Lexapro and Pramipexole and to Treat Major Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Lexapro and Pramipexole and to Treat Major Depression (NCT00086307), a randomized phase 2 study. The primary outcome is Montgomery Asberg Depression Rating Scale (MADRS).

2. Clinical Question

3. Trial Design

The registry describes a randomized, quadruple-masked, parallel-group phase 2 study sponsored by the National Institute of Mental Health. Its status is completed. The registry lists enrollment as 39 and describes three treatment groups: pramipexole, escitalopram, and escitalopram with pramipexole. The posted arm summaries list 13 participants in each group.

The brief summary describes a 9-week study. Before treatment, participants were tapered off antidepressants or other prohibited medications and remained drug-free for 1 week. They were then assigned to a 6-week treatment period. Participants were scheduled for clinic visits for health and symptom assessments, including depression and anxiety ratings, adverse events, and sexual functioning. Blood and urine samples were also collected periodically, according to the summary.

The registry gives start and primary completion dates of 2004-06 and 2010-03, respectively. These dates describe the registry record and do not by themselves explain the duration of recruitment or follow-up. The record does not report the allocation ratio, the precise masking roles, the randomization procedure, or the original statistical design assumptions. Those details should not be inferred from the study's randomized and quadruple-masked labels.

The eligibility description narrows the population to adults with a current major depressive episode and a history of inadequate response to an SSRI, within the stated limits on previous antidepressant failures for that episode. The posted results should therefore be read in the context of that selected study population, rather than as a comparison across all people with depression.

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:

Analysis population for the primary outcome, as defined in the registry: All patients with at least one post-baseline measurement were included in the 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 Montgomery Asberg Depression Rating Scale (MADRS), measured over Weekly.

The primary outcome was the Montgomery Asberg Depression Rating Scale (MADRS), reported as a score on a scale. The registry describes MADRS as a 10-item instrument for assessing depression severity. Each item is rated from 0 to 6; the total score can range from 0 to 60, with higher scores indicating more severe depression. Scores of 18 or greater are generally considered to indicate a moderate level of depression, according to the registry description.

A continuous rating scale can capture degrees of symptom severity rather than reducing the outcome to a simple yes-or-no category. In general, repeated symptom ratings can be used to describe changes over time and compare group averages. The registry lists the primary measure's timeframe as weekly and reports least-squares means with standard errors. It states that patients with at least one post-baseline measurement were included in the analysis.

The outcome is a symptom rating, not a direct measure of every aspect of a participant's functioning or experience. The registry record does not report a threshold for what would count as a clinically important difference in MADRS score. The score summaries should therefore be interpreted as measured depression severity, with attention to the scale's direction and to the uncertainty around group comparisons.

7. Results Posted in the Registry

• Escitalopram: mean 26.10 versus 22.63 on control; difference 3.47 (95% CI -0.71 to 7.65), standardized effect d = 0.64; z-test p = 0.103.
• Escitalopram and Pramipexole: mean 29.45 versus 22.63 on control; difference 6.83 (95% CI 2.32 to 11.33), standardized effect d = 1.17; z-test p = 0.003.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous, so the posted arm summaries use means rather than event counts. The primary record describes the group estimates as least-squares means and their dispersion as standard errors. A least-squares mean is a model-based group estimate; it can differ from a simple arithmetic average when a model accounts for other features of the data. The registry identifies mixed models as the method for the sponsor-reported analyses, but does not report the model specification, the covariance structure, the exact time points included, or the covariates used. The interpretation of model-based estimates must remain limited by that missing detail.

A standard error describes the estimated precision of a mean, not the spread of individual observations. The posted standard errors are 1.508 for pramipexole, 1.507 for escitalopram, and 1.732 for the combination. In general, the standard error reflects uncertainty in estimating a group mean; it should not be read as a range containing most participants' scores. The registry does not provide standard deviations for the arm summaries.

The comparison calculations use the pramipexole arm as the reference and report mean differences for the other arms. A positive difference means that the comparison arm's reported mean is higher than the reference mean. Since higher MADRS scores represent more severe depression, the direction of a positive difference corresponds to a higher average severity score in the comparison arm. This sign convention is important: a difference is not inherently favorable or unfavorable without considering which group is subtracted from which and the scale's direction.

