Telephone-linked Home-based Exercise Training in PD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Telephone-linked Home-based Exercise Training in PD (NCT02898740), a randomized na study. The primary outcome is Change in Hamilton Rating Scale for Depression (HAM-D) From Baseline to 6 Months.

2. Clinical Question

This is a question about the measured change in depression-scale scores during the study period, not a direct measure of every aspect of mood, functioning, or quality of life. The registry record identifies the study as randomized and parallel-group, but does not report how the random sequence was generated or how participants were assigned. It also does not specify which participants or study personnel, if any, were masked. Those details matter when assessing how allocation, expectations, and outcome measurement might have influenced the comparison.

3. Trial Design

The registry record describes a randomized, single-masked, parallel-group study of telephone-linked, home-based exercise training in Parkinson’s disease. The listed intervention arms were structured exercise and health education. Enrollment was 40 participants: 21 in the Exercise arm and 19 in the Health Education arm. The record lists the study status as completed and its phase as NA.

The primary outcome was change in HAM-D score from baseline to 6 months. The record defines change as the 6-month score minus the baseline score, so a negative change represents a lower score at 6 months than at baseline. The outcome is reported as a mean, with dispersion labeled Full Range. The analysis method is listed as Mixed Models Analysis, and the reported comparison is a net mean difference between the two groups.

Several design details that would help readers assess conduct and interpretation are not reported in the registry record. These include the randomization procedure, allocation concealment, the identity of the masked group, intervention dose and delivery details, adherence, protocol deviations, and how outcome data were collected. The record does not describe the rationale for the enrollment target or the planned statistical analysis in enough detail to determine whether the reported model and contrast were prespecified. It also does not identify the analysis population, such as whether the analysis included everyone as randomized or only participants with available outcome measurements.

The study dates in the record are a start date of 2017-02-01 and a primary completion date of 2018-10-31. These dates describe the registered study timeline; they do not, by themselves, establish when every participant completed follow-up or when the analysis was conducted.

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

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 Change in Hamilton Rating Scale for Depression (HAM-D) From Baseline to 6 Months, measured over Baseline and 6 months.

The primary endpoint is change in the Hamilton Rating Scale for Depression (HAM-D) from baseline to 6 months. The HAM-D is described in the registry record as a widely used and accepted measure for evaluating depression severity. Scores range from a minimum of 0 to a maximum of 50, with higher scores indicating more severe depression.

The change score is defined as the 6-month score minus the baseline score. Under that definition, a negative change means the measured score was lower at 6 months than at baseline, while a positive change means it was higher. A change score is one way to express within-participant movement over the specified period, although the registry record does not provide individual participant trajectories or explain whether baseline score was included as a separate model term.

HAM-D is used because it provides a structured measure of depression severity rather than relying only on a broad description of mood. The score remains a scale measurement; it does not itself identify why a score changed or establish how a participant experienced that change. The registry record does not provide a threshold for interpreting the clinical importance of a change, nor does it report supporting outcomes that would show whether HAM-D change corresponded to other aspects of participants’ health or daily lives.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry names Mixed Models Analysis as the method for the primary comparison. In general, a mixed model can represent outcomes measured across participants and can account for relationships among repeated measurements when the model is specified to do so. Its exact form depends on the outcome structure, included terms, assumptions, and handling of incomplete observations. The registry record does not report those model details, such as the fixed effects, random effects, covariance structure, adjustment variables, or estimation method. It is therefore possible to describe the stated method, but not to reconstruct the analysis specification from the record.

What does a mean change describe?

A mean change summarizes the average difference between follow-up and baseline scores within an arm. For this endpoint, the direction is defined as follow-up minus baseline. The Exercise arm’s reported mean change is -1.4, and the Health Education arm’s is 2.0. These averages do not imply that every participant changed in the same direction. Individual outcomes can vary around an arm mean, and the posted full ranges show that observed changes extended on both sides of zero in each arm.

The registry labels the arm-level dispersion as Full Range. A range conveys the lowest and highest reported values under the registry’s presentation, but it does not quantify the typical distance of observations from the mean. It is distinct from a standard deviation, which describes spread around the mean, and from a confidence interval, which expresses uncertainty about an estimate. Since standard deviations are not reported, the range should not be converted into one or used as if it were a standard error.

How should the mean difference be read?

The reported net mean difference is -3.4, with a 95% confidence interval from -6.8 to -0.0. In general, a mean difference expresses a contrast between group means, but its direction depends on which group is subtracted from which. The record lists the groups as Exercise and Health Education and identifies Exercise as the control group, without explicitly defining the contrast order. Readers should retain that uncertainty when interpreting the sign. The estimate should also be read in the context of the arm-specific changes rather than as a standalone description of either arm.

A confidence interval gives a range of values that, under the analysis assumptions, is compatible with the observed data and model. It reflects uncertainty in the estimated contrast, not the range of individual participant changes. A narrower interval generally indicates greater precision than a wider one, but the registry does not provide a prespecified threshold for a clinically important difference. Statistical precision and clinical importance are separate questions: an interval and p-value do not establish whether a difference would matter to participants.

