eIMPACT Trial: Modernized Collaborative Care to Reduce the Excess CVD Risk of Older Depressed Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for eIMPACT Trial: Modernized Collaborative Care to Reduce the Excess CVD Risk of Older Depressed Patients (NCT02458690), a randomized phase 2 study. The primary outcome is Brachial Artery Flow-Mediated Dilation (FMD) at 12 Months.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group phase 2 trial sponsored by Indiana University. The registry records the trial as completed, with an enrollment of 216 participants. It describes single masking; the identity of the masked party is not reported in the registry record. Participants were primary care patients with depression and no history of cardiovascular disease.

One group received eIMPACT for 12 months. The intervention used a collaborative stepped care approach, including computerized and telephonic cognitive-behavioral therapy and select antidepressant medications. The comparator was usual care: patients and their primary care providers were informed about the depressive disorder diagnosis and follow-up was encouraged, without restrictions on the care received. The registry describes the primary care clinics as using a team care approach, but further details of usual care are not reported in the registry record.

The study objective was to test whether delivering the modernized depression intervention before cardiovascular disease onset reduced future cardiovascular risk. Artery function was assessed by ultrasound before and after the one-year treatment period. Brachial artery flow-mediated dilation at 12 months was designated the primary outcome. The posted primary analysis includes all 216 randomized participants through multiple imputation, although 195 had post-treatment brachial FMD data.

The registry record does not report the trial's original sample-size justification, its prespecified statistical analysis plan, or details of the masking procedure. Those omissions limit what can be concluded about the design decisions beyond the randomized, parallel-group structure and the stated primary question.

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

Analysis population for the primary outcome, as defined in the registry: 195 of 216 randomized participants had post-treatment brachial FMD data. Because multiple imputation was employed, all 216 randomized participants were included in this analysis.

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 Brachial Artery Flow-Mediated Dilation (FMD) at 12 Months, measured over 12 months.

The primary endpoint was brachial artery flow-mediated dilation (FMD) at 12 months, expressed as the percent increase in brachial diameter. FMD measures how much the artery's diameter increases after a temporary change in blood flow. The registry describes measuring the artery with high-resolution ultrasound after a period of rest, then calculating the maximum percentage increase at 60- or 90-seconds after cuff deflation.

The trial used artery function as an intermediate measure relevant to its question about future cardiovascular risk. The study objective was to assess whether depression care delivered before cardiovascular disease onset changed this measure. In general, a continuous physiological measure can provide a quantitative way to compare groups, but it is not itself a count of cardiovascular events or a direct measure of future disease. A result for FMD therefore addresses the registry's specified vascular-function endpoint, not every possible consequence of the intervention.

The reported arm values are means with standard deviations. At 12 months, the mean was 2.28 in the eIMPACT arm and 2.44 in usual care. The recalculated mean difference, defined as eIMPACT minus usual care, was -0.16 percentage points. The registry record does not report a numerical estimate of the change from baseline in the primary outcome.

7. Results Posted in the Registry

• eIMPACT: mean 2.28 versus 2.44 on control; difference -0.16 (95% CI -0.74 to 0.42), standardized effect d = -0.07; z-test p = 0.588.

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.

Depressive Symptoms at 12 Months

• 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:
  • eIMPACT: mean 1.12 versus 1.65 on control; difference -0.53 (95% CI -0.75 to -0.31), standardized effect d = -0.65; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for eIMPACT than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

High-Frequency Heart Rate Variability (HF HRV) at 12 Months

• 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:
  • eIMPACT: mean 5.39 versus 5.35 on control; difference 0.04 (95% CI -0.44 to 0.52), standardized effect d = 0.02; z-test p = 0.871.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between eIMPACT and the control. The interval gives the range of effects that remain compatible with the data.

Interleukin-6 (IL-6) at 12 Months

• 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:
  • eIMPACT: mean 0.73 versus 0.77 on control; difference -0.04 (95% CI -0.11 to 0.03), standardized effect d = -0.15; z-test p = 0.267.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between eIMPACT and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes span depressive symptoms and several physiological measures, including heart-rate variability, interleukin-6, high-sensitivity C-reactive protein, β-thromboglobulin, and platelet factor 4. The numerical comparisons reported for depressive symptoms, heart-rate variability, and interleukin-6 show a mixed pattern rather than a uniform shift across all the measured domains. Numerical result details for the other registry-listed secondary measures are not reported in the registry record.

