The Improving ATTENDance to Cardiac Rehabilitation Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Improving ATTENDance to Cardiac Rehabilitation Trial (NCT03646760), a randomized na study. The primary outcome is Number of CR Sessions Completed Within 6 Months in Patients Randomized to HYCR vs. Patients Randomized to Traditional CBCR (Usual Care)..

2. Clinical Question

The clinical question concerns whether changing how rehabilitation is delivered is associated with a different amount of attendance. The trial does not compare rehabilitation with no rehabilitation: both groups were assigned to a rehabilitation approach. The hybrid model combines in-person care with remote or home-based sessions, while the comparison group is center-based rehabilitation.

The registry lists several cardiovascular conditions, but it does not report the distribution of those diagnoses among participants or specify eligibility details in the outcome record. Nor does it describe the primary-outcome population beyond the randomized groups. Those omissions limit how specifically the result can be related to a particular diagnosis or participant subgroup.

The primary measure is attendance, not a direct measurement of clinical benefit. Exercise capacity and quality of life are listed as secondary outcomes, but the central comparison is the number of sessions completed within 6 months. The outcome can describe participation in rehabilitation; by itself, it does not establish what a difference in attendance means for health or whether one delivery model is preferable for an individual.

3. Trial Design

The registry identifies this as a completed, randomized, parallel-group trial with no masking. Participants were assigned to HYCR or CBCR. Enrollment was 282: the posted primary-outcome summaries include 142 participants in HYCR and 140 in CBCR. The registry lists the phase as NA. The sponsor is Henry Ford Health System.

The intervention is behavioral rather than a drug or device treatment. The registry summary characterizes HYCR as a combination of center-based and remote- or home-based cardiac rehabilitation, tailored to individual needs with telemedicine support. CBCR is described as center-based cardiac rehabilitation and is identified in the primary outcome title as traditional care or usual care. Detailed arm descriptions are not reported in the registry record.

The primary endpoint is the number of cardiac rehabilitation sessions completed within 6 months, with the timeframe recorded as baseline to 6 months. Results are presented as means with standard errors. The registry also lists three secondary outcomes, each assessed over baseline to 6 months: change in six-minute walk distance, peak oxygen uptake, and Dartmouth COOP total score. The posted numbers contributing to those secondary outcomes are smaller than the primary-outcome group counts, and the reasons for the differences are not reported in the registry record.

The registry records the study as completed, with a start date of 2019-03-01 and a primary completion date of 2023-12-30. It does not report the original sample-size justification, allocation-concealment procedures, the detailed randomization procedure, or a full statistical analysis plan. No masking is recorded; who, if anyone, was unaware of assignment for specific measurements is not reported in the registry record. Random assignment supports a comparative design, but these missing operational details constrain assessment of how assignment and follow-up were implemented.

The results are group summaries rather than a detailed participant-level account. The record does not report baseline characteristics, adherence to assigned delivery, crossovers, or the number of participants with each outcome missing. Those details would help readers understand the composition of the analyzed groups and the handling of incomplete outcome information, but they 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.

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 Number of CR Sessions Completed Within 6 Months in Patients Randomized to HYCR vs. Patients Randomized to Traditional CBCR (Usual Care)., measured over Baseline to 6 months.

The primary endpoint counts cardiac rehabilitation sessions completed within 6 months. It operationalizes attendance as the amount of rehabilitation participants received, rather than simply whether they started rehabilitation or were assigned to a particular delivery model. The registry presents the endpoint as a mean number of sessions for each randomized group, with standard errors as the dispersion measure.

Attendance is relevant to a trial comparing delivery formats because the opportunity to attend may depend on how sessions are offered. A hybrid option may allow some sessions to occur remotely or at home, while CBCR is center-based. The posted attendance outcome therefore addresses whether the assigned approaches were accompanied by different mean session counts over the defined follow-up period.

