Caduet vs Usual Care in Subjects With Hypertension and Additional Risk Factors: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Caduet vs Usual Care in Subjects With Hypertension and Additional Risk Factors (NCT00407537), a randomized phase 4 study. The primary outcome is Framingham 10-year Risk of Total Coronary Heart Disease (CHD) at Month 12.

2. Clinical Question

The endpoint is a predicted-risk score rather than a count of cardiovascular events during follow-up. It uses gender-specific prediction equations incorporating age, diabetes, smoking, blood pressure categories, and total cholesterol and LDL cholesterol categories. The central statistical question is therefore about the difference in estimated risk scores at the stated assessment time, not whether the trial observed fewer CHD events. The record also lists secondary risk-score and blood-pressure outcomes, but the primary comparison is the Month 12 Framingham 10-year risk of total CHD.

3. Trial Design

This was a Phase 4, randomized, parallel-group trial with no masking recorded. The registry identifies the design as cluster randomized, so allocation was made at the cluster level rather than by treating every participant as an independent randomization unit. The record does not describe what constituted a cluster, how clusters were formed, or how many clusters took part. Those design details matter when interpreting both the analysis and its uncertainty.

The trial compared an open-label Caduet-based treatment strategy with usual care in clinical practice. Caduet was described as an amlodipine besylate/atorvastatin calcium single-pill combination added to the usual care regimen followed by investigators. The usual-care arm is named, but its specific treatment practices are not described in this record. The enrollment figure is 1531. The study started in 2007-03, reached primary completion in 2009-10, and is marked completed.

The primary endpoint was Framingham 10-year risk of total CHD at Month 12. The record describes a baseline and Month 12 timeframe and states that the score was derived at the end of study treatment. Its analysis population was the Full Analysis Set, defined as participants in either group with blood pressure and lipid data from at least one follow-up visit. The number contributing to the primary outcome analysis is not reported in the registry record.

Cluster randomization has implications beyond the choice of treatment contrast. Participants within the same cluster may have more similar outcomes or care than participants in different clusters. An analysis that treats all individuals as independent can therefore misstate uncertainty. The registry record explicitly cautions against a simple individual-level re-analysis and presents the sponsor's analysis for this outcome. Details of cluster-level allocation, cluster adjustment, and any correlation structure are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Prior to group assignment of investigators (who were the unit of randomization), participants were consented and screened for enrollment eligibility; participant selection for enrollment took place before investigator assignment. A total of 1,531 participants were enrolled and attended baseline visit; 1,461 participants received treatment.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full Analysis Set (FAS) included all subjects, in either treatment group, who had blood pressure and lipid data from at least 1 follow-up visit. Number of participants analyzed refers to number of participants contributing to 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 Framingham 10-year Risk of Total Coronary Heart Disease (CHD) at Month 12, measured over Baseline, Month 12.

The primary endpoint is the Framingham 10-year risk of total coronary heart disease at Month 12, expressed as percent risk. It is a prediction derived from gender-specific equations, not a direct measurement of whether a participant experienced a CHD event during the trial. The score uses age, diabetes, smoking, blood pressure categories, total cholesterol, and LDL cholesterol categories. The registry states that the coefficients were used to derive the score at the end of study treatment.

A predicted-risk score can provide a common way to combine several cardiovascular risk factors into an estimate of longer-term risk. In this trial, it allows the treatment groups to be compared on a calculated risk measure at a specified follow-up assessment. The outcome therefore reflects the inputs and structure of the prediction equation as well as participants' measured risk-factor data. It should not be read as the observed frequency of CHD events over the prediction horizon.

The registry gives baseline and Month 12 as the timeframe, but the posted primary result is a model-estimated difference at Month 12. It does not provide the group-specific baseline and Month 12 means in the registry record for this explanation. The result also does not report separate contributions from the score's component risk factors. Interpretation should stay focused on the modeled contrast in the composite predicted-risk measure.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is described as a least-squares mean difference from a mixed-effect linear model. These terms indicate that the reported comparison is model-based rather than simply a subtraction of two raw group averages. A linear model relates an outcome to treatment-group membership and any other modeled terms. The registry does not list all terms or covariates in the model, so the precise adjustment set is not reported in the registry record.

