Pilot Study to Evaluate Escitalopram in Obsessive-compulsive Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pilot Study to Evaluate Escitalopram in Obsessive-compulsive Disorder (NCT00215137), a randomized phase 2 study. The primary outcome is Yale Brown Obsessive Compulsive Scale.

2. Clinical Question

This distinction matters when interpreting the research question. The randomized comparison was not described as enrolling all 14 participants from the outset. Thirteen completed the open-label phase, and 12 of those participants showed at least minimal improvement, defined in the record as CGI-I < 3, and agreed to enter the randomized, double-blind phase. Those participants were assigned to escitalopram or placebo. The record does not report results that would establish how the groups differed on the Yale Brown Obsessive Compulsive Scale, nor does it report a treatment-effect estimate or its precision. The clinical question can therefore be described from the design, but the posted results do not answer it with a reported comparison.

3. Trial Design

The registry classifies this as a phase 2, randomized, triple-masked, parallel-design study. Its arm descriptions give a more detailed staged account: all 14 enrolled patients entered an open-label escitalopram phase, and 13 completed that phase. One patient was terminated early due to side effects. Of the 13 completers, 12 demonstrated at least minimal improvement and agreed to continue into the randomized phase. These 12 were randomized to escitalopram (n=5) or placebo (n=7).

The distinction between the open-label and randomized phases is important. The first phase exposed participants to escitalopram without a randomized placebo group. The subsequent randomized comparison was restricted to those who completed that phase, met the stated improvement criterion, and agreed to continue. It therefore addresses a selected group of participants rather than all people who began the study. The record does not report why the remaining participant who completed the open-label phase did not enter randomization.

The registry’s masking classification is triple, while the arm description calls the randomized phase double-blind. Which people were masked, and how masking was implemented, are not reported in the registry record. The design is identified as parallel, meaning the randomized groups received their assigned treatments alongside one another rather than crossing over between them. The record lists escitalopram and placebo interventions.

The study is marked completed. Its start is listed as 2004-10 and its primary completion as 2009-01. The registry lists Duke University as sponsor. Further details about the protocol, recruitment procedures, treatment adherence, or the reason for the interval between the listed dates 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.

Open Label Escitalopram

Reasons for not completing this period:

Randomization Period

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.

Population: Only the open label treatment phase is being reported per reviewer's instructions. The other arms are for the second phase of the study.

6. Primary Endpoint

The registry lists the primary outcome as Yale Brown Obsessive Compulsive Scale, measured over Open Label Phase Baseline,Randomization Phase Baseline or Beginning.

The primary measure is the Yale Brown Obsessive Compulsive Scale, abbreviated YBOCS. The registry describes it as a clinician-administered measure of OCD severity. Scores can range from a minimum of zero to a maximum of forty, and higher scores indicate more severe OCD symptoms. A continuous scale of this kind can represent symptom severity in a graded way, rather than reducing participants to only broad categories such as improved or not improved.

The measure is relevant to the stated study purpose because the study concerns OCD symptoms and the YBOCS assesses their severity. In general, a symptom scale can help quantify changes over time and differences between groups, provided the relevant scores and analysis are reported. A mean describes the group’s average score, while a standard deviation describes how spread out individual scores are around that average. Neither summary alone shows the full distribution of scores.

The listed timeframe refers to baseline in the open-label phase and baseline or beginning of the randomization phase. These are distinct points in the study sequence. The record does not report YBOCS means, standard deviations, changes from baseline, between-group contrasts, confidence intervals, or p-values. It also does not reduce the primary outcome’s several categories or time points to one comparison. Consequently, the endpoint and its measurement properties can be described, but the posted results do not permit a numerical account of symptom change or a randomized treatment contrast.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary measure as a continuous outcome summarized by a mean and standard deviation. It also describes a randomized parallel comparison after an open-label phase. Those features help explain what statistical summaries could mean, but the analysis methods used for the primary outcome are not reported in the registry record. General statistical explanations below should not be read as a claim that a particular method was used in this study.

What does a mean describe?

In general, a mean is the arithmetic average of the observed values in a group. For a symptom scale, it gives a compact description of the group’s average recorded severity at a specified time. The mean does not show how individual participants scored, whether values cluster tightly, or whether a small number of unusually high or low scores influence the average. The registry lists the YBOCS as a mean outcome, but does not report the actual means at the stated time points.

What does a standard deviation add?

In general, a standard deviation describes the spread of individual observations around a group mean. A smaller standard deviation indicates that observations are more concentrated around the mean; a larger one indicates greater dispersion. It is not the same as a confidence interval: a standard deviation describes variation among observations, whereas a confidence interval describes uncertainty in an estimated quantity. The record identifies standard deviation as the dispersion measure, but reports no YBOCS standard deviations.

Why does randomization matter?