Session count is a process or participation measure. It does not directly measure exercise capacity, quality of life, cardiovascular events, or the clinical benefit of rehabilitation. Those are distinct questions; the registry lists exercise capacity and quality of life as secondary outcomes. The attendance endpoint can indicate how much participation was recorded, but it cannot by itself establish why attendance differed, whether session counts were clinically important, or whether any difference translated into other outcomes.

The endpoint’s definition is limited in the registry record. It does not specify how a completed session was defined, how remote and center-based sessions were counted relative to one another, or how incomplete follow-up was handled. It also does not report a threshold that would distinguish a meaningful from an unimportant change in session count. These details matter when interpreting the meaning and comparability of the group averages.

7. Results Posted in the Registry

• Center Based Cardiac Rehabilitation (CBCR): mean 27.60 versus 28.70 on control; difference -1.10 (95% CI -33.81 to 31.61), standardized effect d = -0.01; z-test p = 0.947.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Improvement in Exercise Capacity, as Measured by Distance Walked During the Six Min Walk (6MW) Test, in Patients Randomized to HYCR vs Patients Randomized to CBCR.

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Center Based Cardiac Rehabilitation (CBCR): mean 50.00 versus 46.00 on control; difference 4.00 (95% CI -9.34 to 17.34), standardized effect d = 0.08; z-test p = 0.557.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Center Based Cardiac Rehabilitation (CBCR) and the control. The interval gives the range of effects that remain compatible with the data.

Improvement in Exercise Capacity, as Measured by Peak Oxygen Uptake (VO2), in Patients Randomized to HYCR vs Patients Randomized to CBCR.

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Center Based Cardiac Rehabilitation (CBCR): mean 1.90 versus 2.30 on control; difference -0.40 (95% CI -1.20 to 0.40), standardized effect d = -0.14; z-test p = 0.330.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Center Based Cardiac Rehabilitation (CBCR) and the control. The interval gives the range of effects that remain compatible with the data.

Improvement in Quality of Life (QOL), as Measured by the Total Score for Dartmouth COOP, in Patients Randomized to HYCR vs Patients Randomized to CBCR.

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Center Based Cardiac Rehabilitation (CBCR): mean -3.40 versus -3.90 on control; difference 0.50 (95% CI -0.87 to 1.87), standardized effect d = 0.10; z-test p = 0.476.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Center Based Cardiac Rehabilitation (CBCR) and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover exercise capacity and quality of life, rather than attendance. The posted comparisons show estimates in different directions across these measures, and each has an interval that includes no difference. This broad pattern does not establish a consistent advantage for either delivery approach, but neither does it establish that the approaches have the same effects. Each measure has its own scale, interpretation, and uncertainty.

The secondary outcomes relate to the primary result by addressing other possible dimensions of rehabilitation experience and outcome. Session attendance is a measure of participation, while walking distance, peak oxygen uptake, and quality of life describe different constructs. Similar-looking results across distinct measures cannot be combined into a single conclusion without a prespecified method for doing so. Nor can the secondary outcomes explain why the primary attendance averages were close.

The numbers contributing to the secondary outcomes are lower than the primary group counts for some measures. The registry record does not report why, whether the same participants contributed to each measure, or how missing measurements were handled. This means the secondary results may be less precise and may reflect different analyzed subsets; the available record does not permit a fuller evaluation of that possibility.

The secondary comparisons are unadjusted, and the registry does not report a multiplicity procedure or a hierarchy specifying how those outcomes were to be interpreted. When multiple outcomes are examined without adjustment, a single apparently notable result can arise by chance more often than when considering one outcome alone. That general concern is a reason to give unadjusted secondary results less weight than a prespecified primary analysis, not a reason to dismiss their descriptive information.

The registry lists no sponsor-reported analyses for the secondary outcomes. Therefore, the posted summary comparisons should not be treated as confirmatory evidence of benefit or lack of benefit. They are best read as exploratory estimates whose uncertainty, population differences, and relationship to the primary attendance question all remain relevant. The results do not support a conclusion about the overall clinical effect of either rehabilitation approach.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Improvement in Exercise Capacity, as Measured by Distance Walked During the Six Min Walk (6MW) Test, in Patients Randomized to HYCR vs Patients Randomized to CBCR.
• Improvement in Exercise Capacity, as Measured by Peak Oxygen Uptake (VO2), in Patients Randomized to HYCR vs Patients Randomized to CBCR.
• Improvement in Quality of Life (QOL), as Measured by the Total Score for Dartmouth COOP, in Patients Randomized to HYCR vs Patients Randomized to CBCR.