This pattern is not a substitute for the primary FMD result. Secondary measures can provide context about symptoms and biological processes, but they address distinct outcomes and may not move together. The primary comparison concerns brachial artery FMD at 12 months; the secondary outcomes do not establish a primary-outcome effect when that primary comparison is uncertain.

Secondary findings generally carry less weight when many outcomes are assessed, especially if the analysis plan and multiplicity procedures are not clear. Each additional test creates another chance of an apparently noteworthy result arising through random variation. The registry record does not report whether secondary analyses were adjusted for multiple comparisons or identify a hierarchy among them. The numerical secondary results should therefore be read with caution and in context, rather than as a single confirmatory package of evidence.

The registry reports ANCOVA p-values for the three secondary results with numerical details, while the simple recalculations use unadjusted arm summaries. The record does not give the adjusted estimates and confidence intervals for those analyses. The differences between outcome scales and analysis methods also mean that the secondary estimates should not be combined informally into one overall treatment effect.

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

• Depressive Symptoms at 12 Months
• High-Frequency Heart Rate Variability (HF HRV) at 12 Months
• Interleukin-6 (IL-6) at 12 Months
• High-Sensitivity C-Reactive Protein (hsCRP) at 12 Months
• β-thromboglobulin at 12 Months
• Platelet Factor 4 (PF4) at 12 Months

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a measurement on a numeric scale rather than a category or event count. The registry reports each arm's mean and standard deviation. A mean describes the center of the observed values; a standard deviation describes how spread out individual values are around that center. These summaries help distinguish the average outcome from the variability among participants.

What the mean difference describes

The recalculated mean difference is defined as the eIMPACT mean minus the usual-care mean. A negative value indicates a lower mean in the eIMPACT arm for this outcome, while a positive value indicates a higher mean. The estimate of -0.16 is an unadjusted comparison calculated from the posted group summaries. It is not a baseline-adjusted treatment estimate and should not be conflated with the trial's ANCOVA analysis.

In general, an unadjusted comparison summarizes the separation between group means without accounting for other variables. ANCOVA, or analysis of covariance, compares groups while adjusting for one or more covariates. This can account for measured differences in variables included in the model and may improve precision. Which covariates were used, how they were coded, and whether the model included baseline FMD are not reported in the registry record. The sponsor's reported primary method is ANCOVA, but its adjusted effect estimate and confidence interval are not reported.

What the confidence interval contributes

A confidence interval describes the uncertainty around an estimated effect under the assumptions of the analysis. It gives a range of effects compatible with the data and method at the stated confidence level. The recalculated interval runs from -0.738 to 0.418. Because it includes zero, it is compatible with no mean difference; because it also includes values on both sides of zero, it is compatible with either direction of difference. The interval's width is important: the point estimate alone would conceal the uncertainty about the size and direction of the underlying effect.

The recalculated p-value of 0.588 tests the unadjusted comparison against a null hypothesis of no mean difference, using the assumptions of that calculation. A p-value measures how unusual data at least as extreme as those observed would be if the null hypothesis and model assumptions held. It is not the probability that the null is true, nor a measure of the size or importance of an effect. The registry's ANCOVA p-value is 0.60, which pertains to the sponsor's reported adjusted analysis.

Why variation matters

Standard deviations of 2.04 in eIMPACT and 2.29 in usual care describe meaningful variation in individual FMD values around the respective means. When outcomes vary substantially, a given difference between group averages is harder to estimate precisely unless the study has enough information. General statistical reasoning therefore treats both the estimated group separation and its variability as relevant to precision. No extra calculation is needed to recognize that a mean difference should be read together with its confidence interval.

The standardized effect size reported for the primary comparison is -0.074. Standardized effects express the mean difference relative to a measure of outcome variability, making them less dependent on the original scale. They can help compare magnitudes across settings, but they do not replace the original-unit estimate, and their interpretation depends on the chosen standardization. For clinical interpretation, the original FMD scale and an appropriate meaningful-difference threshold would be needed; no such threshold is reported in the registry record.

Missing outcome data and imputation

Post-treatment FMD data were available for 195 of 216 randomized participants. The registry says multiple imputation was employed and that all 216 randomized participants were included in the primary analysis. In general, multiple imputation uses a statistical model to account for missing values by creating and analyzing multiple plausible completed datasets, then combining the results to reflect imputation uncertainty. The imputation model, variables included, assumptions about the missingness process, number of imputations, and sensitivity analyses are not reported in the registry record. Those details matter because imputation does not make missing observations equivalent to directly observed measurements; conclusions depend in part on the model and assumptions.

