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 Telehealth Pulmonary Rehabilitation for Hispanic and African-American Patients Admitted With Exacerbation of COPD (NCT03007485), a randomized na study. The primary outcome is Composite of COPD Hospital Readmission/Death Within 6 Month of Discharge.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03007485 |
| Phase | NA |
| Condition | COPD Exacerbation |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Double |
| Enrollment | 266 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Northwell Health |
| Results posted | Yes |
2. Clinical Question
Population
Hispanic and African-American patients admitted to hospital with COPD exacerbation. The registry lists a planned enrollment of 266.
Intervention
Telehealth-delivered pulmonary rehabilitation, described as exercise and education delivered remotely, with exercise at home supervised by a respiratory therapist through videoconference.
Comparator
Standard pulmonary rehabilitation. The registry identifies the two arms as Telehealth-delivered Pulmonary Rehabilitation and Standard of Care.
Primary question
How does the primary composite of COPD-related hospital readmission or death within 6 months of discharge compare under the prespecified analysis? A single estimate for that comparison is not reported in the registry record.
The study addresses access to pulmonary rehabilitation after hospitalization for COPD exacerbation. The registry's background describes barriers that may limit participation in center-based rehabilitation, including transportation, insurance, and availability of a rehabilitation center. Telehealth is presented as a way to deliver supervised exercise and education without requiring travel to a rehabilitation facility.
The treatment contrast is telehealth pulmonary rehabilitation versus standard pulmonary rehabilitation, not rehabilitation versus no rehabilitation. The registry describes the primary outcome as a composite endpoint and says that the TelePR and SPR arms were combined as prespecified in the study protocol for the primary outcome. It also describes logistic regression to compare the outcome in terms of an odds ratio of event rate. The exact primary-outcome estimate, its confidence interval, and arm-specific event counts are not reported in the registry record, so the direction and precision of a primary comparison cannot be assessed from the recorded information.
3. Trial Design
This is a randomized, double-masked, parallel study listed as completed. The phase is recorded as NA. Northwell Health is the sponsor. The registry lists enrollment as 266. The recorded study start is 2017-04-01, and primary completion is 2020-11-15.
The comparison is between standard pulmonary rehabilitation and telehealth-delivered pulmonary rehabilitation. The registry's description of telehealth rehabilitation includes home exercise on a stationary bike, supervision by a respiratory therapist through videoconference, and education delivered remotely. The standard-care arm is described as standard pulmonary rehabilitation. Details about the precise content, duration, or frequency of either program are not reported in the registry record.
The primary endpoint is the composite of COPD hospital readmission or death within 6 months of discharge following hospitalization for COPD exacerbation. The record describes use of all available complete data, with no imputation of missing data. It also states that the TelePR and SPR arms were combined for the primary outcome as prespecified in the study protocol, and identifies logistic regression as the analysis approach. A single primary-outcome comparison is not reported in the registry record.
Secondary outcomes include functional capacity assessed with the 6-minute walk test in the standard pulmonary rehabilitation group and the 2-minute step test in the telehealth group, as well as self-reported quality of life measured around the pulmonary rehabilitation program and over follow-up. The registry lists five outcome measures and one analysis. The exact allocation ratio, masking roles, recruitment locations, and detailed eligibility criteria are not reported in the registry record. The study's completed status indicates that the record marks the study as completed; it does not itself establish the findings of the primary comparison.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Before assignment: In total 266 participants signed the informed consent and were randomized (131 to the TelePR arm and 135 to the SPR arm). Of the 131 randomized to TelePR, 20 were subsequently excluded after randomization because they no longer met inclusion criteria. Of the 135 randomized to SPR, 37 were excluded after randomization because they also no longer met inclusion criteria. Therefore, in total 209 participants were randomized and were included in the intention to treat analysis.
