Drug Therapy for Generalized Anxiety Disorder Among the Elderly: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Drug Therapy for Generalized Anxiety Disorder Among the Elderly (NCT00105586), a randomized phase 4 study. The primary outcome is Response Using Clinical Global Impressions-Improvement Scale (CGI-I).

2. Clinical Question

The study summary describes two interests: determining the efficacy of escitalopram for generalized anxiety disorder and examining whether genetics affect treatment response among elderly individuals. The posted primary result addresses symptom improvement, not a reported genetic comparison. Genetic analyses or results are not reported in the registry record.

This is a question about the comparison recorded for this randomized trial. It is not, by itself, a complete account of how the trial was conducted or how all participants’ outcomes were handled. Detailed eligibility criteria, the treatment regimen, and the statistical plan are not reported in the registry record. The posted figures therefore support a focused reading of the recorded response outcome, while leaving those aspects of the clinical question’s context unspecified.

3. Trial Design

The registry describes a Phase 4 randomized, quadruple-masked, parallel trial comparing escitalopram with placebo. The record lists enrollment of 177 and the status as completed. The study start is recorded as 2004-12, and primary completion as 2008-01. The sponsor is the University of Pittsburgh.

The intervention listed is escitalopram, and the two arms are escitalopram and placebo. The primary outcome is a binary response measure: participants are counted as responders if their Clinical Global Impressions-Improvement Scale score is 1 or 2. The outcome timeframe is Weeks 1-12. The posted arm results list 51.0 responders among 75.0 for placebo, and 69.0 among 69.0 for escitalopram.

Those primary-outcome denominators should not be treated as an explanation of how the enrollment figure relates to the outcome data. The registry also lists different at-risk counts for adverse-event reporting. How these populations were defined, and how people without an available endpoint were handled, are not reported in the registry record.

Randomization means that assignment to trial arms was intended to use a chance-based process; masking means that specified trial participants or personnel were kept unaware of assignments. The record labels masking as quadruple but does not identify which groups were masked or give implementation details. A parallel model means that the arms are compared as separate treatment groups rather than as a crossover sequence. The registry record does not report the randomization procedure, allocation concealment, dose, treatment duration, adherence, or co-interventions.

Quality of Life is listed as a secondary outcome. The registry reports two outcome measures and one posted secondary outcome, but no analyses are listed. The study summary also mentions genetic influences on response, while genetic results 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: 78 excluded from randomization due to ineligibility and/or refusal; 2 excluded after randomization but did not start treatment

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 Response Using Clinical Global Impressions-Improvement Scale (CGI-I), measured over Measured at Weeks 1-12.

The primary endpoint is response on the Clinical Global Impressions-Improvement Scale (CGI-I), recorded as a number of participants. The registry describes cumulative incident response measured at Weeks 1-12. A score of 1, “very much improved,” or 2, “much improved,” is classified as response. The record says the scores were synthesized from anxiety rating scale scores, including the Penn State Worry Questionnaire (PSWQ) and Hamilton Anxiety Scale (HamA).

This endpoint converts an improvement scale into a responder count. That can make a group comparison easier to interpret: it focuses on how many participants crossed the stated improvement threshold rather than comparing every possible scale value as a continuous score. The choice of threshold also matters. A participant just meeting the response definition and one with a score of 1 both count as responders, while other scale scores do not meet the stated definition.

The result is therefore specific to the registry’s responder definition, measurement window, and listed anxiety measures. It is not a report of every symptom change, the full distribution of CGI-I scores, or a quality-of-life result. Detailed procedures for synthesizing the scores, timing of assessments within the window, and handling of missing assessments are not reported in the registry record. The endpoint should be read as the posted binary response outcome, not as a complete description of all dimensions of the participants’ experience.

7. Results Posted in the Registry

• Escitalopram: 100.0% (69/69) versus 68.0% (51/75) on control; risk difference 32.0 percentage points (95% CI 21.4 to 42.6), risk ratio 1.47 (1.26–1.72); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the binary endpoint represent?