What does a least-squares mean difference represent?

A least-squares mean is a model-based mean for a group, interpreted under the model's specified structure. A difference between least-squares means expresses the modeled contrast between groups on the outcome scale. It can differ from a raw difference in observed averages when the model includes other terms or accommodates features of the design. The registry gives the contrast as -4.72 and its 95% confidence interval as -5.55 to -3.89. Because the subtraction order is not stated, the sign cannot be mapped to a named treatment group from this record alone.

Why use a mixed-effect model?

In general, mixed-effect models can represent both fixed effects, such as a treatment contrast, and random effects that account for grouping or other dependence in data. In a cluster-randomized trial, participants in the same cluster may share characteristics or care practices, so their outcomes may be correlated. Accounting for such correlation is important because treating correlated observations as independent can make uncertainty appear smaller than it is. The registry identifies the model as mixed-effect, but the random effects, cluster specification, estimation details, and covariance structure are not reported in the registry record. The record also cautions that an individual-level re-analysis would be misleading, so no simple re-calculation is presented.

How should the confidence interval be read?

In general, a confidence interval gives a range of estimates compatible with the observed data under the model and its assumptions. The reported 95% interval runs from -5.55 to -3.89. It indicates uncertainty around the modeled contrast, rather than a range containing individual participants' risk scores or future trial results. Its position entirely on one side of the null difference is consistent with the reported p < 0.001. The interval does not measure possible errors from incomplete outcome data, an unsuitable model, or other sources of bias.

What does the p-value say?

In general, a p-value describes how unusual results at least as incompatible with a specified null hypothesis would be if that hypothesis and the analysis assumptions held. The reported p < 0.001 is evidence against the null contrast within the sponsor's stated analysis. It is not the probability that the null hypothesis is true, and it does not establish clinical importance. The record does not report whether the p-value was adjusted for testing multiple outcomes or analyses.

What does the endpoint scale imply?

The outcome is a percent-risk score generated by the Framingham prediction equations. The modeled difference is therefore on the score's reported scale, not a count of observed events. No new conversion or group-specific mean can be obtained from the contrast alone. The score incorporates several risk-factor inputs, but the primary result does not separate the contribution of each input. The statistical interpretation should remain about the Month 12 model-estimated risk-score contrast.

Why is a new calculation not presented?

For some trials, posted group summaries can support an unadjusted recalculation. That is not appropriate for this record: its design is cluster randomized, and the registry explicitly says that a simple individual-level re-analysis would be misleading. The result described is the sponsor's mixed-effect model analysis. Planning illustrations based on an observed effect, when used, are design calculations rather than observed trial outcomes or observed power. The assumptions noted for those illustrations are two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation; time-to-event illustrations use total events, while other illustrations use participants per arm. No numerical planning results are stated in the information available for this explanation, and these calculations do not establish the trial's original design rationale.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Caduet", "Usual Care")
n   <- c(748, 655)
m   <- c(12.5, 16.3)
sp  <- c(8.26, 10.51)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Caduet  748  12.500  8.260  11.907  13.093
CARM  Usual Care  655  16.300  10.510  15.494  17.106

11. Analyses Reported by the Sponsor

The primary result is the sponsor's own analysis: a mixed-effect linear model reporting a least-squares mean difference of -4.72, a 95% confidence interval from -5.55 to -3.89, and p < 0.001. The registry's cluster-randomized design makes a simple individual-level re-analysis potentially misleading, and the registry explicitly indicates that only the sponsor's analysis is shown for this comparison. No recalculated estimate is presented alongside it.

The model-based contrast should be interpreted as reported, including its uncertainty interval and p-value. The record does not state the group subtraction order, so the sign cannot be attributed confidently to Caduet or usual care. It also does not give the full model specification or enough information to independently reproduce the analysis. The available result is therefore a report of the sponsor's modeled comparison, not an independently checked calculation from group-level summaries.