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a count summarized as the mean number of sessions completed. A mean describes the arithmetic average in a group. The registry reports standard errors for the primary group means. A standard error describes uncertainty in an estimated mean; it is not the same thing as the spread of individual participants’ session counts. The registry does not give the underlying distribution of attendance, so the extent of skewness, clustering at particular session counts, or outlying values is not reported in the registry record.

What does the group comparison estimate?

The comparison is expressed as a mean difference between HYCR and CBCR. A mean difference summarizes the contrast between group averages on the original outcome scale, here the number of sessions. It does not describe how much participants vary individually, and it does not imply that every participant experienced a difference equal to the group estimate. The comparison interval communicates the uncertainty around the estimated contrast. Because the reported interval crosses zero, it includes no mean difference as well as values on either side.

The record contains an unadjusted p-value for the primary comparison. In general, an unadjusted p-value evaluates compatibility between the observed data and a no-difference hypothesis under the statistical model used; it does not quantify effect size, clinical importance, or the probability that the hypothesis is true. “Unadjusted” means the value has not been adjusted for multiple comparisons. The registry does not report the full model specification or enough detail to identify all assumptions used for the posted primary comparison.

Why does uncertainty matter?

A point estimate is only one summary of a comparison. A confidence interval gives a range of values that are reasonably compatible with the data and the method’s assumptions. A narrow interval would indicate greater precision; a broad interval indicates less precision. The primary interval is broad and includes values favoring either group, so the estimate does not settle the direction or size of a possible difference. A confidence interval is not a guarantee that the true value lies within its particular endpoints, and it does not provide a direct probability statement about the effect.

The primary group summaries use standard errors as the dispersion measure. The secondary outcomes instead report standard deviations. In general, a standard deviation describes variation among individual observations, while a standard error describes uncertainty in a group estimate. These quantities answer different questions and should not be interchanged. Their distinction matters when comparing reported summaries or trying to understand how variable participants’ outcomes were.

What does “unadjusted” imply for secondary results?

The registry lists secondary comparisons and unadjusted p-values. When several outcomes are tested, the chance of at least one apparently unusual result can increase if each result is considered separately without accounting for multiplicity. An adjustment or a prespecified hierarchy can address that issue, but the registry does not report such a procedure. The secondary findings should therefore be interpreted cautiously, particularly because the registry lists no sponsor-reported analyses for them. The outcome summaries remain descriptive evidence, but isolated p-values should not be treated as confirmatory findings.

Another general issue is over-dispersion: for count outcomes, the variability can be larger than a simple model would expect. If that occurs, a model that assumes less variability may underestimate uncertainty and produce intervals or tests that are too optimistic. The registry reports means and standard errors for session counts but does not report a distributional assessment or a model designed to address over-dispersion. It is therefore not possible to determine from the registry record whether this issue was evaluated. This is a general statistical consideration, not a claim that over-dispersion occurred in this trial.

What analysis details are absent?

The registry record does not report whether the primary comparison was adjusted for baseline characteristics, whether it followed an intention-to-treat principle, or how participants with incomplete outcome data were handled. It does not state whether assumptions were checked, whether any transformations or alternative models were used, or whether there were interim analyses. These matters affect the interpretation of the inferential results, and they should not be inferred from the reported group summaries alone.

The record also does not report a prespecified statistical analysis plan or the original rationale for the planned enrollment. Consequently, it is unclear whether the posted analysis was the planned primary analysis, how endpoint-specific populations were defined, or how the trial’s original design addressed the expected size and variability of the attendance measure. These unknowns do not invalidate the posted descriptive values, but they limit assessment of the analysis choices and the intended strength of inference.