In general, random assignment is intended to make treatment groups comparable on average with respect to measured and unmeasured characteristics. It does not guarantee that two small groups will be balanced in every respect. The randomized phase assigned 12 participants, with 5 allocated to escitalopram and 7 to placebo. Randomization supports interpretation of a between-group comparison within the randomized population, but it does not remove the selection that occurred before randomization. Participants first had to complete the open-label phase, meet the stated improvement criterion, and agree to continue.

How would a treatment contrast be interpreted?

In general, a treatment contrast compares outcome values between assigned groups at a defined time or compares changes from a defined baseline. The interpretation depends on the precise outcome definition, time point, and analysis population. A difference between group means would describe the contrast on the scale’s units; a confidence interval would show the range of values compatible with the data under the model and method used. A p-value, if reported, would summarize how compatible the observed data are with a specified null model. None of these outcome statistics is reported for the YBOCS in the registry record, so a contrast, its uncertainty, and its statistical significance cannot be described.

Why can small groups yield uncertain estimates?

In general, estimates based on few participants are often sensitive to individual observations, and their uncertainty can be substantial. A mean may be calculated for a group even when the group is small, but the value alone does not establish precision. The number of participants in each randomized group is therefore relevant context, not a substitute for a confidence interval or other uncertainty measure. Since the registry does not report outcome estimates or uncertainty intervals, the degree of precision cannot be assessed from the posted results.

What does a planning illustration represent?

In general, sample-size planning asks how many participants might be needed under chosen assumptions to detect a target effect with a specified operating characteristic. The registry’s stated assumptions describe a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The target effect is defined as a stated fraction of the observed effect. These are planning assumptions, not a description of the analysis actually conducted. A planning illustration based on an observed effect also depends on the reliability of that effect estimate. It should not be interpreted as observed power or as evidence that the trial demonstrated the target effect. The numerical planning estimates and the observed YBOCS effect are not reported in the registry record.

Why can over-dispersion matter in general?

For outcomes analyzed with models that assume a particular relationship between the mean and variance, over-dispersion means that the observed variability exceeds what the model expects. Ignoring it can make uncertainty estimates too small and statistical evidence appear stronger than warranted. This is a general consideration for suitable outcome models; the YBOCS is described as a continuous measure summarized by a mean and standard deviation, and the registry does not report a model or an over-dispersion assessment. It would therefore be inappropriate to claim that over-dispersion was present or that any particular adjustment was used.

The registry records two analyses, but their definitions, models, covariates, and relationship to the outcome time points are not reported. The statistical plan, including any prespecified primary contrast or approach to repeated measurements, is also not reported in the registry record. As a result, methodological interpretation must remain limited to the design and summaries that are explicitly described.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies one primary outcome measure, the YBOCS, but the outcome entry refers to several categories or time points and is not reduced to one comparison. It lists baseline in the open-label phase and baseline or beginning of the randomization phase. The record also notes two analyses, without describing what distinguishes them. The exact outcome categories, analysis time points, contrasts, or hierarchy among analyses are not reported in the registry record.

Multiplicity arises in general when several outcomes, time points, or statistical comparisons are examined. The more comparisons considered, the greater the chance that at least one may appear statistically notable by chance if no adjustment or prespecified interpretive strategy is used. How this issue should be handled depends on the analysis plan and the purpose of the comparisons. The statistical plan and any multiplicity procedure are not reported, so it is not possible to determine how the study addressed this issue.

Missing-data information is also limited. The arm descriptions report that 13 of 14 patients completed the open-label phase and that one patient was terminated early due to side effects. They report that 12 of the 13 completers entered the randomized phase. Reasons the other completer did not proceed, outcome availability for each participant, and the handling of missing observations are not reported in the registry record.

The record does not identify the analysis population for the YBOCS, such as whether all randomized participants were included or whether another rule was used. It does not describe imputation, exclusions, or sensitivity analyses. These omissions limit assessment of whether the reported analysis, if results were available, represented the randomized groups as assigned or depended on the availability of observed measurements.

11. 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 1% reporting threshold.

The safety counts are listed separately for the open-label escitalopram group, the randomized placebo group, and the randomized escitalopram group. The registry records 0 participants affected by serious adverse events among 14 at risk in the open-label group, among 7 at risk in the placebo group, and among 5 at risk in the randomized escitalopram group. It also records 0 participants affected by other adverse events in each of those groups, with the same respective numbers at risk.

These entries describe what was recorded for the listed groups; they do not establish that no adverse events of any kind occurred outside the categories or reporting rules used. The record does not give event-level descriptions, severity grading beyond the serious/non-serious grouping, timing, duration, or methods of safety ascertainment. It also does not report the death counts. A null death entry should not be interpreted as a reported count of zero.

The open-label and randomized groups represent different stages and participant sets, so their counts should not be treated as parallel randomized comparisons. The randomized groups were small, and zero recorded affected participants is a descriptive result rather than proof that risk is absent. These figures do not by themselves establish a causal safety difference between escitalopram and placebo.