Overall Study
| Stage | Standard of Care | Telehealth-delivered Pulmonary Rehabilitation |
|---|---|---|
| STARTED | 98 | 111 |
| COMPLETED | 28 | 57 |
| NOT COMPLETED | 70 | 54 |
Analysis population for the primary outcome, as defined in the registry: TelePR and SPR arms were combined as pre-specified in the study protocol and explained below: The primary outcome was a composite of COPD-related hospital readmissions or death within 6 months of discharge. Logistic regression was used to compare the primary outcome in terms of the OR of event rates between arms, in 2 sets of models: * 1: Treatment arm only with no other covariates added to the model * 2: Treatment arm, adjusted for race and clinical site (stratification variables)
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 | Standard of Care | Intervention | Total |
|---|---|---|---|---|
| Participants analysed | 98 | 111 | 209 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 43 | 48 | 91 |
| Age, Categorical: >=65 years | Participants | 55 | 63 | 118 |
| Age, Continuous | years | 65.96 (SD 10.68) | 66.89 (SD 10.80) | 66.42 (SD 10.70) |
| Sex: Female, Male: Female | Participants | 57 | 67 | 124 |
| Sex: Female, Male: Male | Participants | 41 | 44 | 85 |
| Race/Ethnicity, Customized: Hispanic | Participants | 41 | 48 | 89 |
| Race/Ethnicity, Customized: African-American | Participants | 57 | 63 | 120 |
| Region of Enrollment: United States | participants | 98 | 111 | 209 |
6. Primary Endpoint
The registry lists the primary outcome as Composite of COPD Hospital Readmission/Death Within 6 Month of Discharge, measured over 6 months post-discharge from hospitalization following COPD exacerbation.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | 95% Confidence Interval |
| Unit | Hospitalizations/death |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is a composite of COPD-related hospital readmission or death within 6 months of discharge from a hospitalization following COPD exacerbation. It is described in terms of the number of hospitalizations or deaths. The registry says the investigators intended to analyze change in the rate of patients' rehospitalizations following pulmonary rehabilitation, and describes logistic regression for comparing the primary outcome in terms of an odds ratio of event rate.
As a general statistical concept, a composite endpoint combines more than one event type into a single outcome. This can capture more than one clinically important event in one analysis. Its interpretation depends on the components: readmission and death are different events, and a composite result alone would not show whether any overall pattern was driven by one component or the other. Component-specific results are not reported in the registry record.
The registry describes the composite as based on all available complete data, with no imputation of missing data. It also says the TelePR and SPR arms were combined for the primary outcome as prespecified in the study protocol. However, the registry record does not report the primary-outcome event counts by arm, the estimated odds ratio, its confidence interval, or a p-value. It therefore does not support a conclusion about the size or precision of a primary-outcome comparison.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry identifies a primary composite endpoint, a follow-up period of 6 months after discharge, and logistic regression as the analysis method. It does not provide the numerical estimate for the primary outcome. In particular, the event counts by study arm, odds ratio, confidence interval, and p-value are not reported in the registry record. Without those quantities, readers cannot establish the direction, magnitude, or statistical precision of the primary comparison from this record.
The analysis description says that the TelePR and SPR arms were combined for the primary outcome as prespecified in the protocol. That statement does not give an estimate of the comparison between telehealth and standard pulmonary rehabilitation. The record also refers to comparing the primary outcome using an odds ratio of event rate. How the combined-arm description relates to the intended contrast is not fully explained in the registry record. A clear arm-level primary-outcome result is therefore not available for interpretation.
The registry lists a 95% confidence interval as the measure of dispersion for the primary outcome, but does not report the interval itself. In general, a confidence interval communicates the precision of an estimated effect and the range of values compatible with the data under the analysis. Without the estimate and its interval, neither the size of the association nor its uncertainty can be judged. A p-value is also not reported in the registry record, so there is no basis for characterizing statistical significance.
The record includes safety figures separately: serious adverse events affected 3 participants in Standard of Care and 3 in Telehealth-delivered Pulmonary Rehabilitation; other adverse events affected 3 and 13, respectively; and deaths were recorded as 2 and 3, respectively. These descriptive counts do not substitute for the primary endpoint analysis. The registry does not explain whether the safety deaths overlap with deaths in the primary composite, nor does it provide a safety comparison adjusted for follow-up or other factors.
Accordingly, the documented information supports a description of the study design, intended endpoint, analysis approach, and recorded safety counts. It does not support a numerical conclusion about whether the primary composite differed between rehabilitation approaches, how large such a difference might be, or how precisely it was estimated. No inference about a treatment effect should be drawn from the absence of a reported primary estimate.