The recalculated comparisons are simple, unadjusted calculations from the posted arm summaries. They use pooled standard deviations that are marked as approximate, because the posted summaries give standard errors rather than the arm standard deviations used in a conventional pooled calculation. The resulting estimates and p-values are not a reproduction of the registry's mixed-model analysis. The record explicitly contains a discrepancy, so the sponsor's own analysis takes precedence for the sponsor-reported trial result.

A confidence interval gives a range of effect estimates compatible with the data and the method's assumptions. In general, a wider interval reflects less precision. A p-value summarizes how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and the model assumptions; it is not the probability that the null hypothesis is true. Neither measure conveys clinical importance on its own. The posted sponsor analyses report 95% confidence intervals and p-values, while the recalculated comparisons are unadjusted.

Mixed models are commonly used for repeated measurements because observations from the same participant over time are related rather than independent. In general, accounting for that within-person correlation can make estimates and uncertainty more appropriate than treating every measurement as unrelated. The registry's primary timeframe is weekly, and its reported method is Mixed Models Analysis. However, it does not provide enough information to determine the exact repeated-measures structure, how time entered the model, whether a treatment-by-time term was used, or how missing weekly observations were handled.

The distinction between an overall analysis and pairwise analyses also matters. The sponsor reports a mixed-model analysis involving all three groups, followed by post hoc simple effects tests for pairs of groups. In general, testing several comparisons can increase the chance of at least one apparently notable result if the analyses are treated as independent opportunities. Whether a multiplicity adjustment was applied is not reported in the registry record. The post hoc label indicates that those comparisons followed another analysis, but does not itself establish how multiplicity was controlled.

The reported analysis population includes all patients with at least one post-baseline measurement. This is not the same as a statement that every randomized participant contributed every weekly outcome. The registry record does not report the amount or pattern of missing outcome data, the method used to handle missing measurements, or a sensitivity analysis for missingness. These details matter because methods for incomplete repeated data depend on assumptions about why measurements are missing.

Finally, the recalculated effect sizes and planning illustrations are not substitutes for the trial's own analysis plan. The record does not report the original sample-size justification, planned primary contrast, or interim-analysis plan. General statistical concepts can clarify how the displayed estimates work, but they cannot fill those protocol details.

10. Confidence Intervals and Precision

The sponsor-reported mixed-model comparison involving all three groups gives a mean difference (net) of 6.826, with a 95% confidence interval from 1.111 to 12.541. The post hoc comparison of the combination with pramipexole reports the same estimate and interval. The interval is entirely above zero, while its width indicates that the estimate is not exact. A confidence interval is a range generated by a statistical method; it does not guarantee that the true difference falls within that particular interval.

For the post hoc comparison of the combination with escitalopram, the sponsor reports an estimate of 3.353 and a 95% confidence interval from -2.360 to 9.066. For escitalopram compared with pramipexole, the estimate is 3.473 and the 95% confidence interval runs from -1.942 to 8.888. Both intervals include zero. They therefore leave room for no group difference under the reported analysis, as well as for differences on either side of zero. Their width also shows that the direction and magnitude are not precisely determined by these estimates alone.

The simple recalculations from the arm summaries give a different inferential picture from the sponsor's reported analysis, including differences in intervals and p-values. Because the discrepancy is flagged, the sponsor's own analysis takes precedence. The posted arm standard errors and the small group sizes also warrant caution in treating the means as highly precise. The registry record does not report a clinically important difference threshold for MADRS, so statistical precision cannot by itself establish whether a difference would be meaningful to patients.

11. 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 an observed effect, rather than the original trial's sample-size justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set as a stated fraction of the observed effect. The original design assumptions are not reported in the registry record.

The 80% power column gives the illustrative participants per arm for a design with an 80% chance of detecting the specified effect if that effect were the true one and the assumptions held. The 90% power column gives the corresponding planning figure for a 90% chance. These are design probabilities under assumptions, not probabilities that the result is true or that a treatment will have a particular effect.

For the comparison of escitalopram with pramipexole, planning for 100% of the observed effect gives 39 participants per arm for 80% power and 52 participants per arm for 90% power. Planning for 50% of the observed effect gives 154 participants per arm for 80% power and 206 participants per arm for 90% power. For the combination compared with pramipexole, the corresponding figures for 100% of the observed effect are 12 and 16 participants per arm.