The registry describes the primary analysis population as all randomized participants through imputation. Further definitions of analysis populations and any departures from assignment-based analysis are not reported in the registry record. The reported data do not establish whether every secondary endpoint used the same missing-data approach.

How secondary analyses fit in

Secondary outcomes include depressive symptoms, heart-rate variability, inflammatory markers, and platelet-related measures. These outcomes can help describe possible pathways or additional effects, but they are not substitutes for the primary endpoint. Each additional comparison creates another opportunity for a result to appear unusual by chance. A multiplicity adjustment can control or otherwise address this issue; whether any adjustment was planned or used is not reported in the registry record. Secondary results should therefore be interpreted as supporting or exploratory evidence unless their analysis plan and error-control approach are clear.

11. Confidence Intervals and Precision

The recalculated 95% confidence interval for the primary mean difference is -0.738 to 0.418 percentage points, with the difference defined as eIMPACT minus usual care. It covers values on both sides of zero. This means the posted summaries do not pin down whether the underlying group difference is negative or positive, and they do not rule out a difference of some size in either direction. The point estimate of -0.16 is only one value within that broader range.

Precision concerns how narrowly an estimate is localized, not whether the estimate is important or whether an intervention is beneficial. A narrow interval would indicate greater precision; a wide interval indicates more uncertainty about the size of the effect. Here, the interval spans from -0.738 to 0.418. The registry record does not state a clinically important threshold for FMD, so the interval cannot be classified as excluding or including a known clinically important effect.

The interval is a recalculated, unadjusted interval from posted summary statistics. It is not the confidence interval from the registry's ANCOVA analysis. The registry reports the ANCOVA p-value as 0.60 but does not report its adjusted effect estimate or interval. Consequently, the recalculated interval is useful for understanding the raw summary comparison, but the sponsor's model-based result remains the trial's reported analysis.

Confidence intervals rely on assumptions about the data and method used. Their interpretation does not account automatically for unreported analysis choices, missing-data assumptions, or multiplicity. The primary outcome used multiple imputation for all 216 randomized participants despite post-treatment data being available for 195. Details needed to assess the contribution of imputation to precision are not reported in the registry record.

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 planning figures are participants per arm for a continuous-outcome comparison, not total event counts. They are illustrative calculations based on the observed effect, rather than the trial's original design justification. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The planning effect is a stated fraction of the observed effect.

The 80% power column gives the approximate number of participants per arm that would yield an 80% chance of detecting the selected effect under those planning assumptions, if that effect were the true one. The 90% power column gives the corresponding per-arm count for a 90% chance. Power describes a design's long-run probability of detecting a specified effect using a particular test and assumptions; it is not the probability that the study's conclusion is correct.

The rows planning for 100%, 75%, and 50% of the observed effect illustrate how the assumed effect size changes the required sample. When the effect to detect is smaller, more participants are needed to distinguish it from random variation at the same significance and power targets. The figures for the primary outcome are 2,887 and 3,865 participants per arm for planning for 100% of the observed effect; 5,133 and 6,871 for planning for 75%; and 11,548 and 15,459 for planning for 50%, respectively for 80% and 90% power.

These are planning illustrations, not observed power. They do not establish what the trial was designed to detect, and they should not be interpreted as the trial's actual sample-size target. The registry record does not report the original sample-size assumptions or justification.

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 = -0.160; pooled standard deviation σ = 2.170.
2. n = 2 × (1.96 + 0.842)² × 2.170² / -0.160² = 2,886.9, rounded up to 2,887 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 <- -0.16                      # observed difference in means
sigma <- 2.16977151129213                     # 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(2458690)
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 = 2,887 (80% power), 3,865 (90% power)
 75% of observed effect: n per arm = 5,133 (80% power), 6,871 (90% power)
 50% of observed effect: n per arm = 11,548 (80% power), 15,459 (90% power)
power.t.test: 2887.8 
Simulated power at n = 2,887 per arm: 0.792 (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. Analyses Reported by the Sponsor

The registry reports an ANCOVA primary analysis with p = 0.60. This is the sponsor's reported inferential result for the comparison of eIMPACT with usual care. The adjusted effect estimate, confidence interval, and model covariates are not reported in the registry record, so the magnitude and precision of that adjusted comparison cannot be reconstructed from the reported p-value alone.