9. Statistical Methods Explained
The registry describes a randomized, double-masked, parallel design and a primary composite of COPD-related hospital readmission or death within 6 months of discharge. It identifies logistic regression as the method used for the primary outcome and describes the comparison in terms of an odds ratio of event rate. The actual model output and supporting event counts are not reported in the registry record.
What does logistic regression estimate?
As a general statistical explanation, logistic regression models the odds of a binary outcome, such as whether a participant experienced an event during a defined period. An odds ratio compares the odds in one group with the odds in another. It is not the same quantity as a difference in event probabilities or a ratio of probabilities. Interpretation depends on the outcome definition, the groups being compared, and whether the model includes adjustment variables.
The registry does not report the odds ratio, confidence interval, p-value, or any covariates used in the model. It also does not report arm-specific primary-outcome counts. Thus, the analysis method can be described, but its numerical result and the degree of adjustment cannot be evaluated. The phrase “event rate” appears in the analysis description, but additional details about the outcome coding and model specification are not reported in the registry record.
What is the role of the composite endpoint?
As a general concept, a composite endpoint treats any of its component events as a qualifying outcome. Combining COPD-related readmission and death may capture more than one kind of adverse clinical outcome during follow-up. However, the composite does not automatically give equal clinical or statistical importance to each component. The observed composite can be influenced more by a component that occurs more often. Component-specific counts or estimates would help clarify the pattern, but they are not reported in the registry record.
The registry says the TelePR and SPR arms were combined for the primary outcome as prespecified in the study protocol. It also says logistic regression was used to compare the primary outcome by odds ratio of event rate. The relationship between combining the arms and the precise comparison represented by the regression is not sufficiently detailed in the record to reconstruct the contrast. The arm descriptions identify standard and telehealth rehabilitation, but no primary-outcome estimate comparing those arms is reported.
What does the confidence interval contribute?
The primary outcome is associated with a stated 95% confidence interval, but the interval's numerical limits are not reported. In general, a confidence interval conveys uncertainty around an estimated quantity. A narrow interval usually indicates greater precision than a wide interval, while an interval that includes values corresponding to no difference may be compatible with a range of interpretations. These principles cannot be applied to this study's primary endpoint without the numerical estimate and limits.
A confidence interval does not by itself explain whether an observed difference is clinically important. It is interpreted alongside the outcome definition, estimate, study design, and limitations. Here, the registry record does not report the primary estimate or interval, so precision and the range of compatible effects cannot be assessed.
How should the missing-data description be read?
The registry states that the primary composite used all available complete data and that missing data were not imputed. In general, an analysis of available complete cases uses records with sufficient information for the outcome analysis; it does not fill missing values with modeled or substituted values. This approach can reduce the information available for analysis. If missingness is related to health status or study group, the analyzed participants may differ from those without complete information. The registry does not report the amount or pattern of missing primary-outcome data, so the possible effect on the estimate cannot be assessed.
The endpoint description also states that, without adherence added, an offset term was omitted in the logistic regression. The record does not explain this modeling choice further or describe a version of the model that included adherence. Details such as the covariates, assumptions, diagnostic checks, handling of repeated observations, or model-fitting procedures are not reported in the registry record. General explanations of logistic regression do not fill those gaps.
What can be learned from planning calculations?
Planning sample-size calculations, when presented, are illustrations based on assumed design choices and an effect size. They are not observed outcomes and are not a substitute for the study's estimated effect and confidence interval. The stated planning framework describes a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect expressed as a fraction of the observed effect. Time-to-event illustrations are described as total events, while other illustrations are described as participants per arm. These assumptions do not establish the trial's original design rationale, which is not reported in the registry record. A planning calculation also should not be interpreted as observed power or as evidence that the primary comparison was conclusive.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists five outcome measures and one analysis. The named secondary outcomes include functional capacity assessed using the 6-minute walk test and the 2-minute step test, and self-reported quality of life measured around pulmonary rehabilitation and during follow-up. The primary endpoint is itself a composite of COPD-related hospital readmission or death. The record does not describe a multiplicity-adjustment procedure, a hierarchy of hypotheses, or how interpretation across outcomes was controlled. Those details are not reported in the registry record.