The primary measure is a responder count. Each participant is classified according to whether the CGI-I score met the stated response threshold: 1 for very much improved or 2 for much improved. A binary endpoint has two possible classifications for this analysis—response or no response—even though the underlying clinical scale has more than two possible values. The posted arm counts describe participants classified as responders; they do not show the full distribution of scale scores.

In general, turning a scale into a binary outcome simplifies a comparison and gives a threshold-based interpretation. It also discards information about differences within each classification. The registry’s response definition is therefore essential to understanding what the result represents. The record says the CGI-I scores were synthesized from anxiety rating scale scores, including PSWQ and HamA, but details of that synthesis are not reported.

What does a risk difference say?

The registry reports a risk difference of 0.32, comparing escitalopram with placebo. In general, a risk difference is the absolute contrast between the probabilities of the event in two groups. Here, the event is meeting the responder definition. The sign indicates direction according to the comparison as reported; a positive value means the response probability is estimated to be higher in the escitalopram arm.

The registry also reports a risk ratio of 1.471, with a confidence interval from 1.259 to 1.718. In general, a risk ratio compares the event probabilities as a relative measure. The risk difference and risk ratio offer distinct perspectives: one describes an absolute contrast and the other a relative contrast. Neither should be interpreted without its endpoint definition, observation window, and uncertainty.

How should the p-value be read?

The posted comparison reports p < 0.001. In general, a p-value describes how often data at least as incompatible with a specified null model would be expected if that model and the test assumptions held. It does not give the probability that the null hypothesis is true, the probability that the observed effect is due to chance, or the size of the effect. It also does not establish whether an outcome is important to participants.

The registry does not report the detailed test specification, whether the comparison was prespecified, or whether it was adjusted for other factors. The p-value should consequently be read as the value attached to the posted comparison, not as evidence that every possible analysis or outcome yielded the same conclusion.

What role does randomization play?

The trial is labeled randomized and parallel. In general, random assignment aims to make the treatment groups comparable at the point of assignment, including with respect to factors that may not be measured. A parallel trial compares groups assigned to their respective arms. The registry also labels masking as quadruple, but does not identify which people were masked or provide procedures for maintaining masking. The record does not describe the randomization sequence or allocation concealment.

Randomization is a design feature, not a guarantee that every later analysis is free of bias. Outcome availability, deviations from assigned treatment, and the choice of analysis population can affect interpretation. Those details are not reported in the registry record. In particular, it is not clear from the posted information how the listed outcome denominators relate to enrollment or to the adverse-event denominators.

What do the planning assumptions mean?

The registry’s sample-size assumptions describe planning illustrations based on the observed effect, rather than the trial’s original design. They specify a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect set as a stated fraction of the observed effect. In general, such calculations illustrate the number of participants per arm that might be needed under the chosen assumptions. They do not measure the power actually achieved by this trial, and they are not observed results.

The registry record does not report the original sample-size justification, the assumptions used when the trial was designed, or interim analyses. No numerical planning results are stated in the registry entries described for this analysis, so no particular planning sample size can be interpreted. The planning assumptions should not be mistaken for an analysis plan or evidence about trial conduct.

What is not specified about the statistical analysis?

No analyses are listed in the registry record. The record therefore does not give details of a statistical analysis plan, adjustment for covariates, handling of repeated measurements, subgroup testing, or missing outcome data. It does not identify an intention-to-treat, per-protocol, or other analysis population. These are not minor labels: they describe which participants and measurements enter an analysis and how the comparison is estimated.

In general, a confidence interval describes a range of estimates compatible with the data under a specified method and assumptions. Its width reflects statistical precision under that method, but not all sources of uncertainty, including measurement limitations or incomplete reporting. Here, the reported comparison and interval can be described, while the absence of a detailed analysis specification limits independent appraisal of how they were obtained.

10. Confidence Intervals and Precision

The registry reports a risk difference of 0.32 and a confidence interval for that estimate. The interval’s lower and upper bounds are both above zero. The exact bounds are not repeated because the registry gives them with more than three decimal places. For the relative comparison, the reported risk ratio is 1.471, with a confidence interval from 1.259 to 1.718.

In general, a confidence interval represents uncertainty about an estimated quantity under a particular statistical method and its assumptions. It is not a range containing a fixed percentage of individual outcomes, and it does not mean that values outside the interval are impossible. An interval that excludes the null value can be consistent with a statistically detectable difference under the corresponding testing procedure, but it does not determine whether that difference is important in practice.

