. The start and primary completion entries are 2013-03 and 2016-07. The record does not provide a detailed statistical analysis plan, original sample-size justification, or interim-analysis information.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intention to treat analysis including only participants that received any treatment (6 participants that were enrolled never returned for treatment)

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Hip Outcome Score - ADL Subscale at 24 Months, measured over 24 months.

The primary endpoint is the Hip Outcome Score Activities of Daily Living subscale at 24 months. The registry describes the Hip Outcome Score as an instrument intended to assess higher-level activities, including activities relevant to athletics, and notes prior evidence supporting its use in people assessed after hip arthroscopy. For this trial, the endpoint focuses on reported ability to perform activities of daily living.

The subscale contains 19 questions, each scored on a five-point scale from 0, meaning unable to do, to 4, meaning no difficulty. The final score is calculated by dividing the raw score by the number of questions answered multiplied by 4, then multiplying by 100. Scores range from 0 to 100%, with lower scores associated with greater limitation in activities of daily living. These properties make the measure interpretable as a continuous score, with higher values indicating less reported limitation on the described scale.

A patient-reported score can capture aspects of function that are not represented by an event count alone. It also has limitations: interpretation depends on which questions participants answer and on how the scale reflects their experience. The registry posts group least-squares means with 95% confidence intervals at 24 months. It does not post a direct treatment contrast for this endpoint, and the registry record does not report how missing item responses or missing 24-month assessments were handled beyond its stated analysis-population description.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: it is a score on a scale rather than a count of events. The registry reports a least-squares mean for each arm and a 95% confidence interval. A mean describes the group’s average score; a least-squares mean is generally an estimate from a statistical model, potentially adjusted for terms in that model. The model terms and adjustment details are not reported in the registry record, so the exact meaning of the adjustment cannot be reconstructed from the posted outcome entry.

A confidence interval expresses the uncertainty of an estimate under the statistical procedure used. It is not a guarantee that the true value lies within the particular interval, and it does not describe the spread of individual scores. A narrower interval would generally indicate greater precision for the estimated mean, while a wider interval indicates less precision. Here, the registry gives separate intervals for the two arm means, but not a confidence interval for their difference. Separate arm intervals cannot substitute for a direct interval on the contrast because the uncertainty of the contrast depends on the joint comparison and its analysis.

Random allocation can help make treatment groups comparable at the start of a trial, on average, when implemented appropriately. It does not ensure that every measured feature is balanced, and it does not remove uncertainty from a modest sample. Details of sequence generation and allocation concealment are not reported in the registry record. The record describes a randomized parallel design, but does not report a direct inferential analysis of the primary outcome. For that reason, the posted arm estimates are descriptive evidence rather than a complete test of the comparative question.

What does the analysis population imply?

The primary outcome description calls the analysis intention to treat but restricts inclusion to participants who received any treatment. It also specifies that six enrolled participants never returned for treatment. In usual statistical usage, intention-to-treat analysis retains participants in their randomized groups regardless of treatment received; the registry’s stated restriction is narrower than that conventional meaning. The record does not clarify the treatment of other protocol deviations or missing outcome assessments. Readers should therefore use the population definition actually stated rather than assume a full randomized-population analysis.

The primary endpoint estimates use 37 participants in each arm. The adverse-event record instead lists 40 participants at risk in each arm. These denominators answer different descriptive questions and arise from distinct posted outcome records. They should not be combined into one sample size or treated as if every endpoint had identical completeness.

How should the secondary recalculation be read?

The global-rating comparison includes a recalculated risk difference and an unadjusted p-value based on posted participant counts. A risk difference expresses the difference in proportions between two groups; its sign depends on the order of groups used. A confidence interval for a risk difference describes uncertainty in that absolute contrast. When an interval includes zero, the data are compatible with no difference as well as with effects in either direction, under the calculation’s assumptions. The posted interval for this recalculation spans zero. This is not a sponsor-reported analysis: the record lists no reported analyses, and the figures are labelled as recalculated.

Unadjusted means the calculation does not account for other variables or for a multiplicity plan. It should not be read as a preplanned confirmatory test unless the registry documents that status. The record does not report the original analysis specification for this secondary outcome. No p-value or direct contrast is posted for the continuous primary endpoint.

What do the planning illustrations represent?

The sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design and not observed power. They assume a two-sided significance level of 0.05, equal allocation, and a normal approximation; the requested effect is a stated fraction of the observed effect. Such calculations illustrate how a chosen target effect and statistical assumptions relate to a possible future sample size. They do not show the probability that this completed trial detected an effect, nor do they justify the trial’s original enrollment. The original sample-size justification is not reported in the registry record.

More generally, planning based on an observed effect can be unstable because the observed estimate itself is uncertain. Choosing a smaller target effect leads to a larger planning requirement in the posted illustrations, but those numbers remain conditional on their assumptions. They do not account for any unreported design features, missing outcomes, or analysis choices. Interim analyses and any associated stopping rules are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures, identifies secondary outcomes, and posts one secondary comparison with a recalculated unadjusted p-value. It does not report a multiplicity adjustment, a testing hierarchy, or a plan specifying which secondary outcomes were confirmatory. When several outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase. An unadjusted secondary result does not account for that issue and should not be interpreted as though it were the sole planned test.

The primary outcome entry gives arm-level least-squares means and confidence intervals but no direct contrast or p-value. The record therefore does not state whether a formal primary hypothesis test was conducted or what inferential method was used for it. A detailed statistical analysis plan is not reported in the registry record. Nor are interim analyses or stopping rules reported. Their absence from the record is not evidence that none occurred; it means the registry does not document them.

Missing-data handling is also incompletely described. The primary outcome population includes participants who received any treatment and excludes six enrolled participants who never returned for treatment. The record does not specify the handling of any additional missing 24-month outcome values, item-level missingness, or other departures from the planned analysis population. It does not report imputation, sensitivity analyses, or assumptions about missing outcomes. The endpoint sample counts therefore describe the analyzed observations but do not explain how incomplete follow-up may have affected the estimates.

The registry’s intention-to-treat label should be read together with its stated eligibility restriction for the analysis. Conventional intention-to-treat analysis generally retains participants according to randomized assignment, including those who do not receive treatment. Since six enrollees who never returned were excluded from the stated primary population, the posted definition is not a full all-randomized analysis as usually understood. The record does not explain why that population was selected or provide a comparison of results under alternative populations.

These reporting gaps constrain interpretation, but they do not establish that the analyses were inappropriate. They mean that readers cannot verify key aspects of the planned inferential approach from the registry record.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event record lists 40 participants at risk in each group. No serious adverse events are recorded in either group. Other adverse events are recorded for two participants in each group. The death count is one in the arthroscopic surgery group and zero in the physical therapy group.

These are descriptive counts, not a comparative safety analysis. The registry does not provide serious-event terms, and the record does not describe event timing, severity categories beyond the serious designation, attribution to treatment, or statistical comparisons. The death entry should not be assumed to be caused by treatment; a causal relationship is not reported. Likewise, zero recorded serious events does not establish that the risk is absent. With the posted denominators and counts, the safety information is limited to what was recorded for these participants, and it should not be used to infer comparative safety beyond the trial’s reported observations.

Safety denominators differ from the 37 participants per group in the primary endpoint analysis. That distinction reflects the separate outcome records and means safety counts should not be interpreted as if they use the primary endpoint population. The registry record does not report further detail needed to assess the completeness of adverse-event collection or how events were adjudicated.

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

The posted primary outcome describes functional scores at 24 months among participants included in the registry’s stated analysis population. The physical therapy group’s least-squares mean is higher than the surgery group’s mean, and higher scores indicate less limitation on the described scale. However, arm-specific means do not by themselves quantify the treatment contrast. The registry does not report a direct primary-outcome estimate, interval for that contrast, or p-value.

Accordingly, the results do not establish that one strategy is better, worse, equivalent, or noninferior. Equivalence and noninferiority require an appropriately prespecified margin and analysis; neither is reported in the registry record. A non-significant result, even if one were available, would not by itself establish equivalence. Likewise, a numerical difference between group estimates is not automatically evidence of a meaningful or reliable treatment effect.

The trial’s population is also important to the scope of interpretation. Participants had already been determined to be surgical candidates. Findings from this selected group should not automatically be extended to people who were not considered candidates, to other patient groups, or to different surgery or rehabilitation protocols. The physical therapy group received a supervised in-clinic program, while the surgical group received postoperative rehabilitation as well as surgery. The comparison concerns those care pathways as described, not surgery compared with the absence of rehabilitation.

The registry reports a follow-up period of two years and a primary outcome at 24 months, but the record does not describe the completeness of all follow-up assessments or the handling of additional missing measurements. The six enrolled participants who never returned were excluded from the stated primary analysis population. This limits how confidently the posted estimates can be generalized to all enrolled participants.

