Study of Antidepressants in Parkinson's Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study of Antidepressants in Parkinson's Disease (NCT00086190), a randomized phase 3 study. The primary outcome is Change in Hamilton Depression Rating Scale (HAM-D) Scores.

2. Clinical Question

3. Trial Design

The registry describes a completed phase 3, randomized, quadruple-masked, parallel-group study sponsored by the University of Rochester. The recorded start is 2005-06 and primary completion is 2009-11. The planned comparison involved three groups: paroxetine, venlafaxine extended release, and placebo. The registry describes the treatment comparison as lasting 12 weeks during a double-blind phase. It records 115 participants randomized overall.

The primary outcome population is described as all randomized participants, analyzed according to the intention-to-treat principle. The posted primary outcome groups contain 42 participants assigned to paroxetine, 34 assigned to venlafaxine extended release, and 39 assigned to placebo. These are the group sizes attached to the reported HAM-D summaries; details about any participant-level exclusions, missing measurements, or deviations from assigned treatment are not reported in the registry record.

The primary endpoint was change in Hamilton Depression Rating Scale scores between 0 weeks and 12 weeks. The posted group results are means with standard deviations. The registry also lists secondary measures covering depression, anxiety, sleep, psychiatric symptoms, and Parkinson’s disease symptoms. Numerical results are available for some of those secondary outcomes, but analyses for the secondary outcomes are not listed as reported analyses in the registry record.

Allocation concealment, the specific randomization procedure, medication dosing, and the original statistical analysis plan are not reported in the registry record. Nor are details of masking procedures or the identities of the parties included in quadruple masking described there.

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: 115 subjects were randomized to receive either Paroxetine, Venlafaxine ER or placebo. All randomized participants were included in analysis, in accordance to intention-to-treat principle.

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 Change in Hamilton Depression Rating Scale (HAM-D) Scores, measured over from the beginning (0 weeks) to end (12 weeks) of the double-blind phase.

The primary endpoint is change in HAM-D score from the beginning of the double-blind phase at 0 weeks to its end at 12 weeks. HAM-D is a depression-rating scale. The registry states that scores range from 0-50 and that higher scores represent more significant depression. It also gives score ranges for mild, moderate, severe, and very severe depression. The posted outcome is a mean change for each assigned group, rather than the number of participants who crossed a threshold or met a definition of response.

Change scores are used to describe how the measured symptom score moved during the study period. A negative change represents a lower score at the end than at the beginning; given the registry’s description of the scale, that direction is a reduction in the measured depression score. The primary group means were -13.0 for paroxetine, -11.0 for venlafaxine extended release, and -6.8 for placebo.

Comparing each medication group’s mean change with the placebo group’s mean change gives an estimate of the between-group difference in change. This comparison is useful because improvement over time in the placebo group is also part of the observed study course. The registry does not report a threshold for a clinically important HAM-D change, or a separate rule for defining an individual participant as improved. The mean results therefore describe group-level score changes and should not be read as individual response rates.

7. Results Posted in the Registry

• Paroxetine: mean -13.00 versus -6.80 on control; difference -6.20 (95% CI -6.77 to -5.63), standardized effect d = -4.77; z-test p = <0.001.
• Venlafaxine Extended Release: mean -11.00 versus -6.80 on control; difference -4.20 (95% CI -4.82 to -3.58), standardized effect d = -3.12; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in Montgomery-Asberg Depression Rating Scale (MADRS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Paroxetine: mean -13.60 versus -6.60 on control; difference -7.00 (95% CI -10.33 to -3.67), standardized effect d = -0.92; z-test p = <0.001.
  • Venlafaxine Extended Release: mean -10.90 versus -6.60 on control; difference -4.30 (95% CI -7.77 to -0.83), standardized effect d = -0.57; z-test p = 0.015.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paroxetine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Venlafaxine Extended Release than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Beck Depression Inventory II (BDI-II)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Paroxetine: mean -9.70 versus -5.20 on control; difference -4.50 (95% CI -7.55 to -1.45), standardized effect d = -0.64; z-test p = 0.004.
  • Venlafaxine Extended Release: mean -9.60 versus -5.20 on control; difference -4.40 (95% CI -7.59 to -1.21), standardized effect d = -0.64; z-test p = 0.007.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paroxetine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Venlafaxine Extended Release than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Geriatric Depression Rating Scale (GDS)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Paroxetine: mean -6.90 versus -2.80 on control; difference -4.10 (95% CI -6.87 to -1.33), standardized effect d = -0.64; z-test p = 0.004.
  • Venlafaxine Extended Release: mean -6.90 versus -2.80 on control; difference -4.10 (95% CI -7.01 to -1.19), standardized effect d = -0.65; z-test p = 0.006.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Paroxetine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Venlafaxine Extended Release than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The numerical secondary results available in the registry cover MADRS, BDI-II, and GDS changes. Across these measures, mean changes were negative in the placebo group as well as in both medication groups, and the medication-group means were more negative than the placebo mean. This broad pattern is directionally consistent with the primary HAM-D comparison. It suggests that the pattern is not confined to the primary scale among the secondary numerical results posted.