The reported intervals indicate a positive direction for both stated comparisons, while still allowing uncertainty about the size of the contrast. Precision should be judged in relation to the endpoint, the trial population, and the quality of the underlying measurements. The registry record does not provide the detailed analysis methods needed to assess all assumptions behind the intervals, nor does it report the handling of missing data or the analysis population.

Confidence intervals for a binary comparison do not describe variation among individual participants or guarantee that the same contrast would appear in another trial. The primary result is also bounded by the particular CGI-I response threshold and Weeks 1-12 timeframe. The interval cannot resolve questions about longer-term outcomes, quality of life, genetic effects, or safety beyond the posted event counts.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures and identifies Quality of Life as a secondary outcome. It lists no analyses. The posted primary comparison is accompanied by a p-value, but no p-values or detailed results for the secondary outcome are reported in the registry record. Genetic influences are mentioned in the study summary, but no genetic analyses or results are reported.

Multiplicity arises when several outcomes, time points, subgroups, or statistical comparisons are examined. In general, conducting multiple tests can increase the chance that at least one result appears statistically notable even if there is no underlying effect. Statistical plans may address this through prespecified outcome hierarchies or adjustment procedures. Whether such procedures were planned or used in this trial is not reported in the registry record. The available information does not support judging whether the reported p-value was adjusted for multiple comparisons.

Missing-data handling is also not reported. The registry does not explain whether participants without an endpoint assessment were excluded, counted as nonresponders, handled using an imputation method, or addressed in another way. The posted response denominators should not be used to infer what happened to participants outside those denominators, or to reconstruct the analysis population from enrollment.

The registry record does not identify an intention-to-treat or per-protocol analysis, describe protocol deviations, or specify whether outcome assessors and participants remained masked throughout. It also does not report interim analyses or a detailed statistical analysis plan. These omissions matter because they affect the interpretation of the reported comparison and the extent to which it can be independently reproduced.

Accordingly, the primary result can be described as posted, but the role of multiplicity, missing observations, and analysis-population choices remains uncertain. Those details are not reported in the registry record, rather than evidence that a particular method was or was not used.

12. 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 non-serious adverse event terms

The registry lists zero serious adverse events in the placebo group, with 92 at risk, and zero in the escitalopram group, with 85 at risk. It lists no serious adverse-event terms. Deaths are not reported in the registry record. These are safety figures as recorded and should be distinguished from the primary response counts, which use different group denominators.

The registry also lists other adverse events in both groups: 59 among 92 at risk for placebo, and 65 among 85 at risk for escitalopram. The record does not give descriptions of those other events in the listed safety data. The category should not be interpreted as equivalent to serious adverse events, nor do the event counts identify severity, timing, duration, or whether an event was judged related to treatment.

In general, a zero count in a trial record means that no event was recorded in that category for the reported group; it does not establish that the risk is absent. Safety interpretation depends on how events were defined, collected, and followed, and on the number and characteristics of participants observed. Those procedures and the relationship between the safety groups and primary-outcome groups are not reported in the registry record. The figures are therefore a limited description of posted events, not a comprehensive safety assessment.

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

The posted data show that the registry recorded more responders in the escitalopram arm than in the placebo arm under the stated CGI-I threshold and Weeks 1-12 timeframe. The reported risk difference is 0.32, with p < 0.001. The record also reports a risk ratio of 1.471, with a confidence interval from 1.259 to 1.718. These statistics describe the comparison as recorded; they do not answer every question about the treatment or the trial.

The result does not, by itself, establish how the response comparison would apply to people outside the trial population. Detailed eligibility criteria are not reported. Nor does it establish the persistence of response beyond Weeks 1-12, because longer-term outcomes are not reported in the primary endpoint description. The endpoint is a threshold-based summary, not a full account of symptom changes or individual experiences.

The trial summary mentions an interest in genetic influences on treatment response, but genetic analyses and results are not reported. Quality of Life is listed as a secondary outcome, but its results are not reported in the registry record. The primary comparison also does not quantify adherence, treatment discontinuation, or other clinical outcomes.

