Optimal Dosing of Ketamine for Procedural Sedation and Analgesia in Children: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Optimal Dosing of Ketamine for Procedural Sedation and Analgesia in Children (NCT02519595), a randomized phase 4 study. The primary outcome is Sedation Efficacy.

2. Clinical Question

The study was motivated by variation in published intravenous ketamine dosing regimens for pediatric emergency-department procedural sedation. Its design compared dose groups rather than testing ketamine against another drug or against no sedation. That distinction defines what the results can address: they characterize outcomes across the three assigned doses in the trial, not the effects of ketamine versus another approach. The registry record does not report the procedures performed, the participants’ baseline characteristics, or further details that would establish how closely the enrolled group represents other clinical settings.

3. Trial Design

The registry identifies this as a Phase 4, randomized, quadruple-masked, parallel-group trial with an enrollment of 182. Participants were assigned to one of three intravenous ketamine dose groups: 1 mg/kg, 1.5 mg/kg or 2 mg/kg. In a parallel design, each participant belongs to an assigned group rather than receiving each dose in sequence. The arm descriptions say that the dose was blinded. The registry’s masking label is quadruple, but it does not identify the four roles or people covered by that label.

The registry lists Children’s Hospital of Michigan as sponsor and gives the study status as completed. The recorded study start is 2010-08, and primary completion is recorded as 2012-08. These dates describe the registry entry; more detailed timing of recruitment, follow-up, or data collection is not reported in the registry record.

The study summary says the three doses were administered using the traditional method of infusion over 30-60 seconds. The primary endpoint was maximum sedation depth on the Ramsay Sedation Scale, assessed after study drug administration; its timeframe is described as approximately 2 hours. Secondary outcomes listed in the record include sedation duration, additional dose, adverse events, and sedation satisfaction. The additional-dose timeframe is described as approximately 1 hour during the procedure.

The primary outcome results give participant counts of 50 for 1 mg/kg, 35 for 1.5 mg/kg, and 40 for 2 mg/kg. These are the numbers attached to that outcome, not a complete account of the enrolled population. How participants were handled in each analysis, and why the outcome-specific counts differ from the enrollment total, are not reported in the registry record. The allocation ratio, detailed randomization procedures, eligibility rules, and any protocol deviations are also 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: 11 consented patients were excluded as sedation and procedure was performed in the operating room rather than in the Emergency Department

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.

Population: Children aged3-18 years who received procedural sedation and analgesia in the Emergency Department

6. Primary Endpoint

The registry lists the primary outcome as Sedation Efficacy, measured over Participants will be assessed after study drug adminsitration and the maximum depth of sedation achieved recorded. Approximately 2 hours.

The primary endpoint is sedation efficacy, recorded as the maximum depth of sedation reached after the study drug was administered. The registry defines the measure using the Ramsay Sedation Scale, which ranges from 1 to 6. At the lower end, a score of 1 describes an anxious, agitated or restless participant; a score of 2 describes a cooperative, oriented and tranquil participant. Higher values represent progressively deeper sedation, with a score of 6 defined as no response to a light glabellar tap or loud auditory stimulus. The record notes that ketamine usually provides a depth of 5 or 6, but that statement is background information, not a separate result from this trial.

Maximum score is useful for describing the deepest sedation observed during the assessment period. It does not, by itself, describe how long that depth lasted, how quickly it was reached, or whether additional medication was needed. Those topics appear among the secondary outcomes. The registry records the primary endpoint as a median with an inter-quartile range and gives an assessment timeframe of approximately 2 hours after study drug administration. The posted arm results include medians and range limits, but the inter-quartile ranges themselves are not reported in the registry record.

As a general statistical point, a scale with ordered categories can be summarized by its median because the median identifies the middle of the observed ordered scores without requiring equal spacing between every score. That makes a median-based comparison a reasonable way to describe the typical recorded depth, while still leaving the full distribution and participant-level experience unrepresented by a single summary.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is summarized by a median for each dose group, with the endpoint described as having an inter-quartile range. The posted results give the group medians and range limits: 5.5 (4.0 to 6.0) for 1 mg/kg, 6.0 (4.0 to 6.0) for 1.5 mg/kg, and 6.0 (5.0 to 6.0) for 2 mg/kg. The registry does not provide the inter-quartile ranges themselves. Range limits and inter-quartile ranges are different summaries: the range spans the smallest to largest observed values, while the inter-quartile range describes the central half of the observations. General statistical interpretation therefore needs to distinguish the posted limits from the stated dispersion measure.

