Ketamine in the Treatment of Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Ketamine in the Treatment of Depression (NCT01558063), a randomized phase 2 study. The primary outcome is Number of Responders 24-hours Post-ketamine Infusion.

2. Clinical Question

3. Trial Design

The registry identifies this as a randomized, triple-masked, parallel Phase 2 trial titled “Ketamine in the Treatment of Depression,” with the official title “The Antidepressant Action of Ketamine: Brain Chemistry.” Columbia University is listed as the sponsor. The condition is Major Depressive Disorder. The record reports enrollment of 38, a start date of 2012-02, primary completion in 2015-05, and completed status.

The study description combines an experimental depression-treatment component with MRI measurement of brain chemistry. The listed intervention arms include five intravenous ketamine doses, from 0.1 mg/kg through 0.5 mg/kg, and a saline solution arm. The saline infusion is described as lasting 40 minutes. MRI scanning is included with the listed dose and saline arms. The brief description also mentions a 0.0 mg/kg placebo dose and an optional second active ketamine infusion for some participants; the posted arm list identifies saline as the comparator and does not show a separate placebo-dose arm.

The registry description says that an optional second scan and 0.5 mg/kg ketamine infusion could be offered when a participant did not respond after the initial infusion, including when the initial treatment was placebo or a dose considered too low. The timing is described only as occurring no later than weeks after the initial scan and infusion, as scheduling permits. It also describes up to 6 months of treatment with the study psychiatrist after ketamine infusion or infusions. How these features were implemented, and whether they affected the posted primary endpoint, are not reported in the registry record.

The record also describes healthy volunteers receiving a single 0.5 mg/kg infusion and MRI scan. It does not provide a separate healthy-volunteer group in the posted primary outcome results or explain how those participants contributed to enrollment or analysis. The registry lists change in glutamate levels and change in gamma-amino butyric acid (GABA) levels as secondary outcomes. Details of the randomization procedure, masking roles, and analysis plan 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.

Overall Study

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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Number of Responders 24-hours Post-ketamine Infusion, measured over Day 1 (post ketamine).

The primary endpoint is the number of participants classified as responders 24 hours after ketamine infusion. The registry says depressive symptom ratings were collected at baseline, Day 1, and Day 3 using HDRS-24, described as a 24-item questionnaire used to indicate depression and guide evaluation of recovery. A responder was defined as an individual whose HDRS score decreased from baseline to 24 hours after treatment. Individuals who did not meet that definition were classified as non-responders.

This is a binary outcome at the participant level: each participant is classified as a responder or non-responder, and the posted group result is a count of responders. A count is straightforward to interpret, but by itself it does not show the uncertainty around a response rate or quantify a treatment effect relative to another group. The registry record does not report a response threshold beyond any reduction in score, nor does it state how missing Day 1 assessments were handled.

The endpoint is intended to capture an early change in depressive symptom ratings after infusion. The record describes MRI measurements of brain chemistry as part of the study, with glutamate and GABA changes listed as secondary outcomes. Those measures are distinct from the primary clinical response classification. The registry’s description of the HDRS total-score range and scoring should be read as a registry description; additional scoring details or endpoint validation information are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is this?

The primary measure is binary at the participant level: a person either meets the responder definition or does not. The registry defines response as any reduction in HDRS score from baseline to 24 hours after treatment. The posted group results are counts of responders, accompanied by group denominators. A count is not the same as an estimated response probability, a risk difference, or an odds ratio. Such comparative quantities require a stated analysis, and none is reported in the registry record.

For a binary outcome, uncertainty depends on both the observed count and the number of participants assessed. Small groups can produce imprecise estimates, even when the observed counts look different. A confidence interval would communicate a range of values reasonably compatible with the data under the statistical method used. The registry record does not report confidence intervals for these results, so their precision cannot be assessed from the counts alone.

What can randomization and masking tell us?