That consistency is supporting context, not independent confirmation by itself. The measures assess related aspects of depression, so their results may be correlated rather than representing wholly separate evidence. The registry lists other secondary measures, including measures of psychiatric symptoms, Parkinson’s symptoms, anxiety, and sleep, but does not include numerical summaries for all those outcomes in the registry record for this account. The record lists 17 outcome measures and 16 secondary outcomes posted; it does not provide a complete set of secondary analyses.

The displayed secondary comparisons and p values are unadjusted recalculations from summary statistics. The registry does not report a multiplicity plan, a hierarchy of secondary endpoints, or analyses identifying which outcomes were intended as confirmatory. When many outcomes are tested, some may appear noteworthy by chance even if there is no underlying effect for those outcomes. Without knowing the prespecified testing strategy and any adjustment, individual secondary p values deserve less weight than a result from a clearly designated primary analysis.

The secondary outcomes can therefore be read as measures that offer additional descriptive context around the primary result. Their broadly similar direction is relevant, but the absence of a reported analysis plan, incomplete numerical outcome reporting, and lack of a stated multiplicity procedure limit how strongly the secondary findings can support a firm conclusion. They should not be treated as a substitute for the primary comparison or as proof that every measured domain changed in the same way.

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

• Change in Montgomery-Asberg Depression Rating Scale (MADRS)
• Change in Beck Depression Inventory II (BDI-II)
• Change in Geriatric Depression Rating Scale (GDS)
• Change in Brief Psychiatric Rating Scale (BPRS)
• Change in Unified Parkinson's Disease Rating Scale (UPDRS)
• Change in Snaith Clinical Anxiety Scale (CAS)
• Change in Pittsburgh Sleep Quality Index (PSQI)
• Change in Unified Parkinson's Disease Rating Scale (UPDRS) - Motor

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant’s HAM-D score can take values across a scale, and the registry reports group means and standard deviations. A mean describes the group’s average change. A standard deviation describes how dispersed individual changes are around that average. These summaries are different: a mean estimates the group’s central change, while a standard deviation characterizes variation among participants. The registry states that all randomized participants were included in the primary analysis according to an intention-to-treat principle, but does not explain how missing measurements were handled.

The medication-versus-placebo contrasts are differences between group mean changes. The posted paroxetine estimate is -6.2, and the posted venlafaxine extended release estimate is -4.2. Both use placebo as the reference group. A negative contrast means the medication group’s mean change is more negative than the placebo group’s mean change. This direction is interpretable alongside the scale description: higher HAM-D scores represent more significant depression, so a negative change reflects a lower score at follow-up. The contrasts describe average differences, not the experience of each participant.

The numerical contrasts, intervals, and unadjusted p values shown in this analysis are recalculated from the posted summary statistics. They are simple, unadjusted comparisons and are not identified as the study sponsor’s analysis in the registry record. The registry has no reported analyses attached to the primary outcome. The sponsor’s inferential method, any covariate adjustment, and the original prespecified model are not reported in the registry record. It is therefore important not to present the recalculated results as though they reproduce a sponsor analysis.