The reported primary comparison uses the Kruskal-Wallis test across the three groups, with p = 0.2. In general, this rank-based test evaluates whether observations across multiple independent groups have similar distributions, using their ordering rather than requiring a normally distributed continuous outcome. It is useful for an ordered scale such as the Ramsay score, particularly when a scale has only a small set of possible values. The test is an overall comparison: it does not identify which particular pair of groups might differ. Pairwise primary comparisons are not reported in the registry record.

A p-value describes how compatible the observed data are with a specified null model, assuming that model and the test’s conditions hold. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The registry reports p = 0.2 for the omnibus primary comparison, but does not give a confidence interval or a direct estimate of an overall difference. In general, confidence intervals communicate the range of effect values reasonably compatible with the data and model; no such interval is reported for the primary comparison.

The additional-dose comparisons are presented as unadjusted recalculations using posted counts. They compare the 1 mg/kg reference group with each higher-dose group using risk differences. A risk difference is the event proportion in one group minus the event proportion in the reference group; negative values in these comparisons indicate a lower observed proportion in the higher-dose group. For 1.5 mg/kg, the recalculated risk difference is -0.131, with a 95% confidence interval from -0.247 to -0.016 and the unadjusted p-value = 0.053. For 2 mg/kg, it is -0.11, with a 95% confidence interval from -0.232 to 0.012 and the unadjusted p-value = 0.099. These are unadjusted pairwise results and are not the trial’s reported primary test.

As a general interpretation, a confidence interval that includes zero for a risk difference is compatible with no difference under the interval’s assumptions, while an interval that excludes zero indicates greater evidence against a zero difference before accounting for multiplicity or other design considerations. The intervals and p-values should be considered together rather than treating a single threshold as the whole interpretation. The reported p-values for these secondary recalculations are unadjusted; the registry does not report a multiplicity adjustment for them. The confidence intervals are also recalculated from the posted summaries, not stated as sponsor analyses.

The record includes planning illustrations based on fractions of the observed additional-dose effect. These are hypothetical sample-size calculations, not observed trial results and not the trial’s original sample-size justification. They use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. General statistical reasoning explains why reducing the effect targeted for detection increases the required number of participants: smaller differences are harder to distinguish from random variation. The registry explicitly characterizes these figures as planning illustrations based on the observed effect, not the trial’s design. The actual protocol’s sample-size assumptions and justification are not reported in the registry record.

Over-dispersion is a general concern when count or binary outcomes vary more between participants or clusters than a simple model assumes. If present, it can make standard errors too small and confidence intervals too narrow unless the analysis accounts for it. The additional-dose recalculations use posted binary counts and do not describe an over-dispersion model. For this endpoint, the registry does not report a sponsor analysis or a model-based adjustment. The primary rank-based analysis is a different method and should not be interpreted through assumptions specific to a binary-outcome model.

What does the overall test tell us?

The Kruskal-Wallis result is an omnibus comparison across all three dose groups. It can indicate evidence of differences somewhere among the group distributions, but it does not quantify the differences or point to a particular pair. The group medians add descriptive context, while the reported p-value gives the test result. Neither alone establishes equality or equivalence.

What are the recalculated comparisons?

The risk differences and their confidence intervals for additional dose are unadjusted recalculations from the posted counts. They are not results from the sponsor’s reported analysis. The registry does not list reported inferential analyses for this secondary endpoint. Therefore, the recalculations are best treated as a transparent descriptive analysis rather than as a substitute for a prespecified trial analysis.

10. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports the sponsor’s overall analysis: a Kruskal-Wallis test across the three dose groups, with p = 0.2. That is the main inferential result posted for sedation depth. The group medians and ranges provide descriptive information, but the record does not include a sponsor-reported confidence interval or pairwise primary comparisons. Accordingly, the primary interpretation rests on the posted omnibus test and the descriptive summaries, without a more precise estimate of differences between particular doses.

The recalculated risk differences concern the secondary outcome of receiving an additional dose. They are simple unadjusted calculations from the posted group counts, not the sponsor’s reported analysis. The registry lists no reported inferential analyses for that secondary endpoint. These recalculations therefore supplement the count summaries but do not replace a sponsor analysis. They also do not alter the primary endpoint result, because they address a different outcome.

