Multi-arm Intervention Diabetes Adherence Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Multi-arm Intervention Diabetes Adherence Study (NCT00754741), a randomized phase 3 study. The primary outcome is Glycated Hemoglobin Levels.

2. Clinical Question

The registry describes a comparison of two approaches intended to improve adherence to diabetes medications among patients with diabetes and poor metabolic control. The primary outcome was not medication adherence itself: it was glycated hemoglobin, reported as a group mean with a standard deviation. Medication adherence and other metabolic outcomes were listed as secondary outcomes.

The study was randomized and used a parallel-group model, with usual care serving as the comparator for each intervention. The central statistical question is whether the observed between-group differences in mean glycated hemoglobin are sufficiently precise to distinguish a difference from no difference. The posted summaries show values close to the usual-care mean, while the recalculated confidence intervals for both comparisons include zero. That pattern describes the available estimates; it does not establish that the approaches are equivalent or that any effect is absent.

3. Trial Design

The registry identifies this as a phase 3, randomized, single-masked, parallel-group trial. It reports enrollment of 1692 and a completed status. The recorded study start was 2008-07, and primary completion was 2010-04. The sponsor was Henry Ford Health System. The registry record does not report who was masked, nor does it give additional details about how randomization or masking was implemented.

The three groups were usual care, adherence information, and adherence information plus motivational interviewing. The usual-care description says physicians received limited training in responding to poor adherence and did not receive patient adherence information through electronic prescribing software for patients assigned to that group. The detailed descriptions for the two intervention groups are not reported in the registry record, so the content and delivery of those interventions cannot be characterized further from this record.

The primary outcome was mean glycated hemoglobin, in percent, at 18 months post randomization. The posted group summaries include a mean and standard deviation for each group. Listed secondary outcomes include medication adherence, lipid-lowering drug adherence, glycated hemoglobin, LDL cholesterol, and an exploratory cardiovascular morbidity and mortality outcome. Some outcome titles appear more than once in the registry list. The record lists 14 outcome measures and 28 analyses, while 12 secondary outcomes are posted.

The registry record does not report the trial’s original sample-size assumptions, the statistical analysis plan, or details such as allocation concealment. Those omissions limit what can be inferred about the original design and the intended analysis. The planning illustrations based on observed effects are a separate calculation and should not be read as the trial’s original design rationale.

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

Reasons for not completing this period:

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 Glycated Hemoglobin Levels, measured over at 18 months post randomization.

The primary endpoint is glycated hemoglobin at 18 months post randomization, summarized as a mean percentage and standard deviation in each group. Glycated hemoglobin reflects blood glucose exposure over time and is used as a measure of metabolic control in diabetes. For a continuous outcome, comparing group means describes the average level recorded in each randomized group at the stated follow-up.

This endpoint is relevant to the study’s stated aim because the registry describes participants as having diabetes and poor metabolic control, and the intervention goal was to improve medication adherence and metabolic control. However, the reported primary outcome is a follow-up group mean. The registry record does not provide baseline glycated hemoglobin values or a description of change from baseline, so the posted summaries alone do not show within-group change over time.

A mean difference near zero indicates that the reported group averages are close. The standard deviation describes variation among observations within each group; it is not the uncertainty around the group mean. Uncertainty in the between-group estimate is instead reflected by its confidence interval. Interpretation therefore depends on both the estimated difference and its precision, as well as on the analysis specified by the trial investigators. The registry record does not report the primary analysis plan in enough detail to reconstruct that analysis fully.

7. Results Posted in the Registry

• Adherence Information: mean 7.91 versus 7.88 on control; difference 0.03 (95% CI -0.15 to 0.21), standardized effect d = 0.02; z-test p = 0.741.
• Adherence Information Plus Motivational Interviewing: mean 7.79 versus 7.88 on control; difference -0.09 (95% CI -0.26 to 0.08), standardized effect d = -0.06; z-test p = 0.294.

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.

Adherence to Oral Anti-diabetic Medications

• 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:
  • Adherence Information: mean 0.74 versus 0.75 on control; difference -0.01 (95% CI -0.05 to 0.03), standardized effect d = -0.03; z-test p = 0.635.
  • Adherence Information Plus Motivational Interviewing: mean 0.74 versus 0.75 on control; difference -0.01 (95% CI -0.05 to 0.03), standardized effect d = -0.03; z-test p = 0.642.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information Plus Motivational Interviewing and the control. The interval gives the range of effects that remain compatible with the data.