The comparison values and planning illustrations are distinct. The former summarize the observed group data and their uncertainty; the latter illustrate the sample sizes that a simple planning calculation would imply under stated assumptions. Planning calculations do not establish what the trial was designed to detect, and they do not measure the probability that this completed trial detected an effect. The original design assumptions are not reported in the registry record.

11. Confidence Intervals and Precision

The primary comparison is a mean difference of -1.1 sessions, with a 95% confidence interval from -33.81 to 31.61. This interval includes zero, meaning that a no-difference value is compatible with the reported uncertainty. It also extends to values on either side, so the estimate is not precise enough to identify a clear direction or tightly constrain the possible size of the difference. The interval should be considered alongside the group means, 28.7 for HYCR and 27.6 for CBCR, rather than treating the point estimate as the complete result.

Confidence intervals express uncertainty in an estimate, conditional on the method and its assumptions. In general, a narrower interval conveys more precision, while a wider interval conveys less. A confidence interval is not a range containing a fixed proportion of individual participants’ outcomes, and it is not a probability statement that the true difference has a particular chance of lying inside the specific reported endpoints.

The unadjusted p-value for the primary comparison is 0.947. It does not show that the groups are equivalent, nor does it indicate the probability that a difference exists. The interval is more informative about the range of differences compatible with the data, though its breadth still leaves substantial uncertainty. The registry does not report a clinically meaningful threshold for session counts, so it is not possible to determine whether any values within the interval would be important in practice.

Secondary confidence intervals also cross zero, but they concern different outcome scales and should not be treated as interchangeable with the primary attendance interval. The registry record does not report an equivalence margin or a precision target for these outcomes. Overall, the posted results support a cautious reading: the primary estimate is close in magnitude to the posted group means, but the interval is too wide to rule out potentially important differences in either direction.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size illustrations are planning calculations based on the observed effect, not the trial’s original design. For these continuous outcomes, the figures are participants per arm, not a total number of events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The calculations consider detecting the stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation indicates would give an 80% chance of detecting the selected effect under its assumptions. The 90% power column gives the corresponding planning target for a 90% chance. Power in this context is a property of a hypothetical design under specified assumptions; the values are not a statement about the probability that the completed trial detected an effect.

The rows labeled as planning for 100%, 75%, and 50% of the observed effect show how the assumed effect size changes the required sample size. Smaller effects need more participants because they are harder to distinguish from random variation. Thus, the figures increase as the planning target moves to a smaller fraction of the observed effect. These are illustrations, not recommendations or evidence that such effects are clinically important.

The calculations use the observed effect as their starting point, so they should not be read as independent confirmation of that effect. In particular, an observed estimate can be imprecise, as the primary confidence interval indicates. The planning figures do not correct that uncertainty, account for missing outcomes or other design features, or establish what enrollment would be needed under a different model. They simply translate the stated assumptions into participants-per-arm illustrations. This is not observed power.

The sample-size figures should be read with the endpoint and its uncertainty in mind. They do not imply that the registry trial was intended to enroll those numbers, nor do they show that the original study was adequately or inadequately planned. The registry record does not report the original design calculation, including its target effect or anticipated variability.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.100; pooled standard deviation σ = 140.121.
2. n = 2 × (1.96 + 0.842)² × 140.121² / -1.100² = 254,716.7, rounded up to 254,717 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -1.1                      # observed difference in means
sigma <- 140.120785650911                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3646760)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 254,717 (80% power), 340,994 (90% power)
 75% of observed effect: n per arm = 452,830 (80% power), 606,211 (90% power)
 50% of observed effect: n per arm = 1,018,867 (80% power), 1,363,974 (90% power)
power.t.test: 254717.7 
Simulated power at n = 254,717 per arm: 0.800 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes or comparisons are examined. If each is tested separately without adjustment, the chance of at least one result appearing unusual by chance can increase. The registry lists a primary attendance outcome and three secondary outcomes, but it does not report a multiplicity adjustment, a testing hierarchy, or a procedure for controlling the overall type I error rate. The unadjusted secondary p-values should therefore be interpreted cautiously and should not be treated as a set of independent confirmatory tests.