The record does not flag a discrepancy between a sponsor analysis and a recalculation. Even so, the distinction in source and purpose matters: the primary test is the analysis reported in the registry, whereas the additional-dose estimates are unadjusted recalculations. A reader should not treat the recalculated secondary findings as if they had been specified as the trial’s primary analysis or as if they were adjusted for multiple outcome comparisons.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and several secondary outcomes, including sedation duration, additional dose, adverse events, and sedation satisfaction. It also reports an overall primary comparison and provides unadjusted recalculated pairwise comparisons for additional dose. The analysis plan does not describe whether the primary endpoint was the sole confirmatory test, how secondary outcomes were prioritized, or whether any adjustment was planned for multiple comparisons. Those details are not reported in the registry record.

Multiplicity is a general issue when many hypotheses are examined. Even when each individual test uses a conventional significance threshold, repeated testing can increase the chance of a false-positive finding across the collection of tests. Adjustments or a prespecified testing hierarchy can address this concern, but they may affect the interpretation and precision of results. No such procedure is reported. The additional-dose p-values are explicitly unadjusted recalculations, so they should be treated as exploratory rather than as multiplicity-controlled evidence.

The registry does not report how missing outcome data were handled, whether participants were excluded from any analysis, or whether an intention-to-treat, per-protocol, or other analysis population was used. The primary results include outcome-specific group counts of 50, 35 and 40, while total enrollment is 182. These figures alone do not explain why the counts differ; the reasons, missingness patterns, and any imputation or sensitivity analyses are not reported in the registry record. It would not be appropriate to assume that all enrolled participants contributed to every endpoint or to infer the treatment of unobserved outcomes.

Other elements needed to assess the statistical plan are also absent from the record, including a detailed protocol, the original sample-size justification, any interim analyses, and prespecified analysis populations. The posted sample-size illustrations are based on the observed secondary effect and are not evidence of the original design assumptions. Without these details, the reader can describe the reported analysis and its limits, but cannot verify whether all analyses followed a prespecified plan or assess the impact of departures from it.

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 posted serious adverse event counts are zero in each dose group: 0 among 50 participants in the 1 mg/kg group, 0 among 35 in the 1.5 mg/kg group, and 0 among 40 in the 2 mg/kg group. These are counts within the group sizes listed for the safety outcome. A zero count means that no serious adverse events were recorded in those groups in the posted results; it does not establish that the underlying risk is zero or that rare events are ruled out.

The registry also reports other adverse-event counts of 5, 5 and 4 for the three groups, respectively. The serious-event and other-event categories should not be conflated. The record’s serious-event count is a limited summary, and the listed results do not provide event-specific descriptions or a comparative statistical analysis. The serious adverse-event terms are not reported in the registry record.

Safety counts should be interpreted with attention to the number of participants observed and the completeness of event reporting. The registry does not report detailed follow-up duration for safety, definitions for event categories, or whether participants with missing safety information were included. These limitations mean the counts describe the posted observations but do not establish a comprehensive comparison of safety across doses.

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

The posted primary results describe maximum sedation depth on an ordered scale. Medians were 5.5 for 1 mg/kg and 6.0 for both 1.5 mg/kg and 2 mg/kg, with the corresponding observed ranges shown in the outcome results. The reported Kruskal-Wallis test across the three groups had p = 0.2. Together, these findings describe high median scores and an overall test that did not provide strong evidence of a difference under the reported analysis.

That result does not prove that the doses are equivalent, nor does it demonstrate that they have identical effects. A non-significant test can reflect similar distributions, limited precision, or other features of the data. The registry does not provide a primary confidence interval or a prespecified equivalence margin, so the range of differences compatible with the primary results cannot be assessed in the same way as it could with an estimate and interval designed for that purpose.

The record also does not establish that one dose is preferable. The primary endpoint is maximum sedation depth, not a combined measure of depth, duration, additional medication, satisfaction, and harms. Some secondary outcomes have posted counts, while numerical results for others are not reported in the registry record. The secondary recalculations are unadjusted and should not be treated as confirmatory evidence for a general claim about dose.

More broadly, this is a comparison of three ketamine doses in a particular trial population and setting. The registry record does not provide enough detail about eligibility, procedures, or participant characteristics to define how far the results extend beyond the enrolled group. Statistical findings describe the outcomes recorded in the trial; they do not by themselves settle the broader clinical question of how to choose a dose in other circumstances.