Adherence to Lipid-lowering Drugs

• 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:
  • Adherence Information: mean 0.69 versus 0.70 on control; difference -0.01 (95% CI -0.05 to 0.03), standardized effect d = -0.03; z-test p = 0.644.
  • Adherence Information Plus Motivational Interviewing: mean 0.69 versus 0.70 on control; difference -0.01 (95% CI -0.05 to 0.03), standardized effect d = -0.03; z-test p = 0.646.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information Plus Motivational Interviewing and the control. The interval gives the range of effects that remain compatible with the data.

Cardiovascular Morbidity and Mortality (Exploratory)

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Adherence Information: 14.1% (80/568) versus 11.1% (63/567) on control; risk difference 3.0 percentage points (95% CI -0.9 to 6.8), risk ratio 1.27 (0.93–1.73); unadjusted two-proportion z-test p = 0.131.
  • Adherence Information Plus Motivational Interviewing: 10.4% (58/556) versus 11.1% (63/567) on control; risk difference -0.7 percentage points (95% CI -4.3 to 2.9), risk ratio 0.94 (0.67–1.31); unadjusted two-proportion z-test p = 0.714.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Adherence Information Plus Motivational Interviewing and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes span medication adherence, metabolic measures, and an exploratory cardiovascular outcome, with follow-up times that differ from the primary endpoint. The available summaries do not suggest a broad pattern of clearly separated group means for the posted adherence measures. The exploratory cardiovascular outcome likewise has uncertain between-group estimates in the unadjusted comparisons. This overview describes the direction and certainty of the posted results without treating different outcome types as interchangeable.

The secondary findings complement, but do not replace, the primary glycated hemoglobin comparison. Adherence measures are closer to the intervention’s stated behavioral aim, while glycated hemoglobin and LDL cholesterol assess metabolic outcomes. The cardiovascular measure is identified as exploratory and was assessed at a later follow-up. A pattern across these outcomes can help describe whether results align conceptually, but it cannot by itself establish a causal pathway or explain why an outcome did or did not differ.

Unadjusted secondary results carry less weight as confirmatory evidence when many outcomes or comparisons are examined. With multiple tests, some apparently notable results may arise by chance even if there is no underlying effect. The registry lists multiple outcomes and analyses, but does not report a multiplicity strategy or identify which secondary analyses, if any, were prespecified as confirmatory. Secondary findings should therefore be treated as supportive or exploratory unless the full analysis plan establishes otherwise. The registry record also does not provide full results for every outcome listed, so this overall reading is limited to the posted summaries.

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

• Adherence to Oral Anti-diabetic Medications
• Adherence to Lipid-lowering Drugs
• Cardiovascular Morbidity and Mortality (Exploratory)
• Glycated Hemoglobin Levels
• LDL Cholesterol Levels
• Glycated Hemoglobin Levels
• LDL Cholesterol Levels
• Adherence to Oral Anti-diabetic Medications

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: glycated hemoglobin is summarized by a mean and standard deviation. The posted comparisons express the difference between an intervention-group mean and the usual-care mean. Under this convention, a positive difference means the intervention-group mean is higher, and a negative difference means it is lower. These signs describe the direction of the estimate, not whether a difference is beneficial or clinically important.

What does a confidence interval express?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. It conveys estimation uncertainty, rather than listing the range of individual participant values. A narrow interval indicates greater precision for the estimated contrast; a wide interval indicates less precision. When an interval for a mean difference includes zero, the data are compatible with no difference under that interval procedure, but they may also be compatible with nonzero differences on either side. This is not proof of equality.

How should a p-value be read?

A p-value describes how unusual results at least as discrepant as those observed would be under a specified null model and its assumptions. The sponsor reported ANOVA analyses for the primary comparisons. The simple recalculations use posted group summary statistics and are unadjusted comparisons; they do not recreate every possible feature of the sponsor’s analysis. A p-value does not tell the reader the probability that the null is true, the probability that the result will replicate, or the clinical importance of an estimate.

What is the role of the standard deviation?

The standard deviation describes spread of individual observations around a group mean. It is distinct from the standard error, which describes the estimated variability of a statistic across repeated samples under a model. Greater variability can make a mean difference harder to estimate precisely, all else being equal. The posted primary summaries report standard deviations of 1.53 in usual care, 1.53 with adherence information, and 1.34 with adherence information plus motivational interviewing.