The posted safety information lists zero serious adverse events in both groups, as well as counts for other adverse events. It does not provide enough detail to characterize those other events or to establish the full safety profile. Absence of a recorded serious event in these groups is not proof that serious events cannot occur.

Finally, the absence of a detailed statistical plan, analysis population, and missing-data methods limits assessment of how the estimate was produced. The findings should be understood as a specific randomized comparison with a stated endpoint and window, accompanied by uncertainty and reporting limitations. They should not be generalized beyond what the registry record reports.

14. Why This Trial Matters Statistically

This trial illustrates how a clinical outcome can be reduced to a responder definition and then compared between randomized arms. The definition is explicit: CGI-I scores of 1 or 2 count as response, measured during Weeks 1-12. That makes the primary result interpretable as a threshold-based comparison, while also making clear that participants’ degrees of improvement within the responder category are not shown in the posted counts.

The record reports both an absolute measure, the risk difference of 0.32, and a relative measure, the risk ratio of 1.471. Looking at both helps distinguish two common ways of expressing a binary outcome comparison. Their interpretation depends on the endpoint and denominator, and neither measure replaces consideration of uncertainty. The registry’s reported confidence intervals and p-value provide statistical context, but not a judgment about clinical importance.

The trial also demonstrates why design information and results need to be read together. Randomization, masking, and a parallel model are recorded, but procedural detail is limited. The registry reports enrollment and arm-specific response and safety figures, yet does not explain analysis populations or missing-data handling. Those gaps prevent a complete reconstruction of the analysis from the record alone.

Finally, a study summary may describe broader aims than the results posted for a particular outcome. Here, the summary mentions genetic influences, while the record lists no analyses and no genetic results. Quality of Life appears as a secondary outcome but has no posted result in this record. Separating the stated research interests from the reported estimates is central to sound statistical reading.

15. 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.

Several trial-specific limitations follow from the available registry record. The population is described as elderly individuals with generalized anxiety disorder, but detailed eligibility criteria and participant characteristics are not reported. It is therefore not possible to assess how closely the enrolled participants represent any particular clinical setting or subgroup.

The endpoint is a binary response classification based on CGI-I scores of 1 or 2, with scores synthesized from PSWQ and HamA. The record does not give the synthesis procedure, the precise assessment schedule within Weeks 1-12, or results for the underlying scales. The responder count therefore cannot show the full pattern of symptom change or how close participants were to the threshold.

The treatment regimen, dose, duration, adherence, and co-interventions are not reported in the registry record. The record labels the trial randomized and quadruple masked but does not describe implementation or identify the masked parties. The analysis population, missing-data method, protocol deviations, statistical model, and adjustment procedures are also not reported. No detailed statistical plan or interim analyses are listed.

Although Quality of Life is listed as a secondary outcome and genetic influences are part of the study summary, results for these topics are not reported. The safety information provides counts but not event descriptions for the other adverse-event category, and deaths are not reported in the registry record. Safety follow-up and event-collection procedures are not described.

These limitations do not negate the posted primary comparison, but they constrain its interpretation and independent appraisal. The result should be kept specific to the outcome, timeframe, and registry information available.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Drug Therapy for Generalized Anxiety Disorder Among the Elderly — NCT00105586.
• Registry record last updated 2017-05-08; status last verified 2017-05.
• Results first posted 2017-05-08.
• 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.

19. Record Summary

The registry record describes a completed Phase 4 randomized, quadruple-masked, parallel trial sponsored by the University of Pittsburgh. Enrollment is listed as 177. The study start is 2004-12 and primary completion is 2008-01.

The primary outcome is CGI-I response during Weeks 1-12, with scores of 1 or 2 classified as response. The record lists 51.0 responders among 75.0 for placebo and 69.0 among 69.0 for escitalopram. The reported risk difference is 0.32, with p < 0.001; the reported risk ratio is 1.471, with a confidence interval from 1.259 to 1.718.

Quality of Life is listed as a secondary outcome. The record lists no analyses and does not report genetic results, missing-data methods, or a detailed statistical plan. Safety figures list zero serious adverse events in each group, with other adverse-event counts also recorded.