A confidence interval is a range of values compatible with the observed estimate under a specified statistical procedure. Narrower intervals indicate greater precision under that procedure; wider intervals indicate more uncertainty about the value of the contrast. The intervals for both primary comparisons lie below zero. That observation describes the posted intervals, but the confidence level and the procedure used to form the registry’s original inferential results are not reported. An interval does not capture every source of uncertainty, such as possible bias, outcome selection, or limitations in how the study was conducted.

A p value evaluates how unusual results at least as discrepant as the observed result would be under a statistical null model, given the test’s assumptions. It is not the probability that the null hypothesis is true, and it is not a measure of effect size. The primary recalculated p values are unadjusted. Since the registry lists several outcomes, interpreting individual p values without accounting for multiple comparisons can overstate how compelling a collection of results is. The registry does not report whether the original study used a multiplicity adjustment or which outcomes were designated in advance for confirmatory testing.

The secondary outcomes use the same general idea of comparing mean change between an assigned medication group and placebo, but the registry reports their group spreads as standard errors rather than standard deviations. A standard error describes the precision of an estimated group mean, whereas a standard deviation describes participant-level variability. They are not interchangeable descriptions of spread. The registry does not list reported inferential analyses for the secondary outcomes, so the displayed contrasts and p values are unadjusted recalculations from summary information, not documented sponsor analyses.

In general, over-dispersion matters when a statistical model assumes less variability than is actually present: standard errors can then be too small, intervals too narrow, and p values too optimistic. The registry does not report a count-data endpoint or a model for over-dispersed counts for the primary result, which is a continuous score change. No claim about an over-dispersion adjustment can be made for this analysis.

Randomization is intended to create groups that can be compared without systematic assignment according to participant characteristics. Masking can reduce the influence of treatment knowledge on participants’ reports, care, or outcome assessment. The registry identifies the study as randomized and quadruple-masked, but does not detail the randomization implementation, masking roles, or whether masking was maintained. These design labels alone cannot establish that every possible source of bias was prevented.

The primary endpoint’s intention-to-treat description is a meaningful part of the analysis record, but it is not a complete account of missing-data methods. The registry does not report the number of missing outcome measurements, how missing observations were treated, or whether sensitivity analyses were performed. It also does not report interim analyses, a stopping rule, or the original rationale for the planned enrollment. Those methods should not be inferred from the posted results.

11. Confidence Intervals and Precision

The primary mean differences are reported with intervals: -6.2 for paroxetine compared with placebo, with endpoints -6.767 and -5.633; and -4.2 for venlafaxine extended release compared with placebo, with endpoints -4.823 and -3.577. Both intervals are entirely below zero. In the direction of the HAM-D scale, the estimates and intervals describe more negative mean change in the medication groups than in the placebo group.

An interval conveys uncertainty about an estimated contrast under the method used to construct it. A narrower interval represents greater statistical precision than a wider interval, but precision does not itself establish that a difference is clinically important or free of bias. The observed group sizes and the variability in the outcome contribute to how precisely a mean difference can be estimated. The registry does not report the confidence level or the original method used to calculate these intervals, and the intervals presented with the contrasts are recalculations from posted summary statistics.

The primary group means are accompanied by standard deviations, while the secondary outcome group means are accompanied by standard errors. These quantities describe different aspects of the data. The standard deviation characterizes participant-level spread; the standard error characterizes uncertainty in an estimated mean. Interpretations of precision should account for that distinction rather than treating all reported spreads as though they measure the same thing.

Even a precise estimate can be limited by the study’s population, duration, measurement choices, or missing-data handling. Conversely, an interval’s width does not tell whether an observed difference matters to participants. The registry does not report a clinically meaningful change threshold for HAM-D. The intervals therefore help describe uncertainty around the posted mean contrasts, but do not by themselves settle clinical importance or the generalizability of the findings.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrative calculations based on the observed effect, rather than the original trial’s design 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 at the stated fraction of the observed effect. The registry does not report the trial’s own original sample-size assumptions or justification.

The 80% power column gives the calculated number of participants per arm for a hypothetical study to have an 80% chance of detecting the effect specified for that row under the stated planning assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding planning number for a higher target chance of detection. These are design illustrations, not estimates of the probability that the completed study detected its effect.