What does the recalculation add?

The recalculated mean differences and intervals offer a transparent, unadjusted comparison derived from the posted means and variability summaries. They are not a replacement for the sponsor’s analysis. The registry reports ANOVA p-values but does not report the model details, covariates, analysis population, or missing-data approach needed to reproduce that analysis. Since the record does not flag a discrepancy, the recalculations can be read as a descriptive cross-check, not as a claim that the sponsor’s analysis is incorrect.

Why can trial analyses differ?

Analyses may differ when they use different populations, adjustments, assumptions about repeated measurements, or approaches to incomplete observations. Randomization supports comparison of assigned groups, but the handling of missing outcome values and deviations from assignment can affect the analysis population and estimate. The registry record does not report these details for the primary endpoint, so it is not possible to determine whether the sponsor used adjusted or otherwise specialized methods beyond the recorded ANOVA label.

What assumptions matter for continuous outcomes?

Methods comparing means rely on assumptions about the data and the sampling distribution of the estimate. For example, an analysis may be sensitive to unusually skewed observations or influential values, depending on the method and sample size. General statistical practice is to examine whether assumptions are reasonable and to consider robust or alternative methods when they are not. The registry record does not report diagnostic checks or sensitivity analyses, so no claim can be made about whether these were performed.

For the exploratory cardiovascular outcome, the measure is a count of participants with the outcome, and the posted comparisons include risk differences. A risk difference describes the difference between the proportions in the two groups. Its uncertainty interval is useful alongside the point estimate because it shows how precisely the contrast is estimated. The sponsor’s reported analysis for this outcome is chi-squared. This is a different outcome type from the continuous primary endpoint and should not be interpreted as if it were a mean comparison.

11. Confidence Intervals and Precision

The recalculated primary confidence intervals include zero for both intervention comparisons. That means the estimates do not precisely separate a nonzero difference from no difference using these unadjusted interval calculations. The intervals also allow values on either side of zero, so they do not establish that the groups are equivalent. The posted means are close, but closeness of point estimates alone is not enough to determine precision or rule out potentially relevant differences.

Confidence intervals are most useful when read together with the estimate and the outcome scale. For glycated hemoglobin, the estimated differences are expressed in percentage units, and the interval describes uncertainty about the difference in group means. The standard deviations describe participant-level variation, not uncertainty in the contrast itself. In general, an interval can be wide even when a point estimate is close to zero, and a small p-value would not itself indicate that an estimate is precise or important.

The registry’s sponsor-reported results are p-values without reported confidence intervals or effect estimates. The intervals are from simple unadjusted recalculations based on the posted summaries. They should therefore be distinguished from intervals that might result from the trial’s original analysis model, which is not described in the registry record. Precision is also limited by unreported information about missing observations, analysis populations, and any adjustments. The available summaries support a cautious reading: the observed means are similar, but the record does not provide enough detail to make a formal equivalence claim or to evaluate every source of uncertainty.

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 figures are planning illustrations based on the observed effects, not the trial’s original design calculations. For these outcomes, the figures are participants per arm. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the effect to detect is the stated fraction of the observed effect. The original trial’s sample-size assumptions are not reported in the registry record.

The 80% power column gives the number of participants per arm that the stated planning method estimates would be needed for an 80% chance of detecting the specified effect if that effect were real and the assumptions held. The 90% power column expresses the corresponding planning target for a 90% chance. These are design calculations for a future study under the listed assumptions, not a description of what the completed trial achieved.

Rows for smaller fractions of the observed effect generally require more participants per arm, because smaller differences are harder to distinguish from random variation. The observed effects used in these illustrations may be unstable estimates, especially when the underlying difference is small; the resulting figures should not be treated as definitive design recommendations. This is not observed power. Observed power would not clarify the uncertainty in the completed trial’s result; the estimates and confidence intervals are more informative for interpreting that uncertainty.

The planning assumptions concern pairwise comparisons between two arms. They should not be read as describing the original three-group randomization or any adjustment for multiple comparisons, since such adjustment is not stated in the assumptions and the original design details are not reported in the registry record.