What does the p-value add?

The registry reports p <0.05. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions were correct. It is not the probability that the null hypothesis is true, nor is it a measure of effect size. The reported threshold does not provide an exact p-value. The registry record does not state whether this test was one- or two-sided, describe any adjustment for multiple testing, or provide the full test statistic.

What assumptions affect the analysis?

Every model relies on assumptions about how outcomes relate to predictors and how errors or repeated measurements behave. A mixed model can accommodate certain data structures, but the method name alone does not reveal whether its assumptions were suitable or whether diagnostics were performed. The registry record does not report checks of model fit, residual behavior, influential observations, or sensitivity analyses. These omissions limit independent assessment of robustness.

Missing data can also affect model-based estimates. The record does not report how many participants had HAM-D measurements at baseline and 6 months, whether any observations were missing, or how missingness was handled. General methods for incomplete repeated outcomes rely on assumptions about the process that generated missing data. Without details on missingness and the chosen approach, readers cannot assess whether those assumptions are plausible in this study.

How do planning illustrations differ from trial power?

General sample-size planning calculations ask how many participants might be needed to detect a chosen effect under specified assumptions. The planning framework associated with this record uses a two-sided significance level of 0.05, equal allocation, and a normal approximation; it considers effects defined as fractions of the observed effect. Such illustrations are planning exercises, not observed power and not evidence that the completed study had a particular probability of detecting an effect. The registry record does not report the trial’s own sample-size justification or enough detail to assess whether its enrollment target was based on a prespecified effect and variance.

In general, planning calculations depend on the target effect, outcome variability, significance level, allocation, and desired power. Choosing a smaller target effect generally requires more information to detect it, all else equal. A calculation based on an observed effect should be interpreted cautiously because that estimate is itself uncertain. The record’s general planning assumptions do not establish that these were the assumptions used to design the study.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a Mixed Models Analysis, with a net mean difference of -3.4, a 95% confidence interval from -6.8 to -0.0, and p <0.05. This is the inferential result reported for the primary outcome. The arm-level means and full ranges are descriptive summaries and should not be substituted for that model-based comparison.

A separately recalculated two-arm comparison is not reported in the registry record. The record does not provide the standard deviations or other information needed to reconstruct an unadjusted comparison from summary statistics, and it does not give the model details needed to reproduce the mixed-model result. As a result, there is no independent numerical recalculation to reconcile with the sponsor’s estimate. The contrast coding is also not stated explicitly, so the sign should be considered together with the group-specific changes.

The registry record does not identify a discrepancy between the reported analysis and a recalculation. This does not establish that the model has been independently verified; it means the posted record supplies the model result but not enough detail for a numerical replication. The sponsor analysis should therefore be described as reported, with its precision and reporting limitations, rather than replaced by an unsupported alternative estimate.

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and one analysis, with HAM-D change from baseline to 6 months identified as the primary endpoint. It lists no secondary posted outcomes. This limits the ability to compare the primary finding with other outcome domains or to evaluate whether different measures tell a consistent story.

A multiplicity plan addresses the possibility of false-positive findings when several outcomes, time points, groups, or analyses are tested. The record does not report whether any multiplicity adjustment was planned or applied, nor does it describe a hierarchy of outcomes or analyses. Since only one outcome measure and one analysis are listed, the record does not expose a broader family of posted tests, but the absence of such listings does not establish what may have been prespecified or conducted outside the registry entry.

Missing-data methods and analysis populations are not reported in the registry record. The entry does not state how many participants contributed baseline and 6-month HAM-D measurements, whether any data were missing, or how incomplete measurements were handled. It also does not say whether the analysis followed randomized assignment, excluded participants without follow-up, or used another population definition. These details affect how readers assess the interpretation of a model-based estimate.

The record does not report interim analyses, stopping rules, or the trial’s sample-size justification. These should not be inferred from enrollment or from planning illustrations. Without those details, the registry entry does not establish what decisions were made during the study or whether the final analysis followed a prespecified plan. This is a limitation of what the record documents, not evidence that particular procedures were or were not used.

12. 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 5% reporting threshold.

The registry reports zero serious adverse events among 21 participants at risk in the Exercise arm and zero among 19 participants at risk in the Health Education arm. It also reports zero other adverse events in each arm and zero deaths in each arm. These are the recorded counts for the groups listed.

Zero reported events means that the registry entry lists none for those categories and groups. It does not establish that no event of any kind occurred, or that the groups had identical safety experience. The record does not provide details about how adverse events were defined, collected, reviewed, or attributed, or specify the period covered by these counts. It also provides no serious-event terms because no serious events are listed.

Event counts should be interpreted in relation to the number of participants at risk and the completeness of reporting. No rates or comparative safety estimates are reported, and none should be inferred from these counts alone. The absence of listed events in this registry entry does not resolve how uncommon events, delayed events, or events not captured by the study’s reporting procedures should be understood.

