Delirium Treatment With Acupuncture in Internal Medicine Departments: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Delirium Treatment With Acupuncture in Internal Medicine Departments (NCT03398928), a randomized na study. The primary outcome is Number of Delirium-free Days During the 7 Days of Evaluation.

2. Clinical Question

The clinical rationale in the record is that delirium frequently occurs in hospitalized older people and treatment options are limited. The investigators proposed that acupuncture might assist with delirium, drawing on its use for agitation in dementia as stated in the brief summary. That rationale is not itself evidence about the trial result. The trial enrolled 81 participants: 50 in the acupuncture group and 31 in the standard-care group. The reported primary comparison concerns delirium-free days, not a direct measure of recovery across every aspect of delirium or hospitalization. Other listed outcomes include length of hospital stay, functional status at discharge, delirium severity, antipsychotic use, pain, sleep, complications, and mortality. The record does not provide numerical results for most of those outcomes.

3. Trial Design

The registry classifies the study as randomized, parallel-group, and quadruple-masked. It records an enrollment of 81 and a completed status. The start date is 2018-01-14, and the primary completion date is 2019-12-01. The sponsor is Bnai Zion Medical Center. The phase is listed as NA.

The registered groups were acupuncture and standard care. The acupuncture arm is described as acupuncture for delirium treatment. Standard care is described as conventional delirium care at the discretion of department medical staff. The official title uses the wording sham-controlled, but the arm descriptions in the registry record identify standard care as the comparator and do not describe a sham arm. The record does not explain that difference in wording. It also does not specify which people were masked, the allocation ratio, or the treatment schedule.

The primary outcome was the number of delirium-free days during the 7 days of evaluation. The stated timeframe includes baseline, day 2, day 3, day 4, day 5, day 6 and day 7 of the study. The outcome description says that delirium status was assessed daily with the Confusion Assessment Method. It defines definite delirium using the first two of the tool’s four features, together with either the third or fourth feature. Subsyndromal delirium is described as not meeting those diagnostic criteria while displaying two or more of the features, including the acute and fluctuating course feature.

The registry also lists secondary outcomes covering hospitalization, function, delirium, medication use, pain, sleep, complications, and mortality. The record does not report detailed eligibility criteria, recruitment methods, a fuller analysis population definition, or a sample-size justification. Those omissions limit what can be concluded about how the enrolled participants were selected and how representative they may be.

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.

6. Primary Endpoint

The registry lists the primary outcome as Number of Delirium-free Days During the 7 Days of Evaluation, measured over At baseline, day 2, day 3, day 4, day 5, day 6 and day 7 of the study.

The primary endpoint is the number of delirium-free days during the 7 days of evaluation. The registry reports this as a median with a full range. Daily delirium assessment was based on the Confusion Assessment Method, a diagnostic tool addressing acute and fluctuating course, inattention, disorganized thinking, and impaired level of consciousness. The endpoint therefore summarizes the number of evaluated days on which a participant was classified as free of delirium under the study’s assessment approach.

A count of delirium-free days can represent the course of delirium over the evaluation period rather than relying on a single assessment. The repeated assessments are relevant because the endpoint description recognizes that delirium can have an acute and fluctuating course. This is a general statistical and clinical rationale for using repeated daily assessments; it does not establish that the endpoint captures every meaningful aspect of delirium or patient experience.

The registry reports medians of 5.5 days for acupuncture and 0.0 days for standard care, with a full range of 0.0 to 6.0 in each group. These group summaries do not by themselves show how individual participants’ values were distributed between the endpoints of the range. The record reports a Wilcoxon (Mann-Whitney) superiority test, but gives no primary comparison estimate or confidence interval. Thus, the reported test addresses evidence of a group difference without quantifying the size or precision of that difference as an effect estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison was analyzed with a Wilcoxon (Mann-Whitney) test. In general, this rank-based method compares the relative ordering of observations between groups. It can be useful when the outcome distribution does not fit the assumptions of a conventional comparison of means. It is not automatically a test of the difference between medians: a median-based interpretation requires additional distributional conditions that the registry record does not discuss. The method’s p-value describes the compatibility of the observed rank pattern with the test’s null hypothesis, not the size of the effect.

The endpoint is a count of delirium-free days, and the registry reports medians and full ranges. A median describes the middle of an ordered group of observations. A full range describes the lowest and highest reported values, but not how often each value occurred or how observations were distributed between those endpoints. Here the medians are 5.5 days for acupuncture and 0.0 days for standard care, and the full range is 0.0 to 6.0 in both groups. Because the record does not give participant-level data or a primary effect estimate, no additional distributional detail can be recovered from these summaries alone.

The reported primary p-value is p < 0.001. In general, a p-value measures how unusual the data, or data more extreme, would be under a specified null model and analysis procedure. It does not quantify the probability that the treatment has no effect. It also does not communicate precision in the way a confidence interval can. No confidence interval or numerical treatment-effect estimate is reported for the primary comparison, so readers cannot use the primary result to assess a range of plausible effect sizes from the registry record.

The record says the outcome was assessed daily using the Confusion Assessment Method, which addresses four features of delirium. Repeated assessment can capture a changing state rather than treating a single observation as the whole course. In general, repeated measures also raise analytic questions: how days are classified, how incomplete assessments are handled, and how correlated observations are incorporated into the endpoint. Here the registered primary outcome is the number of delirium-free days, but further details on missing assessments and derivation rules beyond the stated diagnostic definitions are not reported in the registry record.