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 = 0.030; pooled standard deviation σ = 1.530.
2. n = 2 × (1.96 + 0.842)² × 1.530² / 0.030² = 40,829.9, rounded up to 40,830 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 <- 0.0300000000000002                      # observed difference in means
sigma <- 1.53                     # 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(754741)
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 = 40,830 (80% power), 54,660 (90% power)
 75% of observed effect: n per arm = 72,587 (80% power), 97,173 (90% power)
 50% of observed effect: n per arm = 163,320 (80% power), 218,639 (90% power)
power.t.test: 40830.8 
Simulated power at n = 40,830 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. Analyses Reported by the Sponsor

The registry reports sponsor analyses using ANOVA for the two primary comparisons. The reported p-values are 0.763 for adherence information versus usual care and 0.285 for adherence information plus motivational interviewing versus usual care. The simple unadjusted recalculations based on the posted group summaries give p-values of 0.741 and 0.294, respectively.

These are different analyses: the recalculations use the posted means and standard deviations and do not incorporate model details that are not reported in the registry record. Their results are close in broad interpretation to the sponsor’s reported p-values, but they should not be treated as exact reproductions of the sponsor analysis. The record does not flag a discrepancy between the posted and recalculated results. If a discrepancy were flagged, the sponsor’s own analysis would take precedence.

The sponsor’s p-values do not provide an effect estimate or confidence interval in the posted analysis record. They should be read as evidence from the reported ANOVA tests, not as measures of effect size, clinical importance, or the probability that a hypothesis is true. Interpretation is strongest when the sponsor’s analysis is considered alongside the group means and the uncertainty intervals from the unadjusted recalculation, while recognizing that the original statistical plan is not reported.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists many outcome measures and analyses, including repeated outcome titles, but does not report a multiplicity adjustment or a hierarchy specifying which outcomes were confirmatory. When multiple outcomes or comparisons are tested, the chance of at least one apparently unusual result can increase. General statistical practice is to define the primary outcome and planned contrasts in advance, and to describe whether secondary tests are adjusted or interpreted as exploratory. Without that information, secondary p-values should not be read as if each were the sole confirmatory test.

The record gives a primary outcome and sponsor-reported analyses for its comparisons, but it does not report the full statistical analysis plan. It also does not report whether interim analyses were planned or conducted. No inference about such procedures can be made from the posted record.

Missing-data handling is not reported in the registry record. The posted summaries do not specify how many randomized participants had primary outcome measurements available, why observations might be missing, or whether the analysis used imputation or another approach. Missingness can affect both the population represented by an estimate and the precision of that estimate, particularly if the likelihood of missing data differs across groups or relates to outcome values.

The analysis populations are also not described in enough detail to determine whether the sponsor analyzed participants according to randomized assignment, treatment received, or another population definition. This matters because the estimand—the precise treatment effect a study seeks to estimate—depends on population and handling of events or deviations after randomization. The randomized design is reported, but the registry record does not provide the information needed to assess how the primary analysis preserved or modified that comparison. These gaps do not invalidate the posted results; they limit how fully their methods and potential biases can be evaluated.

16. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry’s adverse-event summary records zero participants affected by serious adverse events in each of the three groups. It also records zero participants affected by other adverse events in each group. These are the posted counts for the listed groups; they do not by themselves establish that no adverse events occurred outside the registry’s definitions or reporting procedures.

Deaths are listed as not reported in the registry record. The adverse-event summary does not include serious-event terms, and further detail about event collection, definitions, follow-up, or attribution is not reported. Consequently, the zero counts should be read as the registry’s posted figures, not as a complete account of all safety monitoring or as evidence about the comparative safety of the interventions. The record does not provide enough information to assess whether event ascertainment or reporting was equally complete across groups.

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

The posted primary results show similar mean glycated hemoglobin values across usual care and the two intervention groups at 18 months post randomization. The recalculated mean differences are small, and both unadjusted confidence intervals include zero. The sponsor-reported ANOVA p-values also do not indicate strong evidence of a between-group difference under those tests. This describes the estimates and analyses recorded; it is not a statement that the interventions are equivalent or that they cannot have an effect.

The results do not demonstrate that medication adherence improved or did not improve, because the primary endpoint is glycated hemoglobin and the available adherence measures are secondary outcomes. Nor do these summaries establish a change from baseline: baseline glycated hemoglobin values and a change analysis are not reported in the registry record. The outcomes occur at different follow-up times and answer different questions.