The registry record also does not report the full statistical analysis plan, including how primary and secondary hypotheses were prioritized or whether additional analyses were planned. It does not report interim analyses. These details are not available in the record and cannot be inferred from the posted outcome summaries. Without a stated analysis framework, it is difficult to determine how the results relate to planned confirmatory questions or whether all analyses are represented.

Missing data are another important uncertainty. The record gives participant counts for each outcome, but it does not explain why those counts differ, how many assigned participants lacked each measurement, or how missing outcomes were handled. It does not state whether an imputation method, complete-case analysis, or another approach was used. The implications depend on why data are missing and how missingness relates to outcomes, neither of which is reported in the registry record.

The analysis population is also not specified. The record does not say whether participants were analyzed according to randomized assignment, whether only participants who attended rehabilitation were included, or whether any per-protocol or other populations were used. These approaches answer different questions and may be sensitive to different biases. The displayed group summaries alone do not resolve which population underlies each outcome.

Finally, details such as baseline adjustment, model assumptions, sensitivity analyses, and handling of protocol deviations are not reported in the registry record. Their absence limits appraisal of robustness and of how the inferential results were obtained. This does not change the values posted, but it means the estimates should be interpreted with awareness that key analysis and missing-data decisions cannot be evaluated from the registry record.

15. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: stroke (0.7%).

The registry lists serious adverse events affecting 1 participant in HYCR and 0 in CBCR. It lists other adverse events affecting 8 participants in HYCR and 3 in CBCR. Deaths are listed for 1 participant in HYCR and 3 in CBCR. These are recorded counts; they should be read as safety summaries rather than as comparative estimates of risk.

The record lists Stroke as the top serious adverse event term, at 0.7%. It does not provide a detailed account of event timing, severity beyond the serious classification, relatedness to the intervention, or whether individuals experienced more than one event. The registry record does not report a statistical comparison of these safety counts or enough detail to determine whether they reflect differences in exposure or follow-up.

Small event counts can be highly uncertain and are not a reliable basis for concluding that one approach is safer. These figures also do not show that an event was caused by either rehabilitation approach. The safety listings are descriptive registry data, and the absence of a listed event in a group does not establish that the event cannot occur.

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

The result does mean that the registry reports similar group averages for the number of cardiac rehabilitation sessions completed within 6 months: 28.7 for HYCR and 27.6 for CBCR. The posted mean difference is -1.1, but its 95% confidence interval ranges from -33.81 to 31.61. The estimate is therefore uncertain, and the interval does not rule out differences in either direction. The unadjusted p-value of 0.947 is consistent with the absence of a clear statistical distinction in this comparison, but it is not proof that the groups are equivalent.

The result does not establish that HYCR and CBCR have identical effects. Demonstrating equivalence or noninferiority requires a prespecified margin and an analysis designed for that question; neither is reported in the registry record. A non-significant comparison alone cannot substitute for such a design. The confidence interval is particularly important because it communicates a broad range of values compatible with the comparison.

The result also does not establish that either rehabilitation approach is effective or ineffective relative to receiving no rehabilitation. Both groups were assigned to rehabilitation. Attendance is the primary endpoint, and it does not directly establish changes in health, exercise capacity, or quality of life. The registry’s secondary outcomes address some of those domains, but their unadjusted comparisons and incomplete analysis details call for caution.

The findings do not determine which format is appropriate for a particular person or setting. The record does not provide enough detail about participant characteristics, preferences, delivery fidelity, access, or subgroup outcomes to support individualized conclusions. Nor do the posted values establish why attendance or other outcomes might differ between approaches.

A useful reading is that this completed randomized trial posted an imprecise comparison of attendance, with group means that are close and a confidence interval that remains wide. The available information can describe the observed summaries and their statistical uncertainty. It cannot settle clinical importance, prove equality, establish a causal mechanism for attendance, or resolve all questions about the secondary outcomes and safety.

