Subdissociative Dose Ketamine for Treatment of Acute Pain in Subjects With Chronic Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Subdissociative Dose Ketamine for Treatment of Acute Pain in Subjects With Chronic Pain (NCT02920528), a randomized phase 3 study. The primary outcome is To Compare the Percentage of Subjects Who Achieved Significant Pain Relief Between the 3 Treatment Groups as Measured by a Visual Analog Pain Scale at 60 Minutes A Decrease of at Least 20 mm on the VAS Will be Considered "Significant" Pain Relief.

2. Clinical Question

The trial also aimed to determine whether ketamine could reduce the amount of subsequent opioid pain medication required for adequate relief. That aim is described in the registry summary, but a result for opioid use is not reported in the registry record. The listed secondary outcome concerns risk for adverse events associated with subdissociative ketamine. The primary comparison is a participant-level threshold outcome: each participant either met the specified pain-reduction criterion or did not. The posted group counts show participants meeting that criterion at the 60-minute assessment; they do not describe the full change in pain scores or how participants felt at other times.

3. Trial Design

The registry identifies this as a Phase 3, randomized, triple-masked, parallel-group trial. It was sponsored by the Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center and is marked completed. The recorded study start was 2017-05-01, and primary completion was 2018-06-22. Enrollment is listed as 106.

Three groups are described: placebo, very low dose ketamine at 0.25 mg/kg, and low dose ketamine at 0.50 mg/kg. The primary outcome counts at 60 minutes are reported for 32 participants in the placebo group, 35 in the very low dose group, and 30 in the low dose group. The primary outcome description says that only participants who completed the 60-minute study and had data recorded for each time point were included in the analysis. The relationship between these analysis counts and the enrollment figure is not further detailed in the registry record.

Randomization and triple masking describe features of the design, but the record does not identify the specific randomization procedure or which roles were masked. It describes a parallel model, meaning the treatment groups are compared as separate arms, rather than participants moving between treatments. The posted primary endpoint is a threshold for pain relief, while the stated broader aims also include subsequent opioid use and adverse events. The reported primary group counts and pairwise comparisons address the pain-relief endpoint; results for subsequent opioid use 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

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 To Compare the Percentage of Subjects Who Achieved Significant Pain Relief Between the 3 Treatment Groups as Measured by a Visual Analog Pain Scale at 60 Minutes A Decrease of at Least 20 mm on the VAS Will be Considered "Significant" Pain Relief, measured over 60 minutes.

The primary endpoint counts participants who achieved clinically significant pain relief by 60 minutes. The registry defines success in advance as a decrease of at least 20 mm on the visual analog pain scale from baseline. This is a binary outcome: a participant either meets the threshold or does not. In the placebo group, 13 of 32 participants met it; in the very low dose ketamine group, 28 of 35 did; and in the low dose ketamine group, 25 of 30 did.

The registry says the threshold was chosen as a minimal amount that might be important to this patient group and was extrapolated from studies of acute pain management in the emergency department. Such a threshold can make an outcome easier to interpret as a defined response rather than relying only on an average score change. It also simplifies the result: differences in the number crossing the threshold are visible, but the endpoint does not show how far each successful participant's score fell beyond the threshold, or the distribution of pain-score changes across participants.

The measure concerns pain relief at a specified assessment time. It does not, by itself, describe relief before or after that assessment, subsequent opioid medication use, or adverse effects. Those are separate questions and require their own reported outcomes.

7. Results Posted in the Registry

• Very Low Dose Ketamine: 80.0% (28/35) versus 40.6% (13/32) on control; risk difference 39.4 percentage points (95% CI 17.8 to 60.9), risk ratio 1.97 (1.26–3.09); unadjusted two-proportion z-test p = <0.001.
• Low Dose Ketamine: 83.3% (25/30) versus 40.6% (13/32) on control; risk difference 42.7 percentage points (95% CI 21.1 to 64.3), risk ratio 2.05 (1.31–3.21); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What kind of outcome is analyzed?

The primary endpoint is a binary participant outcome: whether a decrease of at least 20 mm on the visual analog pain scale occurred by 60 minutes. The posted counts are numbers of participants meeting that threshold, rather than average pain scores. A binary outcome allows the group contrast to be expressed as a difference in proportions, a relative risk, or an odds ratio. Those measures answer related but not identical questions. The risk difference gives the absolute contrast in proportions; the relative risk compares the proportions as a ratio; and the odds ratio compares odds rather than proportions.

What do the risk differences describe?

The registry reports risk differences of 0.394 for very low dose ketamine versus placebo and 0.427 for low dose ketamine versus placebo. Each is an unadjusted comparison of the posted participant counts. The positive direction means that the observed proportion meeting the threshold was higher in the named ketamine arm than in placebo. These are comparisons of the observed groups and do not adjust for other participant characteristics. Whether adjustment was planned or performed is not reported in the registry record.