The trial is labeled randomized, triple-masked, and parallel. In general, random allocation can help make comparison groups similar with respect to measured and unmeasured characteristics at the start of a study. Masking can reduce the influence of expectations or knowledge of assignment on care, assessment, or reporting. These are general design principles, not evidence that every potential source of bias was eliminated in this trial. The registry record does not describe the randomization sequence, allocation concealment, or the parties who were masked. It also does not report whether masking was assessed.

A parallel design generally assigns participants to a group that is followed as a separate treatment arm, rather than having every participant receive each study treatment in sequence. The registry uses the parallel designation, but does not provide enough detail to describe the operational allocation process or the handling of the optional second infusion. The posted results should therefore be read as the reported groups and counts, without assuming unreported design or analysis details.

Why does a missing comparison matter?

Descriptive counts show what was recorded in each group. A comparative analysis would quantify how much the outcomes differ between groups and how uncertain that estimate is. Because the outcome is binary and the groups are small, a selected statistical method can affect the estimate and interval. A test might also address whether observed variation is consistent with a null comparison, but a p-value would not measure the size or clinical importance of a difference. No such comparison is reported in the registry record. The record indicates that values were missing or not estimable for a two-arm comparison of the binary endpoint.

There are several ketamine dose groups and a saline group. In general, making multiple comparisons raises a multiplicity question: the more comparisons considered, the greater the chance of at least one apparently unusual result under a setting where no true differences exist. Investigators may define a primary contrast, adjust for multiple testing, or use another prespecified strategy. The registry record does not report a multiplicity strategy, a primary dose-versus-saline contrast, or a dose-trend model. No one comparison should be treated as the definitive test without that information.

What does over-dispersion mean for this endpoint?

Over-dispersion is a general issue for some count models when observed variation exceeds what the model assumes. It is not the natural first description of these results because the primary endpoint is a count of participants classified as responders within each group, derived from individual binary outcomes. In a binary-outcome analysis, the key modeling choices would instead include the effect measure, any covariate adjustment, and the method used to estimate uncertainty. The registry does not report these choices. There is no basis in the record to claim that an over-dispersion adjustment was or was not needed.

How should planning calculations be read?

Sample-size calculations are planning tools, not summaries of trial results. The stated planning assumptions describe an illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The calculation is framed around detecting a stated fraction of the observed effect. These assumptions are not reported as the original trial’s design justification. A planning illustration based on an observed result is not observed power and does not show that the trial had adequate power for its original objectives. The trial’s own sample-size justification is not reported in the registry record.

In general, power is a design-stage probability calculated under specified assumptions about an effect, variability, significance level, and analysis. It is not a measure of whether a completed study’s result is true. Recalculations based on observed summaries can be useful as planning illustrations, but they depend on the observed effect and may be unstable, especially with small groups. No numerical planning estimates are reported in the registry record, so no such values can be interpreted.

What cannot be reconstructed from the record?

The record does not provide an analysis population definition, a missing-data method, covariate adjustments, confidence intervals, or a formal primary endpoint comparison. It also does not state whether the posted counts were analyzed according to randomized assignment or actual treatment received. These details affect the interpretation of a trial result. Without them, the most defensible account is descriptive: the posted counts record how many participants in each listed group were classified as responders under the stated definition. They do not supply the inferential detail needed to judge comparative precision or statistical significance.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Ketamine Dose 1", "Ketamine Dose 2", "Ketamine Dose 3", "Ketamine Dose 4", "Ketamine Dose 5", "Saline Solution")
n      <- c(5, 6, 8, 5, 9, 5)
events <- c(3, 1, 3, 2, 2, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Ketamine Dose 1  5  3  60.0  14.7  94.7  23.1  88.2  22.3  88.2
ARM  Ketamine Dose 2  6  1  16.7  0.4  64.1  3.0  56.4  3.7  57.9
ARM  Ketamine Dose 3  8  3  37.5  8.5  75.5  13.7  69.4  13.7  70.1
ARM  Ketamine Dose 4  5  2  40.0  5.3  85.3  11.8  76.9  11.8  77.7
ARM  Ketamine Dose 5  9  2  22.2  2.8  60.0  6.3  54.7  6.7  55.6
ARM  Saline Solution  5  0  0.0  0.0  52.2  0.0  43.4  0.4  45.9