The general pattern is that planning to detect a smaller effect requires more participants than planning to detect a larger effect, all else equal. These figures are not observed power: they are planning illustrations, not a retrospective measure of whether the completed study had enough power.

12. 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 = 3.473; pooled standard deviation σ = 5.435.
2. n = 2 × (1.96 + 0.842)² × 5.435² / 3.473² = 38.4, rounded up to 39 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 <- 3.473                      # observed difference in means
sigma <- 5.43536884672972                     # 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(86307)
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 = 39 (80% power), 52 (90% power)
 75% of observed effect: n per arm = 69 (80% power), 92 (90% power)
 50% of observed effect: n per arm = 154 (80% power), 206 (90% power)
power.t.test: 39.4 
Simulated power at n = 39 per arm: 0.798 (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.

13. Analyses Reported by the Sponsor

The simple recalculations use posted group summaries to estimate unadjusted mean differences. The registry separately reports the sponsor's mixed-model analysis and post hoc simple effects tests. These analyses do not match in all inferential details: for example, the unadjusted recalculation for the combination compared with pramipexole has p = 0.003, whereas the sponsor's post hoc comparison reports p = .014 and a 95% confidence interval from 1.111 to 12.541. For escitalopram compared with pramipexole, the unadjusted recalculation has p = 0.103, while the sponsor reports p = .349 and a 95% confidence interval from -1.942 to 8.888.

The recalculations are not adjusted for the trial's repeated measurements or the model-based analysis. They also rely on an approximate pooled standard deviation derived from the posted standard errors. The registry record flags a discrepancy, so the sponsor's own analysis takes precedence when describing the trial's reported results. The record does not give enough model detail to determine exactly why the results differ or to reproduce the sponsor analysis independently.

14. Multiplicity, Missing Data and Analysis Populations

The registry reports one primary outcome measure and four analyses. The sponsor results include a mixed-model analysis involving the three groups and post hoc simple effects tests for group pairs. A sequence of overall and pairwise tests raises a multiplicity question: in general, making several statistical comparisons can increase the chance of a false positive if no adjustment or prespecified testing strategy is used.

Whether the sponsor adjusted the post hoc comparisons for multiplicity is not reported in the registry record. The record also does not state the prespecified primary contrast, the hierarchy of hypotheses, or whether the reported pairwise tests were planned in advance. The word “post hoc” identifies the analyses as follow-up tests but does not reveal whether a correction was applied. The p-values should therefore be read as the values reported for their respective analyses, without assuming that they account for all comparisons.

The stated analysis population included all patients with at least one post-baseline measurement. The registry does not report how many measurements were missing, why they were missing, or how missing measurements were handled in the mixed model. It also does not report sensitivity analyses under alternative assumptions about missing data. In general, repeated-measures methods can use available observations, but their validity depends on the model and assumptions about missingness; the record does not give enough detail to assess those assumptions.

The posted record lists no secondary outcomes. It does not report an interim-analysis plan, stopping rules, or the full statistical analysis plan. These omissions limit assessment of how closely the reported tests followed an original prespecified strategy. They do not establish that the analysis was inappropriate; they mean that the relevant details are unavailable in the registry record.

15. 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 5% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event figures report zero affected participants in each of the three groups: pramipexole, escitalopram, and escitalopram with pramipexole. Each group lists 13 participants at risk. The registry lists no serious adverse event terms for this outcome.

A count of zero in these groups means that no serious adverse events were reported in the posted results for the participants at risk. It does not establish that serious events cannot occur, nor does it establish that the treatments have equivalent safety. The group sizes are limited, and the record does not provide a detailed account of event definitions, follow-up completeness, or the duration of safety observation beyond the study description.

The registry reports other adverse events affecting 8 participants in the pramipexole group, 12 in the escitalopram group, and 11 in the combination group. These are listed separately from serious adverse events and should not be conflated with the serious-event counts. Deaths are not reported in the registry record.

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

The posted results describe MADRS group estimates and sponsor-reported comparisons for a randomized, three-group phase 2 study. The reported arm means were 22.63 for pramipexole, 26.1 for escitalopram, and 29.45 for the combination. On this scale, higher values indicate greater depression severity. The sponsor's post hoc comparison between the combination and pramipexole reports a mean difference (net) of 6.826, with a 95% confidence interval from 1.111 to 12.541.