What do confidence intervals and p-values contribute?

As a general statistical explanation, a confidence interval describes uncertainty around an estimated contrast using a specified method. A narrower interval indicates greater precision under that method; a wider interval indicates less precision. An interval is not a range guaranteed to contain the true effect, and its interpretation depends on the analysis assumptions. The posted intervals for both risk differences lie above zero. The associated p-values are reported as p < 0.001 for each comparison. A p-value summarizes compatibility between the data and a null hypothesis under an analysis model; it does not measure clinical importance or prove a treatment effect.

The registry does not name the test or interval method for the reported pairwise estimates. It also does not report an adjusted analysis. The comparison values should therefore be read as the unadjusted results recorded, without assuming a particular model, covariate adjustment, or multiplicity correction.

Why might alternative effect measures look different?

The posted relative risks are 1.969 for very low dose ketamine and 2.051 for low dose ketamine, each compared with placebo. The posted odds ratios are 5.846 and 7.308. Odds ratios can appear farther from the null value than relative risks when an outcome is common; this is a general property of the measures, not evidence of a separate analysis result. A relative risk compares proportions, while an odds ratio compares the odds of the event. Neither should be confused with the risk difference, which expresses the contrast on the proportion scale.

What does randomization contribute?

As a general principle, random assignment can reduce systematic differences between groups at the time of allocation, while masking can reduce some forms of influence during care, assessment, or reporting. The registry describes the design as randomized and triple-masked, but details of the allocation sequence, concealment, and the roles masked are not reported in the registry record. The design label alone does not establish that all potential sources of bias were eliminated.

What analysis population is represented?

The endpoint description states that only participants who completed the 60-minute study and had data recorded for each time point were included in the analysis. This indicates that the posted endpoint counts are not explicitly described as an analysis of every enrolled participant regardless of missing data. The record does not report how many participants were excluded for incomplete follow-up, whether missing values were imputed, or whether an intention-to-treat analysis was conducted. These details matter because excluding participants with incomplete measurements can affect the result if missingness is related to treatment or outcome.

What is not established by the methods record?

The registry does not report a detailed statistical analysis plan, the test underlying the pairwise p-values, a multiplicity procedure, interim analyses, or the handling of any other missing outcomes. Nor is a direct statistical comparison between the ketamine doses reported. These omissions do not change the arithmetic of the displayed counts; they limit how completely the analysis process can be reconstructed. General statistical concepts explain what the measures mean, but the registry record does not provide enough detail to attribute any additional methods to this trial.

10. Confidence Intervals and Precision

The point estimates summarize the observed contrasts, while the confidence intervals indicate how precisely those contrasts are estimated under the analysis method. For very low dose ketamine versus placebo, the risk difference is 0.394, with a confidence interval from 0.178 to 0.609. For low dose ketamine versus placebo, it is 0.427, with a confidence interval from 0.211 to 0.643. Both intervals are entirely above zero, consistent with the positive observed differences and the reported p-values of p < 0.001.

The intervals are not especially narrow relative to their point estimates, so a range of contrast sizes remains compatible with the data. They do not indicate that the point estimates are exact, and they should not be read as a guarantee about the result in another study or population. The group counts are 32 in placebo, 35 in very low dose ketamine, and 30 in low dose ketamine; the endpoint description says only participants completing the 60-minute study with data at each time point were included. The record does not explain the relationship between those analyzed counts and the enrollment of 106 or detail exclusions due to missing data.

Other measures also show uncertainty. The relative-risk intervals are 1.255 to 3.09 and 1.31 to 3.212 for the respective ketamine comparisons. The odds-ratio intervals are 1.97 to 17.35 and 2.22 to 24.06. The odds-ratio intervals span a particularly broad range, illustrating that relative measures can be imprecise even when a test yields a small p-value. The registry does not report the interval method, so its assumptions cannot be checked from the record.

As a general statistical point, precision depends on the number of observed outcomes and the distribution of participants across groups, as well as the variability and analysis method. A confidence interval describes uncertainty in the estimated contrast; it is not a statement that individual outcomes are predictable. The posted results offer evidence about the threshold outcome at the recorded assessment, but they do not pin down a single exact population effect.

11. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm for comparisons between two groups, not total events. They are illustrative calculations based on the observed effect, rather than a report of the trial's observed power. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for each scenario is a stated fraction of the observed effect.