The rows consider planning for the full observed effect, three-quarters of the observed effect, and half of the observed effect. Smaller effects are harder to distinguish from random variation, so the calculations require more participants per arm as the target effect becomes smaller. This is not observed power: the figures do not describe the probability that the completed trial had to detect its observed result. They are prospective planning scenarios calculated from the observed contrast and should not be interpreted as a feature of the original study design.

The planning calculations compare two arms at a time, whereas the trial randomized participants among three groups. They do not account for every possible design feature, analysis choice, or missing-data pattern in the study. Their value is to illustrate how assumed effect size and desired power affect planning under the listed assumptions, not to establish whether the actual enrollment was adequate for every outcome.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -6.200; pooled standard deviation σ = 1.300.
2. n = 2 × (1.96 + 0.842)² × 1.300² / -6.200² = 0.7, rounded up to 1 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 <- -6.2                      # observed difference in means
sigma <- 1.3                     # 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(86190)
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 = 1 (80% power), 1 (90% power)
 75% of observed effect: n per arm = 2 (80% power), 2 (90% power)
 50% of observed effect: n per arm = 3 (80% power), 4 (90% power)
power.t.test: 2.2 
Simulated power at n = 1 per arm: NA (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

The registry reports a primary endpoint and lists several secondary outcome measures. Testing multiple outcomes creates a multiplicity issue: as the number of comparisons grows, the chance of obtaining at least one apparently unusual result by chance may also grow. Statistical approaches can address this by adjusting significance thresholds, controlling an error rate, or specifying a hierarchy of tests. Which approach, if any, was used for the original study is not reported in the registry record.

The p values attached to the recalculated comparisons are explicitly unadjusted. They should be understood as individual comparison results, not as multiplicity-controlled evidence across the family of endpoints. The registry does not report which secondary outcomes were designated as key, whether a testing sequence was prespecified, or whether the primary outcome was the sole confirmatory endpoint. As a result, the pattern across outcomes is useful descriptively but its overall statistical strength cannot be assessed as though a complete testing plan were known.

For the primary outcome, the registry states that all randomized participants were included in an intention-to-treat analysis. That states the intended analysis population but does not explain how missing HAM-D scores were handled. The number of participants with unavailable measurements, reasons for missingness, imputation methods, and sensitivity analyses are not reported in the registry record. Group sizes attached to outcome summaries do not by themselves establish that all randomized participants had complete outcome data.

The analysis population and missing-data details for the secondary outcomes are also not reported in the registry record. The registry does not report interim analyses or stopping rules, nor does it provide the original statistical analysis plan. These omissions prevent a full assessment of prespecification, selective reporting, and the handling of incomplete observations. They do not invalidate the posted summaries, but they limit the strength of claims that can be made from them.

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 registry lists serious adverse event counts of 0 in the paroxetine group, 0 in the venlafaxine extended release group, and 0 in the placebo group. The corresponding numbers at risk are 42, 34, and 39. No serious adverse event terms are listed. Death counts are not reported in the registry record.

These figures describe what the registry recorded for the groups and should not be interpreted as establishing that serious events are impossible or that the groups have equivalent safety. The study groups were limited in size, and the record does not give further detail about event definitions, ascertainment, follow-up completeness, or the time at risk for adverse events.

The registry separately records other adverse events affecting 10 participants assigned to paroxetine, 8 assigned to venlafaxine extended release, and 9 assigned to placebo, with each group’s listed number at risk matching its group size. These counts are distinct from the serious adverse event figures; the registry does not provide event terms in the registry record for this account. Safety interpretation is therefore limited to the recorded counts, without detail on event type, severity among the other events, or comparative statistical analysis.

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

The primary summaries show lower mean HAM-D scores at the end of the 12-week period than at the beginning in all three groups. The medication groups had more negative mean changes than the placebo group, and the unadjusted recalculated differences in mean change were -6.2 for paroxetine and -4.2 for venlafaxine extended release compared with placebo. These describe group averages in the participants enrolled in this study.

The results do not show that every participant’s depression score fell, nor do they provide the proportion who experienced a particular degree of change. Group means can conceal variation in individual outcomes. The HAM-D scale description explains the direction of score changes, but the registry does not specify a minimum clinically important difference or a responder threshold. Statistical differences in group means should not be equated automatically with a clinically meaningful benefit for an individual.

