Improving Medication Adherence Through Graphically Enhanced Interventions in Acute Coronary Syndromes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Medication Adherence Through Graphically Enhanced Interventions in Acute Coronary Syndromes (NCT00399880), a randomized na study. The primary outcome is Adherence to Refills and Medications Scale (ARMS).

2. Clinical Question

3. Trial Design

The registry identifies NCT00399880 as a completed, randomized, single-masked, parallel-group trial sponsored by Vanderbilt University Medical Center. The condition listed is acute coronary syndrome. The study began in 2006-11, and primary completion was recorded as 2008-08. Enrollment was 130. The phase is listed as NA, which is not a phase designation that can be interpreted as a particular development stage from this record alone.

Participants were assigned to a health-literacy intervention or usual care. The intervention description lists illustrated medication schedules, pill boxes, and pharmacist counseling; the usual-care description is blank. The registry marks masking as single, but does not identify which people involved in the trial were masked. The allocation procedure beyond the randomized designation, including how the sequence was generated or concealed, is not reported in the registry record.

The primary outcome was collected approximately 2 weeks after hospital discharge. Its reported analysis population consists of participants who could be contacted and provided outcome data. There were 38 participants with primary outcome data in each group. The secondary self-efficacy outcome has 39 intervention and 42 usual-care observations. These outcome counts do not account in the record for the difference from total enrollment, and the reasons for missing outcome data are not reported in the registry record.

The record includes a primary endpoint and one secondary endpoint, but no reported analyses. The trial’s original sample-size calculation, planned analysis model, covariate adjustments, handling of missing data, interim analyses, and multiplicity strategy are not reported in the registry record. Accordingly, the posted comparisons should be read as the recorded group summaries and unadjusted comparisons, rather than as evidence that a particular unreported analysis plan was followed.

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: Participants who were able to be contacted and provided outcome data.

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 Adherence to Refills and Medications Scale (ARMS), measured over Approximately 2 weeks after hospital discharge.

The primary endpoint is the Adherence to Refills and Medications Scale (ARMS), a validated self-report measure of medication adherence. It is expressed in units on a scale, with a possible range of 12-48. Lower scores indicate better adherence. The measure was recorded approximately 2 weeks after hospital discharge, a period the registry describes as vulnerable for patients returning home and managing medicines.

A self-report scale can capture participants’ accounts of how they take and refill medications, making it relevant to a study intended to support medication-taking behavior. It is not the same as a direct record of medication ingestion, and the registry does not report an objective adherence measure as the primary outcome. The result therefore describes reported scale scores among participants who were contacted and provided outcome data.

For interpretation, the direction of the scale matters: a lower group mean corresponds to better adherence according to the registry description. The comparison is a difference between group means, not the percentage of participants classified as adherent, and it does not convey how individual participants’ scores changed. The registry record does not report a threshold for a clinically meaningful difference on the ARMS, so the practical importance of a numerical difference cannot be determined from the scale range alone.

7. Results Posted in the Registry

• Health Literacy Intervention: mean 12.03 versus 12.18 on control; difference -0.15 (95% CI -1.61 to 1.31), standardized effect d = -0.05; z-test p = 0.841.

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.

Self-Efficacy for Appropriate Medication Use Scale (SEAMS)

• 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:
  • Health Literacy Intervention: mean 32.80 versus 31.50 on control; difference 1.30 (95% CI -1.22 to 3.82), standardized effect d = 0.22; z-test p = 0.313.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Health Literacy Intervention and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary outcome, the Self-Efficacy for Appropriate Medication Use Scale (SEAMS), measured approximately 2 weeks after discharge. Its comparison has a confidence interval that crosses zero and an unadjusted p-value of 0.313. The secondary result therefore does not provide a clear, precise contrast that resolves the uncertainty in the primary ARMS comparison.