Statistical evidence and clinical importance are not the same. A confidence interval describes uncertainty around an estimate under a method, while a clinical interpretation also depends on context and a prespecified judgment about what difference would matter. The registry record does not report a threshold for a clinically important difference or an equivalence margin. It therefore cannot support a formal conclusion about clinical equivalence or the practical importance of a particular difference.

Finally, these registry summaries do not establish how results would apply to patients outside the study population described, or how the interventions were delivered in detail. The record’s incomplete reporting of analysis populations, missing data, and multiplicity further limits interpretation. The most defensible conclusion is restricted to the posted group summaries and reported comparisons: the observed primary means were close, and the available analyses do not clearly distinguish either intervention from usual care on this endpoint.

18. Why This Trial Matters Statistically

This trial raises several useful statistical questions for clinical researchers. It compares more than one behavioral approach with usual care, while focusing its primary assessment on a continuous metabolic outcome. That makes the distinction between an observed mean difference, its uncertainty, and a test result especially important. Similar means are not the same as proof of no effect, and a nonsignificant test is not an equivalence analysis.

The record also illustrates why reporting the analysis plan matters. The sponsor’s ANOVA p-values and simple recalculations based on summary statistics are broadly similar in interpretation, but the posted record does not give enough model detail to reproduce the sponsor’s approach. Reporting effect estimates and confidence intervals from the primary analysis, together with population and missing-data definitions, would make the result easier to assess.

Multiple outcomes create another interpretive challenge. This study lists adherence measures, metabolic outcomes, and an exploratory cardiovascular outcome. These outcomes may help describe different parts of the proposed intervention pathway, but testing many outcomes can make isolated findings less persuasive without a prespecified strategy for multiplicity. Clear labeling of primary, secondary, and exploratory analyses helps readers distinguish confirmatory evidence from hypothesis-generating observations.

Lastly, planning figures based on an observed effect can be informative as illustrations, but they are not a substitute for an original design rationale. Small observed differences can translate into very large planning requirements, while observed effects themselves may be imprecise. The registry’s missing original sample-size assumptions means the planning illustrations should remain separate from judgments about whether the trial was designed appropriately.

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 registry record leaves important trial details unclear. It identifies the design as single-masked but does not state who was masked or describe the masking procedures. It gives limited information about usual care, while detailed descriptions of the two intervention arms are not reported. These omissions make it difficult to assess intervention delivery or the degree to which the groups differed in practice.

The posted primary result is a follow-up group mean. Baseline values, change-from-baseline results, and details of the primary analysis model are not reported. The record does not specify analysis populations, missing-data handling, covariate adjustment, or sensitivity analyses. As a result, the simple recalculations are necessarily less detailed than a full trial analysis and cannot resolve possible effects of incomplete data or model choices.

The registry lists multiple outcomes and analyses, but a multiplicity plan is not reported. Some outcome titles are repeated, and not all listed secondary outcomes have detailed summaries in the posted results. The cardiovascular outcome is labeled exploratory, so it should not be treated as confirmatory evidence. The record also does not report a clinically important difference or an equivalence margin for the primary outcome.

The adverse-event record reports zero affected participants for serious and other events but does not provide event terms or explain ascertainment and reporting. Deaths are not reported in the registry record. Finally, the original sample-size justification and design assumptions are not reported, so the planning illustrations based on observed effects cannot be used to evaluate the trial’s original design rationale.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Multi-arm Intervention Diabetes Adherence Study — NCT00754741.
• Registry record last updated 2015-05-08; status last verified 2015-04.
• Results first posted 2015-04-22.
• 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

NCT00754741 is a completed, randomized, single-masked, parallel-group phase 3 trial sponsored by Henry Ford Health System. The registry reports enrollment of 1692 and a study start of 2008-07, with primary completion in 2010-04. The groups were usual care, adherence information, and adherence information plus motivational interviewing.

The primary outcome was mean glycated hemoglobin, in percent, at 18 months post randomization. Posted means were 7.88 for usual care, 7.91 for adherence information, and 7.79 for adherence information plus motivational interviewing. The sponsor reported ANOVA p-values of 0.763 and 0.285 for the two comparisons with usual care. Simple unadjusted recalculations give estimates close to zero, with confidence intervals including zero.

Secondary outcomes included medication adherence, metabolic measures, and exploratory cardiovascular morbidity and mortality. Key details about the original analysis plan, missing-data handling, analysis populations, multiplicity, and the original sample-size rationale are not reported in the registry record.