The registry describes the primary outcome as using all available complete data and says that missing data were not imputed. It does not report how many primary outcomes were missing, why data were unavailable, whether missingness differed between arms, or whether sensitivity analyses were performed. Without those details, the extent to which incomplete outcome data could affect the primary analysis cannot be assessed.
Three analysis populations are mentioned in the primary-outcome description: intention-to-treat, medically cleared participants, and participants who sat on the bike at least once. The record does not provide numerical results for these populations or explain the exact inclusion criteria and analysis methods used for each. It does not report whether all randomized participants were included in the intention-to-treat analysis, or how participants with incomplete follow-up were handled within each population. These details are not reported in the registry record.
The protocol description says the TelePR and SPR arms were combined for the primary outcome as prespecified, while the study arms are otherwise described separately. The exact statistical contrast and its relation to the stated regression are not fully specified in the registry record. No adjustment variables, subgroup testing plan, interim-analysis plan, or stopping rule are reported.
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 |
|---|---|---|---|
| Standard of Care | 3 | 98 | 3.1% |
| Telehealth-delivered Pulmonary Rehabilitation | 3 | 111 | 2.7% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Standard of Care | 2 of 98 |
| Telehealth-delivered Pulmonary Rehabilitation | 3 of 111 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 1% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Standard of Care | 3 of 98 |
| Telehealth-delivered Pulmonary Rehabilitation | 13 of 111 |
Most frequent serious adverse event terms
| Event term | Organ system | Standard of Care | Telehealth-delivered Pulmonary Rehabilitation |
|---|---|---|---|
| Serious Adverse Event | Cardiac disorders | 1 of 98 | 3 of 111 |
| Serious Adverse Event | Respiratory, thoracic and mediastinal disorders | 2 of 98 | 0 of 111 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Standard of Care | Telehealth-delivered Pulmonary Rehabilitation |
|---|---|---|---|
| Other (Not Including Serious) Adverse Events | Cardiac disorders | 2 of 98 | 9 of 111 |
| Other (Not Including Serious) Adverse Events | Eye disorders | 1 of 98 | 1 of 111 |
| Other (Not Including Serious) Adervse Events | Musculoskeletal and connective tissue disorders | 0 of 98 | 2 of 111 |
| Other (Not Including Serious) Adverse Events | Metabolism and nutrition disorders | 0 of 98 | 1 of 111 |
Serious events with the highest arm-level rate: serious adverse event (2.7%).
The registry reports safety figures separately for Standard of Care and Telehealth-delivered Pulmonary Rehabilitation. Serious adverse events affected 3 participants in Standard of Care, with 98 listed as at risk; the corresponding figures for the telehealth arm are 3 affected and 111 at risk. Other adverse events affected 3 participants in Standard of Care, with 98 at risk, and 13 in the telehealth arm, with 111 at risk. Deaths are listed as 2 and 3, respectively.
The registry also lists “Serious Adverse Event” as the top serious adverse event term, with 2.7%. The record does not explain the denominator used for this percentage or how it relates to the group-level counts. The listed counts are descriptive. They do not establish that an intervention caused an event, and they do not account for differences in follow-up or other characteristics.
The registry does not state whether the deaths listed in the safety information are the same events as deaths included in the primary composite. Nor does it provide a formal statistical comparison of adverse events between groups. These safety figures should therefore be read as reported counts, not as estimates of comparative risk or evidence of causation.
12. What the Result Does, and Does Not, Mean
The registry record describes a randomized comparison of telehealth-delivered pulmonary rehabilitation with standard pulmonary rehabilitation for Hispanic and African-American patients admitted with COPD exacerbation. It identifies COPD-related hospital readmission or death within 6 months after discharge as the primary composite outcome and names logistic regression as the analysis method. Those details define the study question and the intended statistical approach.
The record does not report the primary-outcome estimate, arm-specific primary-outcome counts, confidence interval, or p-value. It therefore cannot establish the direction or magnitude of a primary comparison, or how precisely such a comparison was estimated. The absence of a numerical result is not evidence that the approaches had the same outcome, and it is not evidence that one approach was better. It means the primary effect cannot be interpreted from the registry record.
The safety information gives counts of serious adverse events, other adverse events, and deaths by group. These counts describe what is recorded, but they are not a formal causal or statistical comparison. The record does not make clear whether safety deaths overlap with deaths in the primary composite. Nor does it provide enough detail to assess differences in exposure or follow-up.