The posted results also do not establish how scores would change beyond the recorded study period, whether effects persist, or whether the results generalize to people who differ from the enrolled population. The registry summary does not provide detailed eligibility criteria, participant characteristics, medication doses, or longer-term outcome results. The stated treatment comparison lasted 12 weeks during the double-blind phase.

These comparisons do not settle the effects of treatment on motor symptoms or the other domains named in the registry. Although those outcomes are listed, complete numerical results and reported analyses are not available in the registry record for all of them. Nor do the results provide a formal comparison of harms; the serious adverse event counts are descriptive, and further safety details are not reported.

The appropriate reading is consequently bounded: the posted data describe differences in mean depression-score change over the study period, with reported intervals and unadjusted p values from recalculated summary comparisons. They do not alone establish a complete benefit-harm profile, long-term effects, or individual outcomes.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it compares two active groups with a placebo group while reporting a continuous primary outcome as change in a depression scale. The placebo group’s mean score also changed over the study period. That makes the between-group contrasts important: they distinguish the observed changes in the medication groups from the change that occurred in the placebo group during the same recorded period.

The registry also demonstrates why estimates and precision should be considered together. A mean difference gives the estimated size and direction of a contrast, while an interval conveys uncertainty under the method used. The reported primary intervals lie below zero, but important qualifications remain: the inferential comparisons are recalculations from summaries, the original confidence procedure is not described, and the registry does not include sponsor-reported analyses.

The study further illustrates the interpretive challenges of multiple outcomes. It lists a broad set of secondary measures, and the available numerical secondary results have a direction similar to the primary result. However, the registry does not report an adjustment strategy or a hierarchy for testing outcomes. A consistent descriptive pattern can be informative without making every unadjusted comparison independently conclusive.

Finally, the record distinguishes an intention-to-treat statement from a complete account of analysis methods. The primary analysis population is described, but missing-data procedures, interim analyses, and the original design assumptions are not reported. For biostatistical interpretation, the distinction between what the registry states and what remains undocumented is as important as the numerical estimate itself.

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 gives an outcome period of 12 weeks. It does not provide longer-term results, so the posted changes describe the recorded double-blind phase rather than later outcomes. The population is described as participants with Parkinson’s disease and depression, but detailed eligibility criteria and participant characteristics are not reported in the registry record. Generalizability beyond the enrolled participants cannot be assessed from these summaries alone.

The primary results are group means and standard deviations, and the contrasts are unadjusted recalculations from those summaries. The registry does not report sponsor analyses, the original statistical analysis plan, the method used to construct the intervals, or any covariate adjustment. It also does not report missing-outcome counts or how missing observations were treated, despite describing the primary analysis as intention-to-treat.

Multiple outcome measures are listed, but the registry does not report a multiplicity strategy or identify a confirmatory hierarchy. Numerical results are not available for every listed measure in the record. This limits interpretation of the secondary outcome family and raises uncertainty about how results were selected for reporting.

The HAM-D scale description gives its range and severity bands, but no threshold for a clinically important change is reported. Consequently, the mean differences cannot be translated into an individual responder rate or a definite judgment about clinical importance. Dosing, adherence, details of randomization and masking, and additional-care information are also not reported in the registry record. These limitations constrain interpretation without changing what the posted group summaries state.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Study of Antidepressants in Parkinson's Disease — NCT00086190.
• Registry record last updated 2013-01-04; status last verified 2013-01.
• Results first posted 2013-01-04.
• 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

The registry record describes a completed phase 3 randomized, quadruple-masked, parallel-group study of paroxetine, venlafaxine extended release, and placebo in people with Parkinson’s disease and depression. It records 115 participants randomized and a 12-week double-blind comparison. The primary outcome was change in HAM-D score from 0 weeks to 12 weeks, with all randomized participants included according to the stated intention-to-treat principle.

Mean HAM-D changes were -13.0 for paroxetine, -11.0 for venlafaxine extended release, and -6.8 for placebo. The unadjusted recalculated differences versus placebo were -6.2 and -4.2, respectively, with both p values <0.001. The registry does not list sponsor-reported inferential analyses, a multiplicity strategy, missing-data methods, or the original sample-size rationale. Serious adverse event counts were 0 in each group; death counts are not reported in the registry record.