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

The registry record reports a model-based comparison of change in HAM-D score between the Exercise and Health Education groups. The estimated net mean difference is -3.4, with a 95% confidence interval from -6.8 to -0.0 and p <0.05. At the arm level, the reported mean changes are -1.4 for Exercise and 2.0 for Health Education. These results describe the registered primary outcome and analysis, not every possible effect of either program.

The endpoint measures depression severity on a scale, and its change is calculated as the 6-month score minus the baseline score. The reported group averages therefore summarize score movement over that period. They do not show individual responses, explain why scores changed, or establish the effect on outcomes that were not measured or posted. The record does not define a threshold for a clinically important change, so the statistical result cannot be translated into a claim about clinical importance from this information alone.

The confidence interval conveys uncertainty around the model estimate under its assumptions. The p-value is reported as a threshold, and neither statistic measures the likelihood that a particular participant will experience a given outcome. Interpretation also depends on the model specification, contrast direction, handling of missing data, and analysis population, details that are not reported in the registry record.

The findings should not be taken as proof that the intervention will produce the same outcome in other populations or settings. The record describes 40 enrolled participants in this study, but provides limited information about who they were, how the interventions were delivered, or how consistently participants engaged with them. Generalization and causal interpretation require attention to those design and reporting details, as well as to the uncertainty in the estimate.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it presents both arm-level change summaries and a model-based comparison. Those quantities answer related but different questions. The arm means describe average change within each group, while the net mean difference describes a contrast between groups under the reported analysis. Keeping these distinct helps prevent a within-group change from being mistaken for the comparative effect of group assignment.

The registry also illustrates why direction and contrast coding matter. The primary endpoint’s change definition gives a clear meaning to positive and negative arm-level values, but the exact subtraction order for the net mean difference is not stated. A signed contrast can be misread if its group order is assumed rather than documented. Reading the estimate alongside the arm summaries and noting what the registry does not specify supports a more careful interpretation.

Precision is another central issue. The reported confidence interval expresses uncertainty about the estimated contrast, while the full ranges describe the spread of reported arm-level changes. Those summaries are not interchangeable. The record does not provide standard deviations or model details that would allow a reader to independently reproduce the analysis, so transparent reporting of methods and missing-data handling would be important for evaluating the result.

Finally, the registry reports one primary outcome measure and one analysis, but does not describe multiplicity procedures, analysis populations, or the trial’s sample-size rationale. These omissions do not invalidate the posted estimate, but they constrain what can be concluded about its robustness and place in a broader analysis plan. The study is a useful example of how statistical interpretation depends not only on a point estimate, but also on design detail, uncertainty, and reporting completeness.

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.

Several limitations specific to this trial concern the detail available in the registry record. The intervention is identified as structured exercise, and the title describes telephone-linked home-based training, but the record does not specify the exercise prescription, delivery schedule, telephone support, adherence, or intervention fidelity. The Health Education comparator is likewise not described in enough detail to assess content, contact time, or similarity of attention between arms.

The record calls the study single-masked but does not identify who was masked. It does not describe sequence generation or allocation concealment. Without those details, readers cannot assess how assignment was generated or protected from foreknowledge. The record also does not report participant characteristics beyond the Parkinson’s disease condition, limiting assessment of how representative the enrolled group may be.

Outcome reporting includes the mean change and full range in each arm, but no standard deviations, participant-level data, or baseline score summaries. The reported mixed-model method is not accompanied by the model specification, contrast coding, diagnostics, or sensitivity analyses. The registry does not report missing-data counts or methods, the analysis population, or adherence information. These omissions limit independent replication and assessment of robustness.

The record lists one outcome measure and one analysis, but does not report a multiplicity plan, a sample-size justification, or interim-analysis procedures. It also gives no clinical-importance threshold for HAM-D change. Consequently, the posted statistical result cannot by itself establish the practical meaning of the estimated difference, the completeness of the analysis plan, or how directly it applies beyond the participants and conditions represented in this trial.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Telephone-linked Home-based Exercise Training in PD — NCT02898740.
• Registry record last updated 2019-11-21; status last verified 2019-10.
• Results first posted 2019-11-21.
• 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.

19. Record Summary

NCT02898740 is a completed, randomized, single-masked, parallel-group study titled Telephone-linked Home-based Exercise Training in PD. The condition is Parkinson’s disease. The listed arms are Exercise, described as structured exercise, and Health Education. The registry record lists 40 enrolled participants, with 21 in Exercise and 19 in Health Education.

The primary outcome is change in HAM-D score from baseline to 6 months, defined as the 6-month score minus the baseline score. The registry reports mean changes of -1.4 for Exercise and 2.0 for Health Education, and a mixed-model net mean difference of -3.4 with a 95% confidence interval from -6.8 to -0.0 and p <0.05. The record does not specify the contrast coding or provide enough model detail for independent numerical replication.

The registry reports zero serious adverse events, zero other adverse events, and zero deaths in both arms. It does not report secondary posted outcomes, multiplicity procedures, missing-data methods, analysis populations, or the trial’s sample-size justification.