17. Why This Trial Matters Statistically

This trial raises a statistical question that is easy to oversimplify: a close pair of group averages is not the same as a precise estimate of no difference. The primary means are 28.7 and 27.6 sessions, while the confidence interval for the comparison extends from -33.81 to 31.61. That wide interval is central to interpretation. It shows why readers should examine the range of plausible values rather than treating the point estimate or p-value as a complete summary.

The design also illustrates the distinction between a participation endpoint and a health outcome. Attendance is directly relevant to how rehabilitation is delivered, but it is not itself a measure of clinical benefit. The listed secondary measures address other domains, yet their estimates cannot simply be merged with attendance into a single conclusion. Clear separation of endpoint meaning helps prevent claims that go beyond what each measure records.

Another important issue is the difference between a failure to detect a difference and evidence of equivalence. The posted comparison does not establish that the delivery models have equal effects. An equivalence or noninferiority question requires a suitable design and a prespecified margin, neither of which is reported in the registry record. The example underscores why a non-significant p-value should not be interpreted as proof of sameness.

Finally, the registry’s limited reporting of analysis populations, missing data, multiplicity, and the original design assumptions constrains statistical appraisal. These omissions matter because they shape how much confidence readers can place in the inferential interpretation, not because they change the descriptive means. The trial therefore offers a useful lesson in reading randomized evidence: describe the observed data, assess precision, distinguish exploratory from confirmatory outcomes, and keep the conclusion within the scope of the design and record.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves several trial-specific details unresolved. The intervention is summarized as a hybrid of center-based and remote or home-based rehabilitation, but the arm descriptions are empty. Session delivery, the extent of telemedicine use, and how attendance across formats was counted are not described in the outcome record. This makes it difficult to know how consistently the assigned approaches were implemented or to reproduce the intervention from the registry description.

The primary outcome is recorded over baseline to 6 months, but the record does not explain the operational definition of a completed session or how sessions outside the expected format were handled. It also does not report a clinically meaningful difference for the count. Without a stated threshold, the practical importance of a possible attendance difference cannot be judged from the estimate alone.

The posted primary summaries report means and standard errors, but not the distribution of individual session counts. This limits assessment of the shape and variability of a count outcome. The record does not state whether model assumptions were checked or whether the analysis addressed possible over-dispersion. This is a limitation of what can be assessed from the registry record, not evidence that the analysis was inappropriate.

Outcome-specific participant counts differ, but reasons for those differences and missing-data methods are not reported. Baseline characteristics, participant flow, adherence, crossover, and analysis populations are also not reported in the registry record. These omissions limit evaluation of comparability after randomization and of potential bias from incomplete follow-up.

Lastly, the record does not report the original sample-size justification, statistical analysis plan, multiplicity approach, or interim analyses. The secondary outcomes are unadjusted and lack listed sponsor-reported analyses. These reporting gaps prevent a full assessment of whether all analyses followed a prespecified plan, and they limit the certainty of broader conclusions beyond the descriptive comparisons posted.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: The Improving ATTENDance to Cardiac Rehabilitation Trial — NCT03646760.
• Registry record last updated 2025-07-29; status last verified 2025-07.
• Results first posted 2025-07-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The Improving ATTENDance to Cardiac Rehabilitation Trial is recorded as a completed, randomized, parallel-group study with no masking and enrollment of 282. It compared hybrid cardiac rehabilitation with center-based cardiac rehabilitation. The primary outcome was the mean number of sessions completed within 6 months.

The posted primary means were 28.7 sessions for HYCR and 27.6 for CBCR. The comparison estimate was -1.1, with a 95% confidence interval from -33.81 to 31.61 and an unadjusted p-value of 0.947. The interval is broad and includes no difference, so the result is imprecise rather than evidence of equivalence.

Three secondary outcomes address exercise capacity and quality of life. Their posted unadjusted comparisons also have intervals that include no difference. The registry record does not report the original sample-size justification, full analysis plan, analysis populations, or missing-data methods. The results should therefore be interpreted as registry-posted summaries with important limits on assessment of their analysis and practical meaning.