The confidence interval and p-value indicate statistical evidence under the sponsor's analysis, but they do not resolve questions about cluster handling, outcome completeness, or clinical relevance. Those questions require design and analysis details that are not reported in the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary endpoint alongside several secondary outcomes, including Framingham and European SCORE risk measures at different assessment times, change-from-baseline measures, and blood pressure. It reports 22 outcome measures and 26 analyses. When several outcomes or analyses are examined, the chance of obtaining at least one apparently strong result by chance can increase. Statistical procedures may account for this multiplicity, but the record does not report whether an adjustment was used, which comparisons were prespecified, or how the primary result was protected from multiplicity.

The primary outcome population is the Full Analysis Set. It includes participants in either treatment group who had blood pressure and lipid data from at least one follow-up visit; the number analyzed for the primary outcome is not reported in the registry record. Requiring follow-up data means that the analysis population is defined in part by data availability after baseline. The record does not explain how participants without qualifying follow-up data were handled, why data were missing, or whether missing values were imputed.

The record describes the outcome timeframe as baseline and Month 12 and gives a Month 12 model-based result. It does not report the amount or pattern of missing outcome data at that visit, whether the model used all available repeated measurements, or any sensitivity analyses for missingness. These details matter because conclusions can depend on assumptions about why measurements are unavailable and how incomplete records relate to outcomes.

Other components of the statistical plan are also not reported in the registry record. This includes the sample-size justification, any interim analyses, the complete model specification, and details of the cluster-level analysis. The registry does state planning assumptions for sample-size illustrations, but those are not the trial's reported design rationale. Without the full analysis plan and outcome-completeness information, the reported result can be described, but several sources of uncertainty about analysis choices cannot be evaluated.

13. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: urinary tract infection (0.4%), acute myocardial infarction (0.3%), myocardial ischaemia (0.3%), myocardial infarction (0.3%), chest pain (0.3%).

The adverse-event record lists serious adverse events for Caduet and usual care separately. The number at risk is 779 for Caduet, with 51 affected, and 682 for usual care, with 22 affected. These are the recorded counts; no new percentages or comparative rates are calculated. The record also lists other adverse events, with 356 affected among 779 at risk for Caduet and 289 affected among 682 at risk for usual care.

The serious-event terms listed with percentages are urinary tract infection at 0.4%, acute myocardial infarction at 0.3%, myocardial ischaemia at 0.3%, myocardial infarction at 0.3%, chest pain at 0.3%, and appendicitis at 0.3%. The record does not specify in this information whether the listed terms are pooled across groups or provide group-specific counts for each term. These listed terms should not be read as a complete inventory of all serious adverse events.

Deaths are not reported in the registry record. The count of participants affected by serious adverse events is a safety description, not by itself a causal comparison. The registry information does not provide enough detail about event ascertainment, exposure time, severity beyond the serious-event classification, or statistical comparisons to support a fuller assessment of relative safety.

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

The statistical result shows that the sponsor's mixed-effect linear model estimated a difference in Framingham 10-year risk of total CHD at Month 12. The estimate is -4.72, with a 95% confidence interval from -5.55 to -3.89 and p < 0.001. This describes a modeled contrast in a predicted-risk score. The registry record does not identify the group subtraction order, so the sign cannot be used by itself to say which named group had the lower modeled score.

The score is based on risk-factor information and predicts 10-year CHD risk. It is not an observed count of CHD events during the trial, and the result does not establish a difference in actual event incidence. It also does not demonstrate the practical importance of the score contrast. Clinical importance would require interpretation in context, including information about the score, the population, and the consequences of changes; the registry result alone does not supply that context.

The trial compared a Caduet-based strategy added to usual care with usual care in a cluster-randomized setting. The result concerns that treatment strategy and the measured risk score at Month 12. It should not be generalized automatically to other regimens, settings, or patient groups. The usual-care practices, cluster details, full model specification, and primary analysis sample size are not reported in the registry record.

The confidence interval conveys precision under the sponsor's model assumptions, not certainty about all aspects of the evidence. The record does not describe multiplicity adjustment, handling of missing data beyond the Full Analysis Set definition, or sensitivity analyses. These limitations do not erase the reported statistical result, but they constrain how far it can be interpreted. No medical decision follows from the statistical summary alone.