The primary and secondary measures address related but distinct aspects of medication management: the primary endpoint is a self-reported adherence measure, while the secondary endpoint concerns self-efficacy for appropriate medication use. A relationship between the constructs may be plausible in general, but they are not interchangeable outcomes. The record does not report a planned model linking the two measures, or evidence that a difference in one should be expected to produce a difference in the other.

Secondary outcomes commonly require cautious interpretation because many outcomes or comparisons can create opportunities for apparently notable results by chance. A multiplicity strategy can define how to account for that issue, but one is not reported in the registry record. The posted secondary comparison is unadjusted, and the record does not state whether it was designated as confirmatory, exploratory, or part of a broader testing sequence. Its p-value should not be treated as if it were adjusted for other possible analyses.

In this record, the secondary result adds context but does not overturn or confirm the primary finding. It is best considered supportive descriptive information, not independent proof of a broader effect across medication-related outcomes. The absence of reported secondary analyses beyond this outcome also means the registry does not establish whether other measures were collected but not posted. That uncertainty reinforces the need to avoid drawing a general conclusion from a single secondary measure.

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

• Self-Efficacy for Appropriate Medication Use Scale (SEAMS)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: each participant contributes an ARMS score, and the registry reports a mean and standard deviation for each group. A mean describes the group’s average score. A standard deviation describes how spread out individual scores are around that average. These answer different questions: the mean locates the group’s center, while the standard deviation describes variation among participants. Neither statistic alone describes how certain the estimated difference between group means is.

The primary comparison is a mean difference, calculated as the intervention-group mean minus the usual-care mean. With lower ARMS scores indicating better adherence, a negative difference points toward a lower average score in the intervention group. This direction is about the scale’s score, not a conclusion about clinical benefit. The recorded difference is -0.15. The analysis is unadjusted: it does not account for other measured characteristics in the reported comparison. Whether the trial used another analysis that is not posted is not reported in the registry record.

What does a confidence interval describe?

A confidence interval expresses uncertainty about an estimated quantity, using a method with a stated coverage level when that level is specified. In repeated use of the same method under its assumptions, intervals would capture the underlying quantity at the method’s stated rate. It is not a range of individual participant scores, and it is not a probability statement about a particular hypothesis. The registry supplies an interval for the mean difference, but the confidence level is not reported in the registry record. Its endpoints, -1.615 and 1.315, lie on opposite sides of zero, so the interval does not pin down the direction of the mean difference.

How should the p-value be read?

A p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, a measure of clinical importance, or a probability that the result will be replicated. The posted primary p-value is unadjusted. Its interpretation depends on the analysis and assumptions used, and the registry record does not report the full statistical plan. A large p-value does not demonstrate that the groups are equivalent; demonstrating equivalence would require a prespecified equivalence margin and an analysis designed for that question, neither of which is reported in the registry record.

Why do group sizes and variability matter?

Estimated means become more precise when they are based on more observations and when participant scores are less variable, all else being equal. Here, the primary endpoint has 38 observations in each group, and the reported standard deviations are 2.59 and 3.81. The groups’ observed score distributions may overlap substantially even when their means differ; summary statistics alone do not show individual values or the shape of those distributions. General statistical reasoning also explains why small outcome samples can leave a wide range of plausible mean differences, even when the point estimate is close to zero.

The registry also gives a standardized effect-size summary for the primary comparison, but the underlying metric and its assumptions do not make it a substitute for the scale-unit difference. A standardized measure expresses a mean difference relative to variability, which can help compare effects measured on different scales. It does not supply a clinical threshold for ARMS, and it should not be interpreted as a percentage change or as the fraction of participants who improved. The registry record does not report an interpretation threshold for this endpoint.

What does an unadjusted comparison leave open?

An unadjusted comparison contrasts the observed group summaries without modeling other characteristics. Randomization is intended to support fair comparisons, but it does not guarantee identical characteristics in every realized trial. An adjusted model might account for prespecified baseline variables or improve precision, depending on its design and assumptions. Whether such a model was planned or conducted is not reported in the registry record. No inference should be made that adjustment was or was not part of an unpublished analysis.