The recalculation from the posted arm summaries gives an unadjusted mean difference of -0.16, a 95% confidence interval from -0.738 to 0.418, and p = 0.588. It uses the group means, standard deviations, and sample sizes without covariate adjustment. The registry's ANCOVA and the recalculation therefore describe related but not identical analyses. Their p-values are close, but the registry record does not explain the difference between them.

The recalculation is useful as a transparent check on the raw group summaries and as a way to express uncertainty in the original measurement units. It does not supersede the reported ANCOVA. Since the registry does not provide the adjusted estimate or interval, the recalculated interval should not be presented as if it came from the sponsor's analysis. The record does not indicate a discrepancy between the posted summaries and the reported analysis.

15. Multiplicity, Missing Data and Analysis Populations

The trial has one designated primary endpoint and several secondary outcomes. Testing multiple outcomes raises a multiplicity issue: if every comparison is treated as an independent opportunity to find evidence, the chance of at least one apparently unusual result can increase. Statistical plans may address this through a prespecified hierarchy, adjusted significance thresholds, or other procedures. Whether the trial used any such approach is not reported in the registry record.

The registry lists outcomes beyond the primary FMD measure, including depressive symptoms, high-frequency heart-rate variability, inflammatory markers, and platelet-related measures. It does not report a complete statistical analysis plan, the order in which secondary hypotheses were to be tested, or whether any secondary results were intended to be confirmatory. This makes it difficult to distinguish planned tests from exploratory analyses based on the registry record alone.

Missingness is documented for the primary FMD endpoint: 195 of 216 randomized participants had post-treatment data, and multiple imputation was used so all 216 randomized participants were included in the analysis. The registry record does not report the imputation model, the assumed missingness mechanism, the number of imputed datasets, or sensitivity analyses. Without those details, the robustness of the primary estimate to alternative missing-data assumptions cannot be assessed from the registry record.

Missing-data information for the secondary outcomes and definitions of analysis populations beyond the primary outcome are not reported in the registry record. Nor are detailed covariate specifications for ANCOVA, the handling of protocol deviations, or other analysis decisions. These omissions do not negate the posted results, but they constrain independent assessment of how analysis choices may affect estimates, uncertainty, and the interpretation of multiple outcomes.

16. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization for depression/suicidal ideation (1.9%), emergency department visit for depression/suicidal ideation (1.9%), suspected drug overdose (0.9%), diabetic ketoacidosis (0.9%), study procedures, including depression treatment (0.0%).

The registry reports serious adverse events affecting 5 participants in eIMPACT and 4 in usual care. Other adverse events affected 3 participants in eIMPACT and 1 in usual care. Deaths were reported for 5 participants in eIMPACT and 8 in usual care. These are counts as recorded; the registry record does not provide enough information to infer that the categories are mutually exclusive or to determine the relationship of individual events to treatment.

The listed serious-event terms include hospitalization for depression/suicidal ideation at 1.9%, emergency department visit for depression/suicidal ideation at 1.9%, suspected drug overdose at 0.9%, diabetic ketoacidosis at 0.9%, and study procedures, including depression treatment, at 0.0%. The registry record does not state the denominator or arm attribution for these term percentages in the information reported.

Event counts and listed terms should be read descriptively. They do not by themselves establish causation, comparative risk, or whether an event was judged related to the intervention. Detailed event-level descriptions, attribution, and follow-up context are not reported in the registry record.

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

The posted primary results describe a comparison of mean brachial FMD at 12 months. The unadjusted mean difference was -0.16, and its recalculated confidence interval ranged from -0.738 to 0.418. The registry's ANCOVA analysis reported p = 0.60. Together, these results do not provide a precise estimate of a difference in the primary endpoint, but they also do not prove that the groups had identical outcomes.

The result is specific to the endpoint, follow-up time, population, and analysis reported. It does not establish whether the program changed cardiovascular event rates, because the primary result is a measure of artery function rather than a reported comparison of clinical cardiovascular events. It also does not settle the effect on depression symptoms or other biological measures; those require separate interpretation and are subject to the limitations of secondary analyses.

A non-significant p-value should not be read as proof of no effect. The confidence interval gives a more informative account of the range of effects compatible with the unadjusted data, although it is not the interval from the sponsor's ANCOVA. Conversely, a difference in a secondary measure would not automatically establish a change in the primary outcome or a clinically important change. The registry record does not report a threshold for a meaningful FMD difference.