For each comparison, the planning values consider 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The 80% power column means the number of participants per arm used in the illustration to give an 80% chance of detecting the specified effect under those assumptions, if that effect were real. The 90% power column gives the corresponding per-arm planning number for a 90% chance. These are design calculations, not evidence that the trial had a particular probability of producing a statistically significant result after data collection.

Smaller effects require more participants to detect with the same target power and significance level. Thus, the planning values rise as the assumed effect is reduced from the full observed effect to a smaller fraction. The figures are based on the observed contrasts in this trial, which makes them useful as conditional planning illustrations, not independent estimates of what a future study will observe. They do not replace a design calculation based on a clinically justified target effect and appropriate assumptions for a future study.

The registry's own endpoint description separately records a design rationale: it says a sample size of 96 was required to detect a statistically significant difference among the three groups with 90% power and a = 0.05, using a 20-mm change as the marker for success; it also says 106 were recruited in anticipation of 10% loss due to withdrawal or incomplete data. Those are the trial's recorded planning statements. The pairwise planning illustrations use different stated assumptions and should not be conflated with that original three-group rationale.

12. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.406; treatment proportion p₁ = 0.800; average p̄ = 0.603.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6919 = 1.3561.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6334 = 0.5331.
4. n = (1.3561 + 0.5331)² / (0.3938)² = 23.0, rounded up to 24 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
p0 <- 0.40625                          # control proportion
p1 <- 0.8                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2920528)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 24 (80% power), 31 (90% power)
 75% of observed effect: n per arm = 44 (80% power), 58 (90% power)
 50% of observed effect: n per arm = 101 (80% power), 134 (90% power)
power.prop.test: 23 
Simulated power at n = 24 per arm: 0.836 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The posted results include two pairwise comparisons, each comparing a ketamine arm with placebo, and both are reported with unadjusted p-values of p < 0.001. When multiple hypotheses are tested, the chance of at least one false-positive result can increase if no adjustment or prespecified testing strategy is used. This is a general statistical concern; the registry record does not report whether a multiplicity adjustment was planned or applied. The p-values should therefore be interpreted as the unadjusted values recorded, not as evidence that any particular multiplicity procedure was used.

The registry identifies three treatment groups and describes the original power rationale as detecting a statistically significant difference among groups. The posted comparisons do not include an overall three-group test or a direct comparison between the two ketamine doses. Whether such tests were conducted, and how they were handled in an analysis plan, is not reported in the registry record. The posted pairwise results answer the separate questions of each ketamine dose versus placebo; they do not establish a difference between the doses.

For missing data, the primary endpoint description states that only participants completing the 60-minute study and with data recorded for each time point were included in the analysis. The record does not say how many participants had incomplete data, how missing outcomes were handled, or whether sensitivity analyses were conducted. It does not report whether the primary analysis followed an intention-to-treat principle or another defined analysis population. Those details are important because the analyzed participants may differ from those with missing measurements.

The registry lists enrollment of 106 and endpoint group counts of 32, 35, and 30. It does not explain the full accounting between enrollment and the endpoint analysis. The analysis counts should not be treated as evidence of the reasons for any difference. A detailed protocol or statistical analysis plan could clarify the population definition, missing-data approach, and multiplicity strategy, but those details are not reported in the registry record.

14. 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: dizziness (22.9%), nausea (10.0%), hallucinations (6.7%), anxiety and dizziness (6.7%), anxiety (3.3%).

The serious adverse event figures count participants affected in each group, with the at-risk counts also recorded: 1 of 32 in placebo, 14 of 35 in very low dose ketamine, and 12 of 30 in low dose ketamine. The registry records 0 deaths in each group. For other adverse events, the affected counts are 0 in placebo, 0 in very low dose ketamine, and 0 in low dose ketamine.

These are descriptive counts. They do not, on their own, establish that an event was caused by treatment, nor do they show the timing, duration, severity distribution, or clinical course of each event. The registry lists serious-event terms including Dizziness at 22.9%, Nausea at 10.0%, Hallucinations at 6.7%, Anxiety and Dizziness at 6.7%, Anxiety at 3.3%, and Dysphoria at 3.3%. The record does not make the denominator or group assignment for these term percentages clear in the posted summary, so they should not be attributed to a particular arm from these figures alone.

Safety comparisons should be interpreted with the size and follow-up of the groups in mind. The record does not provide further detail about event ascertainment or statistical comparisons of adverse events.

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

The posted primary results show that more participants in each ketamine group than in placebo met the registry's threshold for significant pain relief at 60 minutes. The unadjusted risk differences are positive for both comparisons, their confidence intervals lie above zero, and both p-values are reported as p < 0.001. These statements describe the recorded endpoint and comparisons; they are not a recommendation about treatment.