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two secondary outcomes: change in glutamate levels and change in GABA levels. It reports three outcome measures, but only the primary endpoint results are described in the posted data used for this interpretation. Results for the secondary outcomes are not reported in the registry record. The record does not state which comparisons were prespecified as primary, whether the several ketamine doses were each intended to be compared with saline, or how multiplicity across doses and outcomes was handled.

Multiplicity matters because considering several comparisons can increase the chance of finding an apparently notable result by chance. A study may address this by specifying a primary comparison, adjusting statistical thresholds, or using a hierarchical testing strategy. The registry record does not report such a plan. Without that information, no single dose-group contrast should be treated as the definitive planned comparison.

Missing-data handling and analysis populations are also not reported in the registry record. The endpoint requires baseline and Day 1 HDRS information, but the record does not say how participants without a usable assessment were classified or whether any observations were imputed. It does not define an intention-to-treat, per-protocol, or other analysis population, nor explain the handling of withdrawals or protocol deviations.

The registry describes an optional second infusion for some participants, but does not report how repeat treatment was handled in the endpoint analysis. The effect of any such treatment on classification, follow-up, or group assignment is therefore unclear. These omissions limit interpretation of both the responder counts and any potential comparisons among groups.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: unrelated medical illness (5.3%), active suicidal ideation (2.6%), antidepressant misuse (2.6%).

The registry reports serious adverse events for an “All Subjects” group, with 4 affected among 38 at risk. It reports other adverse events for 0 affected among 38 at risk, and deaths of 0. These are registry counts; they do not, by themselves, establish whether an event was caused by a study intervention. The record does not provide event-level narratives, timing, severity grading beyond the serious-event designation, or a causal attribution method.

The registry lists Unrelated Medical Illness at 5.3%, Active Suicidal Ideation at 2.6%, and Antidepressant Misuse at 2.6% among the top serious adverse-event terms. The record does not explain the denominator or counting method for each listed percentage, nor whether terms could overlap or reflect repeated events. These terms should be understood as the registry’s listed safety information, not as a basis for inferring causation.

A report of no deaths or no other adverse events in the posted fields does not establish that no such events occurred outside the recorded definitions or reporting period. Safety interpretation depends on how events were collected, classified, and attributed, and on the time under observation. Those details are not reported in the registry record. The figures therefore describe what was posted, rather than providing a complete comparative assessment of safety between treatment groups.

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

The posted primary results describe the number of participants classified as responders 24 hours after infusion in each listed dose group and in the saline group. The response definition is a reduction in HDRS score from baseline to Day 1. The counts offer a direct description of the recorded outcome, but they are not accompanied by confidence intervals or a formal comparison. As a result, the precision of the group results and the uncertainty around differences cannot be assessed from the registry record.

The findings do not establish that ketamine caused the observed responses. Randomization is an important design feature, but the record does not report the analysis method, analysis population, missing-data handling, or the comparison specified as primary. Nor does it report how an optional second infusion was handled in the primary endpoint. Those details are needed to understand what contrast the results represent and how securely it can be interpreted.

The counts also do not establish that response differs by dose or follows a dose-response pattern. Comparing several groups informally is not a substitute for a prespecified statistical analysis. The record does not provide results for the listed glutamate and GABA secondary outcomes, so the posted primary outcome cannot be used to infer changes in those measures or a relationship between brain chemistry and clinical response.