The endpoint data also come from a subset of total enrollment. If the chance of having an observed score is related to participants’ adherence or to their group, comparisons among those with outcome data can differ from comparisons for all randomized participants. This is a general concern about missing outcomes, not evidence that such a pattern occurred in this trial. The reasons for missingness and the method used to handle it are not reported in the registry record. Thus the reported summaries describe observed outcome data, while the extent to which they represent all randomized participants cannot be established from the posted record.

Finally, general statistical methods depend on assumptions about the outcome, observations, and analysis. The registry does not report diagnostic checks, a distributional model, or the original statistical analysis plan. These omissions do not invalidate the posted summaries, but they limit how specifically the comparison’s assumptions and analytic choices can be assessed. The estimates and intervals are best read as concise summaries of the available endpoint data, not as a complete account of every analysis that may have been conducted.

11. Confidence Intervals and Precision

The primary mean difference is -0.15, and its reported confidence interval runs from -1.615 to 1.315. The interval crosses zero, so the data do not establish whether the underlying group difference is negative or positive. In scale terms, negative values correspond to a lower mean ARMS score in the intervention group, while positive values correspond to a higher intervention-group mean. Since lower ARMS scores indicate better adherence, the direction of the estimate has a defined scale interpretation, but the interval does not identify a single direction.

A confidence interval is useful because it displays a range of estimates compatible with the data and method, rather than focusing only on the point estimate. The primary interval includes differences favoring either group and values close to no difference. Its width indicates that the estimate is not highly precise in distinguishing a small difference from other possibilities. The registry record does not report the confidence level, so the interval should be described as the reported confidence interval without assigning it a level not stated in the record.

The unadjusted p-value is 0.841. It does not indicate the probability that the groups are the same, and it does not prove equivalence. The lack of a reported clinically meaningful difference threshold also means that statistical uncertainty cannot be translated into a judgment about practical importance. To make that judgment, a prespecified threshold on the ARMS scale and a clear account of the analysis population would be needed; neither is reported in the registry record.

Precision is relevant in light of the primary outcome counts: 38 participants contributed data in each group. The record gives enrollment of 130 but does not report why some participants did not provide primary endpoint data. Without that information, the interval’s statistical uncertainty is only part of the uncertainty in interpreting the study. The reported interval quantifies uncertainty under its analysis method; it cannot resolve possible limitations related to missing outcomes, measurement, or the applicability of the observed participants to all enrolled participants.

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, not total events. They are illustrations based on the observed effect, not the trial’s original design calculation. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

For the primary ARMS comparison, planning for 100% of the observed effect gives 7,404 participants per arm for 80% power and 9,912 participants per arm for 90% power. Planning for 75% of the observed effect gives 13,163 participants per arm for 80% power and 17,621 participants per arm for 90% power. Planning for 50% of the observed effect gives 29,616 participants per arm for 80% power and 39,647 participants per arm for 90% power.

Power is the probability that a study would meet a chosen statistical decision rule if a particular effect were present and the planning assumptions held. The 80% and 90% columns therefore describe different target probabilities under the illustration’s assumptions: the 90% target requires more participants than the 80% target for the same effect. A smaller effect is harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants. The curve expresses this general relationship across the planning range.

These calculations are not observed power. They are prospective planning illustrations using the observed effect as an input, not a measure of how much power the completed trial had after its results were known. They should not be read as the sample size actually used or as a recommendation for a future study without separate design assumptions and clinical 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 = -0.150; pooled standard deviation σ = 3.258.
2. n = 2 × (1.96 + 0.842)² × 3.258² / -0.150² = 7,403.8, rounded up to 7,404 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 <- -0.15                      # observed difference in means
sigma <- 3.25762183195042                     # 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(399880)
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 = 7,404 (80% power), 9,912 (90% power)
 75% of observed effect: n per arm = 13,163 (80% power), 17,621 (90% power)
 50% of observed effect: n per arm = 29,616 (80% power), 39,647 (90% power)
power.t.test: 7404.8 
Simulated power at n = 7,404 per arm: 0.806 (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, comparisons, time points, or analytic choices are tested. Without a prespecified approach, the chance of at least one apparently unusual result can increase as the number of tests grows. The registry identifies one primary endpoint and one secondary endpoint, but an adjustment method, hierarchy of hypotheses, or other multiplicity strategy is not reported in the registry record. The posted p-values are unadjusted. They should not be read as controlling the overall chance of a false-positive conclusion across all outcomes or analyses that may have been considered.