That estimate is evidence about a measured group contrast under the registry's reported analysis; it is not a description of every participant and does not establish the clinical importance of the difference. The comparison of the combination with escitalopram has a confidence interval from -2.360 to 9.066, while the comparison of escitalopram with pramipexole has an interval from -1.942 to 8.888. Those intervals include zero and leave substantial uncertainty about the size and direction of those contrasts.

The results do not by themselves establish that any regimen works or does not work. The record does not report a clinically important MADRS difference threshold, and it does not provide secondary outcomes. The discrepancy between simple recalculations and sponsor-reported analyses means the sponsor's own analysis takes precedence when describing the trial findings. The available summary cannot resolve why the analyses differ.

The findings apply most directly to the selected population described in the registry: adults with major depressive disorder without psychotic features and a current episode after an inadequate response to SSRI treatment, subject to the listed eligibility criteria. They should not be assumed to describe other populations, longer-term outcomes, or outcomes not recorded in the posted results.

17. Why This Trial Matters Statistically

This study illustrates why a clinical-trial result is more than a comparison of group averages. The outcome is a repeated symptom rating, the sponsor reports a mixed-model analysis, and the record also permits simple calculations from arm summaries. Those approaches can produce related but nonidentical estimates because they rely on different information and assumptions. Recognizing the distinction prevents a recalculated p-value from being mistaken for the trial's own reported result.

The three-arm design also makes the choice of comparison important. The sponsor reports an analysis involving all three groups and post hoc tests between pairs. A reader needs to distinguish the overall model result from a specific pairwise estimate and consider whether multiplicity was addressed. That adjustment is not reported in the registry record.

Precision is central to interpretation. The sponsor's intervals are relatively broad for the pairwise comparisons, and two of them include zero. The posted sample-size illustrations show how planning needs rise when the target effect is smaller, but those illustrations use the observed effects and do not describe the trial's original design rationale. The record does not report that rationale.

Finally, a zero count of serious adverse events in each group is a descriptive result, not proof of equal safety or absence of risk. Together, these features make the study a useful example of reading continuous outcomes, model-based analyses, uncertainty, multiple comparisons, and sparse safety data without overstating what a small phase 2 record can establish.

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 lists enrollment as 39, with 13 participants in each posted arm summary. These limited group sizes mean estimates may be imprecise, as reflected in the reported confidence intervals. The study is labeled phase 2, and its results should not be treated as a definitive assessment across broader populations or longer-term outcomes.

The eligibility description is selective: candidates were adults with major depressive disorder without psychotic features, a current episode of at least 4 weeks' duration, and a history of failed SSRI response, with a limit on previous antidepressant failures for the current episode. This limits how readily the findings can be generalized to people with different diagnoses, episode histories, treatment histories, or demographic characteristics.

The registry describes a 9-week study, a 1-week drug-free period, and a 6-week treatment period. The primary outcome timeframe is weekly, but the record does not report the exact schedule of endpoint assessments used in the mixed model. It also does not provide the model specification, full analysis plan, missing-data methods, or multiplicity adjustment. These omissions limit independent evaluation and reproduction of the sponsor analysis.

The primary outcome is a depression severity rating, and no secondary outcomes are posted. The registry does not report a threshold for clinical importance on MADRS, detailed results for individual symptoms, or longer-term follow-up results. The sponsor's analyses and simple recalculations also differ, and the record does not explain the discrepancy. The sponsor analysis takes precedence, but the missing analytical detail remains a limitation when interpreting precision and robustness.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Lexapro and Pramipexole and to Treat Major Depression — NCT00086307.
• Registry record last updated 2013-01-28; status last verified 2013-01.
• Results first posted 2012-07-12.
• 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

NCT00086307 is a completed randomized, quadruple-masked, parallel-group phase 2 study sponsored by the National Institute of Mental Health. The registry describes a comparison of pramipexole, escitalopram, and the combination in patients with major depression. Enrollment is listed as 39.

The primary outcome was weekly MADRS score, analyzed in patients with at least one post-baseline measurement. The sponsor reported a mixed-model analysis and post hoc comparisons; simple unadjusted recalculations from posted arm summaries differ in some inferential details. Because a discrepancy is flagged, the sponsor's own analysis takes precedence.

The registry reports zero serious adverse events in each group and no secondary outcomes. Details including the original sample-size justification, multiplicity adjustment, and missing-data methods are not reported in the registry record.