The results do not show the complete course of pain over time, the size of score change for every participant, or how durable any relief was. They do not answer the registry's stated question about subsequent opioid medication use, because a result for that outcome is not reported in the registry record. Nor do the pairwise comparisons establish whether the two ketamine doses differ from one another. The primary endpoint uses a threshold, so participants with quite different score reductions may be counted together once they meet it.

Randomization and masking are design strengths in principle, but the record lacks procedural detail about allocation, masking roles, analysis methods, and missing data. The endpoint description says only participants with complete 60-minute study data were included, while the handling and number of incomplete records are not given. The posted counts therefore describe the analyzed participants and cannot by themselves resolve possible effects of missingness.

The safety counts report serious adverse events in each group and no deaths, but they are descriptive and do not establish causation. The record does not offer enough detail to assess all aspects of safety or attribute the listed event terms to specific groups. Overall, the registry results support a comparison of the recorded 60-minute threshold outcome in this trial; they do not settle broader questions about longer-term relief, other outcomes, dose differences, or effects in populations beyond the trial.

16. Why This Trial Matters Statistically

This trial illustrates how the choice of endpoint shapes a clinical comparison. Rather than comparing only average pain scores, the primary outcome defines a participant-level response threshold: a decrease of at least 20 mm on the visual analog scale by 60 minutes. That makes the result interpretable as a count of participants meeting a specified criterion, while leaving out information about the full distribution of score changes.

It also illustrates why effect measures should be read together. The risk difference gives the contrast in proportions, while relative risk and odds ratio express the comparison on different scales. Confidence intervals show uncertainty around these estimates, and the p-values address a different question about compatibility with a null hypothesis. None of these statistics alone conveys every aspect of clinical importance.

The registry records two comparisons against placebo, but no direct comparison between the dose groups. This distinction matters: finding a contrast for each dose versus placebo does not determine whether the doses differ from each other. Likewise, the stated aim of reducing subsequent opioid medication use cannot be evaluated without a posted result for that outcome.

The record also demonstrates why analysis-population and missing-data details matter. Enrollment is listed as 106, while the endpoint group counts are 32, 35, and 30; the endpoint description specifies inclusion of participants with complete 60-minute study data, but does not provide further accounting. Reporting those details alongside estimates and intervals helps readers assess how confidently a result can be interpreted. Here, the reported numbers permit a clear description of the endpoint contrasts, while important analysis and follow-up details remain unreported.

17. 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 limitations are specific to interpreting this record. The primary result is a single threshold endpoint at 60 minutes. It does not describe the magnitude or distribution of pain-score changes across participants, or whether relief continued after the assessment. The registry's broader aim concerning subsequent opioid use does not have a posted result in the record.

The endpoint description restricts analysis to participants who completed the 60-minute study and had data for each time point. The registry does not report the number or characteristics of participants with incomplete data, the reasons for missingness, or sensitivity analyses. Enrollment is listed as 106, but the relationship between that number and the endpoint analysis counts is not explained. This limits assessment of how exclusions may have influenced the results.

The reported comparisons are pairwise against placebo. There is no posted direct comparison between the two ketamine doses, and the record does not specify whether the pairwise p-values were adjusted for multiple testing. The methods underlying the confidence intervals and tests are also not identified. These gaps limit reconstruction of the analysis and interpretation of the inferential results.

The safety summary provides serious adverse event counts and lists event terms, but does not fully describe event timing, attribution, or the denominator and group assignment for the listed term percentages. The record also does not provide enough detail about randomization procedures or which roles were masked to evaluate those design features fully. These are limitations of the registry information available for interpreting the trial, not findings that the study necessarily lacked those procedures.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Subdissociative Dose Ketamine for Treatment of Acute Pain in Subjects With Chronic Pain — NCT02920528.
• Registry record last updated 2021-04-15; status last verified 2021-03.
• Results first posted 2021-04-15.
• 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.

21. Record Summary

NCT02920528, titled “Subdissociative Dose Ketamine for Treatment of Acute Pain in Subjects With Chronic Pain,” is recorded as a completed Phase 3 randomized, triple-masked, parallel-group trial sponsored by the Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center. Enrollment is listed as 106. The groups were placebo, very low dose ketamine at 0.25 mg/kg, and low dose ketamine at 0.50 mg/kg.

The primary endpoint was the number of participants with a decrease of at least 20 mm on the visual analog pain scale at 60 minutes. The posted counts meeting that threshold were 13 of 32 for placebo, 28 of 35 for very low dose ketamine, and 25 of 30 for low dose ketamine. Each ketamine-versus-placebo comparison has an unadjusted p-value of p < 0.001. The record also reports serious adverse event counts and 0 deaths in each group. Results for subsequent opioid use and details of the missing-data and multiplicity approaches are not reported in the registry record.