The randomized design supports a comparative question between assigned care approaches, but interpretation still depends on the quality of outcome measurement, missing-data handling, and the analysis plan. The registry describes multiple imputation for primary FMD and reports the ANCOVA p-value, but key model and imputation details are not reported. The results therefore should be understood as the reported evidence for this trial's specified measurements, not as a complete account of every potential benefit, harm, or longer-term outcome.

No treatment recommendation follows from these data alone. The record supports a cautious description of the measured outcomes and their uncertainty, while leaving unresolved questions about clinical importance, durability, and outcomes not numerically reported in the registry record.

18. Why This Trial Matters Statistically

This trial is statistically interesting because it tests a care-delivery intervention against an objective physiological outcome. Depression care and cardiovascular risk are linked in the trial's stated rationale, and brachial FMD was selected to assess artery function before and after the treatment period. This design makes the primary result a test of a proposed intermediate measure, rather than a direct comparison of cardiovascular events.

The randomized parallel-group design gives a structured comparison between collaborative stepped care and usual care. At the same time, the primary estimate illustrates why statistical interpretation should not stop at a p-value. The unadjusted difference is small relative to its uncertainty, and the interval covers both directions. The registry's ANCOVA p-value likewise does not provide the adjusted magnitude or precision because those estimates are not reported.

The trial also highlights the distinction between a trial's actual design and a retrospective planning illustration based on an observed effect. Planning calculations can show how assumptions about the effect to detect influence participant requirements, but they cannot recover the original rationale when that rationale is not reported. They also cannot be converted into observed power.

Finally, the multiple outcomes make transparency about analysis plans especially useful. Primary and secondary results answer different questions, and repeated testing can make isolated secondary findings less persuasive when multiplicity procedures are unknown. The registry provides a framework for understanding the intervention, primary measure, and some secondary outcomes, while leaving important analytic details unreported. That combination makes careful attention to effect estimates, uncertainty, and the limits of the record essential.

19. 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 primary analysis depends on multiple imputation because post-treatment FMD data were available for 195 of 216 randomized participants. The registry record does not report the imputation model, assumptions, or sensitivity analyses. Those omissions prevent assessment of whether results would be similar under alternative assumptions about missing outcomes.

The adjusted primary analysis is reported as ANCOVA with p = 0.60, but its adjusted estimate, confidence interval, and covariates are not reported. The recalculated unadjusted comparison cannot fill that gap. The registry also does not report a clinical threshold for a meaningful difference in FMD, limiting interpretation of the endpoint's practical importance.

The record does not report the original sample-size justification, detailed statistical analysis plan, multiplicity procedures, or the identity of the masked party. These details would help evaluate the trial's planned sensitivity, the status of secondary findings, and the potential influence of analysis choices. Numerical details for all registry-listed secondary outcomes are not reported in the record, which limits assessment of the full set of outcome results.

Finally, FMD is a physiological endpoint rather than a reported cardiovascular event outcome. The primary comparison therefore addresses artery function at 12 months, not whether participants later experienced cardiovascular disease. The record's reported results do not establish longer-term effects or the relationship between FMD differences and later clinical outcomes.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: eIMPACT Trial: Modernized Collaborative Care to Reduce the Excess CVD Risk of Older Depressed Patients — NCT02458690.
• Registry record last updated 2023-10-19; status last verified 2023-10.
• Results first posted 2020-10-14.
• 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.

23. Record Summary

NCT02458690 is a completed, randomized, single-masked, parallel-group phase 2 trial sponsored by Indiana University, with enrollment of 216. It compared 12 months of eIMPACT, a collaborative stepped care depression program, with usual care among primary care patients with depression and no history of cardiovascular disease.

The primary endpoint was brachial artery flow-mediated dilation at 12 months. The registry reports a primary ANCOVA p-value of 0.60. The unadjusted recalculation from posted arm summaries gives a mean difference of -0.16, with a 95% confidence interval from -0.738 to 0.418 and p = 0.588. The adjusted effect estimate and interval are not reported in the registry record.

Post-treatment FMD data were available for 195 participants; multiple imputation was used to include all 216 randomized participants in the primary analysis. Several secondary measures are listed, but the registry record does not report all analysis details needed to assess multiplicity, missing data, or the full set of secondary results.