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

The record documents a pilot study with an open-label escitalopram phase followed by randomization among participants who completed that phase, showed at least minimal improvement, and agreed to continue. It records that 14 patients entered the open-label phase, 13 completed it, and 12 entered the randomized phase. The randomized allocation was 5 to escitalopram and 7 to placebo. These registry entries describe recruitment, progression, and assignment; they are not evidence of a particular symptom outcome.

The posted results do not report numerical YBOCS findings. Without group means, changes, a between-group estimate, or uncertainty measures, the magnitude and direction of symptom change cannot be stated. The results also do not permit a judgment about how precisely a randomized effect was estimated. The study’s stated aim was to evaluate safety and effectiveness, but an aim is not itself a result.

The design means that the randomized comparison applies to a selected group entering the second phase, not necessarily to all patients who might start escitalopram or to every person enrolled in the open-label phase. A randomized comparison can help distinguish treatment-group outcomes within the population randomized, but only when the relevant results and analysis details are available. The registry record does not report enough information to assess the primary outcome comparison.

The recorded safety counts show no participants affected by serious or other adverse events in the listed groups. That descriptive record should not be translated into a claim that adverse events are impossible or that one treatment is safer. The death figures are not reported. In general, safety interpretation also depends on how events were defined, collected, and monitored, details that are not reported.

Accordingly, the study record supports conclusions about the trial’s structure, eligibility for randomization as described, and the safety counts entered in the registry. It does not support a quantitative conclusion about YBOCS efficacy, a precise comparison between randomized groups, or broad conclusions beyond the selected participants and study procedures described.

13. Why This Trial Matters Statistically

This trial is statistically interesting because its randomized comparison followed an open-label phase rather than beginning with random assignment for all enrolled patients. Such a staged design can address a focused question about participants who continue after an initial treatment period, but it creates a selected randomized population. The distinction between all enrolled participants and those randomized is essential to defining the estimand: the treatment effect that a study is designed to estimate.

The study also illustrates why outcome reporting is necessary for statistical interpretation. A named scale, group sizes, and a randomized design do not reveal the size or precision of an outcome contrast. To interpret the YBOCS comparison, readers would need the outcome values or changes, the analysis time point, an estimate of the between-group contrast, and an uncertainty measure. Those results are not reported in the registry record.

The study’s small randomized groups make uncertainty especially important as a general concern. A point estimate from a small sample can be unstable, and the observed data may be compatible with a broad range of effects. Without a reported estimate and confidence interval, there is no basis for quantifying that uncertainty. Planning illustrations cannot replace trial outcome estimates, and a sample-size calculation is not evidence that a treatment effect was observed.

Finally, the record distinguishes a registry classification of triple masking from an arm description of a double-blind randomized phase. Noting such differences helps readers avoid assuming details about who was masked or how masking operated. The record is valuable for clarifying the study’s staged structure and limits, while its missing outcome and analysis details constrain what can be concluded statistically.

14. 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 main trial-specific limitation in the available record is the absence of numerical results for the primary YBOCS outcome. The scale and its scoring direction are described, but no means, standard deviations, changes, treatment contrast, confidence interval, or p-value is reported. The primary outcome also refers to multiple categories or time points without one comparison being identified, preventing a focused account of the primary result.

The randomized sample consisted of 12 participants, with 5 assigned to escitalopram and 7 to placebo. Those participants had completed the open-label phase, shown at least minimal improvement, and agreed to continue. This staged selection limits the population to which the randomized comparison directly pertains. The record does not report the characteristics or outcomes of participants who did not proceed to randomization, beyond noting that one open-label completer did not enter the randomized phase.

Important analysis details are absent. The registry does not report the statistical model, analysis population, handling of missing measurements, prespecified contrast, or multiplicity strategy. It records two analyses but does not describe them. The masking classification and arm description use different terms, and details about masking roles are not reported. The record also gives limited safety detail beyond group counts, with no event descriptions or death counts.

These limitations do not establish that the trial was conducted improperly; they describe what cannot be evaluated from the registry record. The missing outcome summaries and analysis information make it impossible to assess the magnitude, precision, or robustness of the primary result from the posted record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Pilot Study to Evaluate Escitalopram in Obsessive-compulsive Disorder — NCT00215137.
• Registry record last updated 2014-07-23; status last verified 2013-08.
• Results first posted 2013-10-29.
• 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.

18. Record Summary

The registry describes a completed phase 2 randomized parallel study of escitalopram for OCD, with an open-label phase followed by a randomized phase. Fourteen patients entered the open-label phase, 13 completed it, and 12 who showed at least minimal improvement and agreed to continue were randomized to escitalopram (n=5) or placebo (n=7). The primary measure was the YBOCS, a clinician-administered severity scale on which higher scores indicate more severe symptoms.

The record lists zero participants affected by serious adverse events and zero affected by other adverse events in each group; death counts are not reported. Numerical YBOCS results, analysis details, and the statistical plan are not reported in the registry record. The study’s design and participant flow can be described, but its primary randomized symptom comparison cannot be quantified from the posted results.