The primary endpoint population is described as participants who could be contacted and provided outcome data. The primary outcome count is 38 in each group. The secondary outcome has 39 intervention and 42 usual-care observations. These outcome-specific counts differ, but the registry does not explain why observations are missing, whether participants were excluded for particular reasons, or whether missingness differed by assignment. The method used to handle missing data, including whether any imputation was performed, is not reported in the registry record.

The record lists 130 enrolled participants, but it does not provide a complete accounting of the randomized population by arm for each endpoint. The adverse-event section records 65 participants at risk in each group; that information does not explain how the endpoint analysis populations were formed. The analysis population definition beyond the primary outcome description, such as intention-to-treat or per-protocol status, is not reported in the registry record.

The statistical analysis plan, any covariate-adjustment strategy, interim analyses, and prespecified handling of protocol deviations are also not reported in the registry record. The record lists no reported analyses. This leaves uncertainty about whether the posted comparisons were the only analyses performed or whether they represent a subset of a larger analysis plan. These absences are limitations of the public record, not evidence that a particular procedure was or was not used. They constrain evaluation of selection, missing-data, and multiplicity risks.

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 0% reporting threshold.

The adverse-event record lists 65 participants at risk in each group. It records zero serious adverse events and zero other adverse events in both the health-literacy intervention and usual-care groups, and zero deaths in each group. No serious adverse-event terms are listed.

These figures describe what was entered in the registry record. They do not establish that no event of any kind occurred outside the recorded categories, nor do they show whether events were actively collected or how they were defined. The record does not provide event-specific descriptions, severity criteria, follow-up details, or a statistical comparison between groups. It also does not explain how the adverse-event risk populations relate to the smaller endpoint populations.

Accordingly, the entries can be reported as zero recorded events in the listed categories, but they should not be taken as proof of absence of risk or as a comparative safety conclusion. The safety information is limited to the counts and categories recorded.

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

The posted primary data show similar average ARMS scores in the two observed groups, with a mean difference of -0.15. The intervention group’s mean is lower, and lower scores indicate better adherence on this scale. However, the reported confidence interval spans values on both sides of zero, and the unadjusted p-value is 0.841. The data therefore do not establish a clear direction or precise size for a difference in mean self-reported adherence.

This is not the same as demonstrating that the groups are equivalent. Equivalence requires a prespecified range of differences that would count as sufficiently similar and an analysis designed to test that question. The registry record does not report such a margin. Nor does the result show that the intervention has no effect in every participant or setting. It summarizes the observed scale scores for participants who provided data under the recorded trial conditions.

The ARMS is a self-report measure collected approximately 2 weeks after discharge. The result does not directly document medication ingestion, longer-term adherence, clinical events, or the effect of individual intervention components. The registry does not report a clinically meaningful threshold for the scale, so the practical importance of a particular score difference cannot be established from the posted means alone.

The endpoint data are available for fewer participants than total enrollment, and the reasons are not reported in the registry record. The uncertainty interval captures statistical uncertainty under the analysis that produced it; it does not account for every possible consequence of missing outcomes or measurement limitations. The record also does not report an adjusted analysis, original sample-size justification, or multiplicity strategy. Each absence limits how far the estimates can be generalized or interpreted as confirmatory.