The planning assumptions described for sample-size illustrations are not the original design justification, and such illustrations are not observed power. They do not prove that the study was adequately powered or that a nonsignificant result would indicate no effect. No p-value or inferential comparison is reported in the registry record.

In short, the posted data support a descriptive reading of responder counts and registry safety fields. They do not establish clinical benefit, lack of benefit, comparative safety, or the clinical importance of any group difference. Further trial methods and analyses would be needed for those conclusions.

14. Why This Trial Matters Statistically

This trial is statistically interesting because it combines a clinical binary endpoint with several ketamine dose groups and a saline comparator. The primary outcome is easy to state—whether an individual’s HDRS score decreased by Day 1—but the simplicity of the classification does not remove uncertainty. Counts from small groups can be imprecise, and comparisons across several doses require a clearly defined analysis strategy.

The record illustrates the difference between describing outcomes and estimating effects. The posted responder counts show what was recorded in each group. Without a comparison estimate and confidence interval, readers cannot gauge the magnitude or precision of differences. Without a reported analysis plan, readers also cannot determine whether a particular dose contrast was the prespecified focus or one of several comparisons considered.

Design features such as randomization and masking are relevant, but they do not answer every analytical question. The record does not report who was masked, how allocation was implemented, or how the optional second infusion was handled. These details are especially relevant when a short-term endpoint follows an initial infusion and additional treatment may be offered to some participants.

The trial also includes MRI-based brain chemistry measurement and lists glutamate and GABA changes as secondary outcomes. This creates an important distinction between clinical symptom response and biological measurements. The registry record does not post results for those secondary outcomes, so it does not permit a statistical assessment of their relationship to the primary endpoint.

Finally, planning calculations based on an observed effect should not be confused with the study’s original design or with observed power. The record makes clear that its planning assumptions are illustrative. This distinction helps prevent a retrospective planning exercise from being mistaken for evidence about the certainty or validity of the trial’s findings.

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.

The primary endpoint definition is broad: any reduction in HDRS score from baseline to Day 1 qualifies as a response. The registry does not report a minimum reduction needed to count as clinically meaningful, or additional details about score handling. The record describes HDRS-24 and gives a total-score range, but does not provide further scoring or assessment procedures. This makes it difficult to judge how sensitive the responder classification is to small score changes.

The posted primary results are counts only. No confidence intervals, comparative effect estimates, or inferential tests are reported. The record also does not identify a primary dose-versus-saline contrast or a strategy for handling multiple dose comparisons. These omissions limit conclusions about the magnitude, precision, and statistical significance of group differences.

The record does not explain the relation between the healthy volunteers described in the study summary and the posted outcome groups. It also describes an optional second active infusion for some nonresponders, but does not state whether those participants’ primary outcome data reflect the first infusion alone or how repeat treatment was handled. This limits clarity about the intervention exposure represented by the endpoint.

Missing-data procedures, analysis populations, and the treatment of withdrawals or protocol deviations are not reported in the registry record. The listed secondary outcomes have no posted results in the information available for interpretation. These are additional gaps alongside the absence of a detailed statistical analysis plan and original sample-size justification.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Ketamine in the Treatment of Depression — NCT01558063.
• Registry record last updated 2019-12-10; status last verified 2019-12.
• Results first posted 2019-07-16.
• 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 describes a completed, randomized, triple-masked, parallel Phase 2 study sponsored by Columbia University, with enrollment of 38 and a focus on Major Depressive Disorder. The study examined intravenous ketamine at five listed doses and saline, alongside MRI assessment. It also describes healthy volunteers and an optional second ketamine infusion for some nonresponders.

The primary outcome is the number of responders 24 hours after infusion, defined as a reduction in HDRS score from baseline. The posted results give responder counts by dose and saline group, but no confidence intervals or formal comparative analysis. The registry lists glutamate and GABA changes as secondary outcomes without posting their results. Missing-data methods, analysis populations, multiplicity handling, and the original sample-size justification are not reported in the registry record.