15. Why This Trial Matters Statistically

This trial is statistically notable because the randomized allocation was at the cluster level, while the primary endpoint was measured on individual participants. That combination makes the unit of allocation and the unit of measurement distinct. In general, outcomes within a cluster may be correlated, and an analysis needs to account for that dependence to estimate uncertainty appropriately. The registry identifies a mixed-effect linear model and cautions against a simple individual-level re-analysis, underscoring why the design cannot be reduced to a basic comparison of participant-level summaries.

The primary outcome also illustrates the difference between a predicted-risk score and a clinical event endpoint. Framingham 10-year CHD risk combines multiple risk-factor inputs to produce a long-term prediction. The reported result concerns the score at Month 12; it does not report observed CHD events over the prediction horizon. Distinguishing predicted risk from observed outcomes is essential when describing what a trial result measures.

The record further highlights why statistical significance and practical importance are separate questions. The estimate, confidence interval, and p-value describe the sponsor's model-based comparison and its uncertainty under the analysis. They do not tell readers whether the difference is meaningful in clinical practice, nor do they establish which group had the lower score when contrast coding is not stated.

Finally, interpretation depends on transparent reporting of the analysis population, missing data, multiplicity, and cluster handling. The record defines the Full Analysis Set but does not report the number contributing to the primary outcome or the detailed methods for incomplete data. These omissions make the posted result interpretable at a high level while limiting independent assessment of its robustness.

16. Limitations

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

The registry record leaves several design and analysis details unresolved. Although the trial is identified as cluster randomized, it does not say what the clusters were, how many clusters were enrolled, how allocation was implemented, or whether cluster sizes varied. The result is attributed to a mixed-effect linear model, but the random-effects structure, covariance assumptions, model covariates, and estimation procedure are not reported. These omissions limit the ability to assess whether the analysis fully reflects the cluster design.

The contrast is reported without stating which group was subtracted from which. As a result, the negative estimate cannot be assigned a direction for Caduet versus usual care. Group-specific primary endpoint means, baseline values, and the number contributing to the endpoint analysis are not included in the information available for this explanation. The Full Analysis Set definition is stated, but further detail about exclusions and missing-data handling is not reported in the registry record.

The registry lists multiple outcomes and analyses but does not report a multiplicity strategy. It also does not report the sample-size justification, interim analyses, detailed protocol, or sensitivity analyses. The usual-care regimen and the specifics of cluster-level delivery are not described. The record's serious adverse-event summary does not provide group-specific counts for each listed event term, and deaths are not reported in the registry record.

These are limitations of what can be assessed from the registry record, not proof that the underlying trial lacked procedures or analyses. The result remains the sponsor's reported model-based estimate, with the stated confidence interval and p-value, but independent evaluation of its assumptions and robustness is constrained by the missing detail.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Caduet vs Usual Care in Subjects With Hypertension and Additional Risk Factors — NCT00407537.
• Registry record last updated 2021-01-28; status last verified 2021-01.
• Results first posted 2011-11-11.
• 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.

20. Record Summary

NCT00407537 is a completed Phase 4, randomized, parallel-group, cluster-randomized trial of a Caduet-based strategy versus usual care in people with hypertension and additional cardiovascular risk factors. The Caduet strategy used an amlodipine besylate/atorvastatin calcium single-pill combination added to usual care. The record lists enrollment of 1531 and dates of 2007-03 to 2009-10.

The primary endpoint was Framingham 10-year risk of total CHD at Month 12. The sponsor's mixed-effect linear model reported a least-squares mean difference of -4.72, a 95% confidence interval from -5.55 to -3.89, and p < 0.001. The comparison's subtraction order is not identified in the registry record, and the cluster-randomized design is why no simple individual-level re-analysis is presented.

The record also lists secondary risk-score and blood-pressure outcomes, adverse-event counts, and serious-event terms. Important analysis details—including the full model specification, multiplicity handling, and missing-data methods—are not reported in the registry record.