The secondary self-efficacy result does not resolve the primary uncertainty. Taken together, the posted results describe the measured outcomes and their uncertainty, rather than supporting a definitive claim about the intervention’s effectiveness or lack of effectiveness. No medical conclusion follows from these statistics alone.

17. Why This Trial Matters Statistically

This trial raises a useful statistical question about how to evaluate behavioral support delivered during a transition from hospital to home. The intervention combined visual medication schedules, pill boxes, and pharmacist counseling, while the primary outcome captured participants’ self-reported medication adherence. That pairing makes it important to distinguish a change in a measured scale from a change in medication-taking behavior itself. Statistical interpretation begins with what the endpoint actually records.

The primary estimate is close to zero relative to its reported uncertainty, and the interval includes differences in both directions. This illustrates why an observed point estimate and a p-value do not provide a complete account of evidence. The interval describes how imprecisely the mean difference is estimated, while the scale’s scoring direction supplies context for interpreting either sign. Neither can answer whether a difference matters clinically without a meaningful threshold established for the measure.

The record also illustrates how enrollment and outcome-data counts answer different questions. Enrollment was 130, whereas the primary comparison includes 38 participants per group. The available record does not explain this difference. Recognizing the analysis population and the uncertainty about missing outcomes is essential when assessing how representative the result may be of all randomized participants.

Finally, the registry contains no reported analyses and does not describe the original sample-size calculation, multiplicity approach, or missing-data plan. Those omissions do not change the numerical summaries, but they limit evaluation of the inferential framework around them. Transparent reporting of estimands, analysis populations, missingness, and prespecified decisions is important for readers to distinguish a descriptive result from a fully documented confirmatory analysis.

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 important trial-specific details unresolved. It does not describe the usual-care condition, identify who was masked, or report how randomization was generated or concealed. It also does not provide detailed eligibility criteria, baseline characteristics, or the delivery and adherence of the intervention components. These omissions limit assessment of the contrast between groups and the context for the observed outcomes.

The primary outcome was reported for 38 participants in each group, while enrollment was 130. The endpoint population is described as participants who could be contacted and provided data, but reasons for missing primary outcome data and their distribution across the trial are not reported in the registry record. The secondary outcome has different group counts. Without a missing-data account or analysis-population definition, it is not possible to determine from the record how the posted results relate to all randomized participants.

The primary measure is self-reported and was collected approximately 2 weeks after discharge. The record does not provide a longer-term outcome or an objective adherence endpoint. It also does not report a clinically meaningful ARMS threshold, so the scale difference cannot be classified as practically important using information in the record.

The analysis plan, adjusted models, interim analyses, original sample-size justification, and multiplicity handling are not reported in the registry record. No analyses are listed as reported. The adverse-event entries are limited to zero recorded events in the listed categories, without further event definitions or follow-up detail. These reporting gaps do not show that particular methods or events were absent; they limit what can be verified from the registry information.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Improving Medication Adherence Through Graphically Enhanced Interventions in Acute Coronary Syndromes — NCT00399880.
• Registry record last updated 2017-09-25; status last verified 2017-07.
• Results first posted 2017-09-25.
• 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

NCT00399880 is a completed randomized, single-masked, parallel-group study of a health-literacy intervention compared with usual care among people hospitalized with acute coronary syndrome. Enrollment was 130, and the study was sponsored by Vanderbilt University Medical Center. The intervention included illustrated medication schedules, pill boxes, and pharmacist counseling.

The primary outcome was mean ARMS score approximately 2 weeks after discharge. The observed means were 12.03 for the intervention group and 12.18 for usual care; the mean difference was -0.15, with a reported confidence interval from -1.615 to 1.315 and an unadjusted p-value of 0.841. The comparison is imprecise and does not establish the direction or practical importance of a difference.

One secondary self-efficacy endpoint is posted. The registry record does not report an analysis plan, missing-data methods, multiplicity strategy, or original sample-size justification. Safety entries list zero serious adverse events, other adverse events, and deaths in both groups.