As a general statistical principle, a study's design and analysis plan describe how investigators intend to answer a question; they do not themselves show the result. A complete interpretation would require the primary estimate and uncertainty interval, details of the analyzed populations and missing outcome data, and enough information to understand the exact group contrast. The registry record does not report those elements. Consequently, the findings should be described as unavailable in this record rather than inferred from the study design, the adverse-event counts, or the fact that the study is marked completed.
13. Why This Trial Matters Statistically
This study is statistically notable because its question concerns access to pulmonary rehabilitation after hospitalization, and its intervention changes how rehabilitation is delivered rather than replacing rehabilitation with no program. The registry describes remote exercise supervision and education as a way to reduce the need for travel. The randomized design is intended to support a comparison between the telehealth and standard-care approaches, while the endpoint focuses on readmission or death after discharge.
The primary endpoint is a composite, so interpretation would depend on both the combined outcome and its component events. A single composite estimate could summarize whether either qualifying event occurred, but it would not show whether readmission or death principally shaped the result. Component-specific information is not reported in the registry record. This matters because the components are different outcomes and may not carry the same meaning for participants or clinicians.
The record also makes clear why statistical transparency matters. It identifies logistic regression and a 95% confidence interval as part of the analysis description, but does not report the model estimate or interval. Without those results, readers cannot judge either the size or precision of the primary comparison. Naming a method is not a substitute for reporting its output.
The stated complete-data approach, with no imputation, is another important interpretive detail. Missing outcomes can affect both precision and the representativeness of the analyzed data. The number and pattern of missing outcomes are not reported in the registry record, preventing assessment of this issue.
These features make the study question relevant for biostatistical interpretation even though the primary numerical result is unavailable in the record. They illustrate the need to distinguish the clinical question, the planned analysis, observed estimates, and uncertainty. The study's completed status and its documented design do not resolve the missing primary-outcome result.
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 the primary outcome is that the registry record does not report a single primary-outcome comparison. No event counts by arm, odds ratio, confidence interval limits, or p-value are reported. Although the record names logistic regression and a 95% confidence interval, the numerical results needed to assess effect size and precision are absent.
The analysis description says the TelePR and SPR arms were combined for the primary outcome as prespecified in the protocol, but the exact group contrast is not fully clear from the registry record. This makes it difficult to determine how that description relates to the comparison between telehealth and standard pulmonary rehabilitation. The registry does not provide enough model detail to resolve the question.
The complete-data analysis with no imputation is described, but the number and pattern of missing primary outcomes are not reported. The record mentions intention-to-treat, medically cleared participants, and participants who sat on the bike at least once, but does not provide results or detailed definitions for those populations. Covariates, model diagnostics, sensitivity analyses, and multiplicity procedures are also not reported in the registry record.
The primary endpoint combines readmission and death, but component-specific results are not reported. Safety counts are available, yet their relationship to the primary composite is not explained, and no formal comparative safety analysis is described. The registry also does not report detailed intervention schedules, adherence summaries, or the roles of those masked in the double-masked design. These are limits on what can be learned from the record, not evidence that any particular aspect of the study was conducted inadequately.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Telehealth Pulmonary Rehabilitation for Hispanic and African-American Patients Admitted With Exacerbation of COPD — NCT03007485.
- Registry record last updated 2022-03-10; status last verified 2021-12.
- Results first posted 2022-03-10.
- 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
NCT03007485 is a completed, randomized, double-masked, parallel study sponsored by Northwell Health. The registry lists enrollment as 266, a start date of 2017-04-01, and primary completion on 2020-11-15. The study compares telehealth-delivered pulmonary rehabilitation with standard pulmonary rehabilitation in patients admitted with COPD exacerbation.
The primary endpoint is COPD-related hospital readmission or death within 6 months of discharge. The registry describes logistic regression, all available complete data, and no imputation of missing data. It also describes combining the TelePR and SPR arms for the primary outcome as prespecified. The primary numerical comparison and its confidence interval are not reported in the registry record.
Secondary outcomes address functional capacity and self-reported quality of life. Safety counts include serious adverse events, other adverse events, and deaths in each group. These counts are descriptive and are not a substitute for the missing primary-outcome estimate.