The most defensible reading is therefore descriptive and bounded: the registry provides estimated group scores and their uncertainty for the analyzed participants, but it does not provide the direct primary comparison needed to draw a formal conclusion about the relative effects of the two strategies.

13. Why This Trial Matters Statistically

This trial is statistically informative because it compares two active treatment pathways in a randomized parallel design and reports a patient-reported functional endpoint at a defined follow-up time. The treatment strategies are clinically distinct, yet the surgery pathway includes postoperative rehabilitation. That distinction matters when defining the estimand: the comparison is between the assigned pathways as conducted, rather than between a procedure and no accompanying care.

The record also illustrates why arm-level results and comparative results are not interchangeable. Group means and confidence intervals describe each group, but a direct contrast is needed to estimate the difference of interest and its uncertainty. Without that contrast for the primary outcome, readers cannot assess the primary comparative question using the posted primary statistics alone. This is a reporting limitation, not a reason to infer an unreported test or result.

Population definitions matter as much as endpoint choice. The registry calls the primary analysis intention to treat while restricting it to participants who received any treatment. It explicitly excludes six enrolled participants who never returned for treatment. Clear reporting of randomized, treated, analyzed, and followed populations helps readers understand what each estimate represents and assess possible selection introduced by missing participation.

The record also distinguishes primary and secondary evidence. A recalculated, unadjusted secondary comparison may offer exploratory context, but it is not a substitute for a reported primary analysis and does not account for multiple outcomes. The registry’s lack of a documented multiplicity plan, missing-data methods, and original sample-size rationale limits reconstruction of the intended inferential framework.

Finally, the planning illustrations are not observed power and are not proof of the trial’s design adequacy. They depend on stated assumptions and an effect drawn from observed data. Keeping those calculations separate from the trial’s actual results prevents planning quantities from being mistaken for evidence about what the completed study established.

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 important trial-specific limitation in interpreting the primary result is the absence of a posted direct comparison between the two arm means. The registry gives least-squares means and confidence intervals for each group, but not a contrast estimate, its confidence interval, or a primary p-value. This prevents a direct assessment of the magnitude and precision of the primary treatment difference from the registry results.

The analysis population warrants care. Although described as intention to treat, it includes only participants who received any treatment and excludes six enrolled participants who never returned for treatment. The record does not report outcomes for those excluded participants or explain the reason for the restriction. It also does not describe how other missing endpoint data were handled, so the potential impact of incomplete 24-month measurement cannot be assessed from the registry record.

The population was restricted to people already determined to be surgical candidates. That narrows the applicability of the results. The treatment pathways also differ in more than the procedure itself: the surgery arm included standardized postoperative physical therapy, while the comparison arm received a supervised impairment-based program. The registry does not provide enough detail to evaluate how differences in treatment intensity, adherence, or co-interventions were handled.

Other limitations are reporting gaps rather than established flaws. The registry does not report who was masked, how allocation was generated or concealed, a detailed statistical analysis plan, multiplicity handling, original sample-size justification, or interim analyses. Numerical results for the listed IHOT33 secondary outcome are not reported in the registry record. The safety record provides counts but little detail on event characterization or attribution. These omissions constrain independent evaluation and should not be filled by assumptions.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Physical Therapy Program Versus Surgery for Femoroacetabular Impingement: Randomized Clinical Trial — NCT01993615.
• Registry record last updated 2020-08-14; status last verified 2020-07.
• Results first posted 2020-08-14.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

The registry describes a completed randomized, parallel-group trial of arthroscopic surgery versus supervised physical therapy for femoroacetabular impingement. Enrollment was 80, and participants were limited to those already considered surgical candidates. The surgery pathway included postoperative physical therapy; the other group received a supervised impairment-based in-clinic program described as six weeks in duration.

The primary endpoint was the Hip Outcome Score Activities of Daily Living subscale at 24 months. Arm-specific least-squares means and 95% confidence intervals are posted, based on 37 participants in each arm. The stated primary analysis excluded six enrolled participants who never returned for treatment. No direct primary comparison is reported in the registry record.

The registry lists secondary outcomes, but numerical IHOT33 results are not reported. The global rating comparison is a recalculation rather than a listed sponsor analysis. Safety counts show no serious adverse events recorded, with other events and deaths reported as described in the safety information. Important analysis details, including missing-data methods, multiplicity handling, and the original sample-size justification, are not reported in the registry record.