14. Why This Trial Matters Statistically

This trial illustrates how an outcome’s scale shapes the statistical question. Sedation depth was recorded on the Ramsay Sedation Scale, an ordered measure with a defined upper limit. Medians and a rank-based omnibus test are natural summaries for ordered scores, but the concentration of medians at 6.0 also highlights a general limitation of bounded scales: once observations reach the maximum category, the measure cannot distinguish further differences in the underlying construct.

The registry also makes a useful distinction between an overall three-group primary test and pairwise secondary comparisons. The primary Kruskal-Wallis result addresses the group distributions collectively, whereas the additional-dose recalculations compare each higher-dose group with the 1 mg/kg group. Different outcomes, estimands and analysis methods cannot be collapsed into a single claim about dose. This matters when readers encounter multiple endpoints and see individual estimates alongside an overall primary result.

The posted record further shows why source and status of an analysis matter. The primary result is the sponsor-reported test; the additional-dose risk differences are unadjusted recalculations from posted counts. The sample-size figures are planning illustrations based on observed effects, not the trial’s original design assumptions and not observed power. Keeping these categories distinct helps prevent exploratory summaries from being mistaken for prespecified confirmatory results.

Finally, the incomplete reporting of some listed outcomes and of analysis-plan details limits what can be inferred. The record supports a clear description of the primary endpoint and selected event counts, but not a full reconstruction of the statistical plan or all secondary findings. That gap is important when judging precision, multiplicity, missing data, and generalizability.

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 limitations are specific to the information recorded for this trial. The primary endpoint is an ordinal score summarized by group medians, but the inter-quartile ranges stated in the endpoint definition are not reported. The registry does provide range limits, which do not convey the same information about the spread of the central observations. The primary analysis is an omnibus Kruskal-Wallis test, with no reported confidence interval, pairwise primary estimates, or fuller description of the score distributions.

The score has an upper bound of 6, and the reported median is 6.0 in both higher-dose groups. This raises a measurement consideration: a ceiling can obscure distinctions among observations at the top of a scale. The registry does not report analyses addressing this issue, nor does it provide participant-level data for evaluating the shape of the distributions.

The outcome-specific primary sample sizes are smaller than the enrollment total, but the record does not explain the difference. Missing-data handling and analysis populations are not reported. The registry also does not give detailed eligibility criteria, participant characteristics, procedures performed, allocation ratio, or the roles covered by quadruple masking. These omissions constrain assessment of risk of bias and applicability.

Reporting is incomplete across the listed secondary outcomes. Numerical results for sedation duration and satisfaction are not reported in the registry record. The additional-dose comparisons are unadjusted recalculations rather than sponsor-reported analyses, and no multiplicity method is reported. The registry also does not report a detailed protocol, sample-size justification, interim analyses, or prespecified analysis plan. These limitations do not negate the posted summaries, but they restrict the strength and breadth of conclusions that can be drawn from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Optimal Dosing of Ketamine for Procedural Sedation and Analgesia in Children — NCT02519595.
• Registry record last updated 2016-10-12; status last verified 2016-08.
• Results first posted 2016-10-12.
• 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

NCT02519595 is a completed Phase 4 randomized, quadruple-masked, parallel-group trial sponsored by Children’s Hospital of Michigan. It enrolled 182 participants and compared intravenous ketamine doses of 1 mg/kg, 1.5 mg/kg and 2 mg/kg for procedural sedation and analgesia in children in the emergency department. The recorded study start is 2010-08, and primary completion is 2012-08.

The primary endpoint was maximum sedation depth on the Ramsay Sedation Scale, assessed after study drug administration over an approximately 2-hour timeframe. The posted medians were 5.5, 6.0 and 6.0, respectively, and the reported Kruskal-Wallis comparison had p = 0.2. Secondary outcomes listed include sedation duration, additional dose, adverse events and sedation satisfaction. The record posts additional-dose counts and adverse-event counts, but does not report numerical results for every listed secondary outcome.

The registry does not report detailed analysis-population rules, missing-data handling, multiplicity procedures, or the original sample-size justification. The additional-dose estimates and associated planning illustrations are recalculations, not sponsor-reported primary analyses or observed power estimates.