The recalculated length-of-stay comparison uses group means and standard deviations. The group means are 16.4 days for acupuncture and 18.3 days for standard care; the standard deviations are 14.7 and 15.0 days, respectively. The unadjusted recalculated mean difference is -1.9 days, with a 95% confidence interval from -8.57 to 4.77 and p = 0.577. A mean difference expresses the contrast in group averages. A confidence interval describes a range of values compatible with the data and model at the stated confidence level; it is not a range containing a fixed proportion of individual participants. This interval spans both negative and positive differences, so the estimated direction is not precise.

That secondary comparison is labeled recalculated because it is a simple, unadjusted calculation from posted summary statistics. It is not a substitute for an adjusted sponsor analysis, and no sponsor-reported analysis for length of stay is listed in the registry record. The recalculation should be read as an approximate group comparison based on the reported means, standard deviations, and group sizes, not as a claim about the trial’s original analysis plan.

The sample-size illustrations are planning calculations based on the observed length-of-stay effect, not the trial’s own design justification. The stated assumptions are a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation. The illustrations consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. For 80% power, the corresponding figures are 955, 1,697, and 3,818 participants per arm; for 90% power, they are 1,278, 2,272, and 5,111 participants per arm. These are planning illustrations, not observed power and not evidence that the trial was designed around those assumptions. The registry record does not report its own sample-size justification.

In general, planning estimates depend strongly on the assumed effect size and model. Smaller target effects require larger samples when other assumptions are held fixed. An observed effect from a small trial can be unstable, so using it as the basis for future planning may be misleading. The illustrations should therefore be treated as conditional scenarios rather than definitive recommendations. They concern the length-of-stay recalculation, not the primary delirium-free-days outcome.

Multiplicity arises when multiple outcomes or comparisons are tested. In general, conducting many tests without adjustment increases the chance that at least one apparently notable result will occur by chance. The registry lists several secondary outcomes but does not report a multiplicity procedure or a testing hierarchy. As a result, unadjusted secondary p-values, where available, would carry less confirmatory weight than a result from a prespecified testing strategy. The numerical record does not provide results for most secondary outcomes.

Over-dispersion is a general concern for count outcomes when observed variability exceeds what a simple count model assumes. It can affect standard errors and confidence intervals if a model does not account for the extra variability. The primary analysis is rank-based, and the registry does not provide a count-regression model or information needed to assess over-dispersion. It is therefore not appropriate to assert that over-dispersion was present or that a particular adjustment was used.

10. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports the sponsor’s Wilcoxon (Mann-Whitney) superiority analysis with p < 0.001. The record does not provide a corresponding effect estimate or confidence interval. The primary result therefore supports a statistical distinction under the reported test, but does not quantify the size or precision of the difference in delirium-free days.

No recalculated primary comparison is presented. The endpoint is summarized by medians and full ranges, and the registry does not include the additional information needed to reproduce a primary effect estimate or its uncertainty from the posted summaries. The reported sponsor analysis is consequently the available primary inferential result.

The length-of-stay contrast is a separate, unadjusted recalculation from posted summary statistics. It gives a mean difference of -1.9 days, a 95% confidence interval from -8.57 to 4.77, and p = 0.577. The registry record lists no sponsor-reported analysis for that secondary outcome. The recalculation does not replace the primary analysis, and it should not be presented as if it were the sponsor’s own inferential result.

The discrepancy flag is false, so no disagreement between a sponsor analysis and a recalculation is identified. More broadly, the registry record does not provide a statistical analysis plan or enough detail to compare analytic choices beyond the reported primary test and the posted secondary summaries. The sponsor’s primary test and the recalculated length-of-stay contrast answer different endpoint questions and should be interpreted separately.

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary outcome and lists several secondary outcomes, but it does not report a multiplicity adjustment, a prespecified testing hierarchy, or a detailed statistical analysis plan. In general, testing multiple outcomes increases the chance of finding at least one statistically notable result by chance if each test is treated as an independent confirmatory test. Adjustments or a hierarchical strategy can control how evidence is interpreted across outcomes. Without a reported approach, secondary findings should be regarded as less confirmatory than the designated primary result.

The record reports a sponsor p-value for the primary comparison, but does not provide a primary effect estimate or confidence interval. It also does not report all secondary outcome results. This limits the ability to assess the overall pattern of evidence and how the analyses may have been prioritized. The registry reports 10 outcome measures and 4 analyses, but does not explain how the analyses map to a prespecified multiplicity strategy.

Details on missing data are not reported in the registry record. The primary endpoint depends on daily assessments, but the record does not say how an absent assessment was treated or whether participants with incomplete assessment schedules were included in the primary analysis. The record also does not report imputation methods, sensitivity analyses, or the number of participants with complete primary endpoint data. These omissions matter because missing assessments could affect the derived count of delirium-free days.

The record does not specify whether the primary analysis followed an intention-to-treat population, a per-protocol population, or another analysis set. It also does not describe post-randomization exclusions or participant flow by analysis. The group counts reported for the primary endpoint are 50 for acupuncture and 31 for standard care, matching the recorded enrollment total, but this alone does not establish how every participant was handled in the analysis.

Randomization can support a fair comparison when allocation is preserved in analysis, but the registry does not provide enough detail to assess deviations, missingness, or the analysis population. No assumptions should be made about the protocol’s plan, interim analyses, or sample-size justification because they are not reported in the registry record.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events among 50 participants at risk in the acupuncture group and zero among 31 participants at risk in the standard-care group. It also reports zero deaths in each group. No serious adverse event terms are listed.

For the separate other-adverse-event category, the record reports 9 affected participants among 50 at risk in the acupuncture group and 7 among 31 at risk in the standard-care group. These are participant counts in the registry’s category; the record does not provide event descriptions, timing, severity details, or an assessment of whether events were considered related to the intervention.

Safety counts are descriptive. A zero count of serious adverse events in the recorded groups does not establish that serious events cannot occur, and the available record does not give enough detail to assess risks beyond the observed trial experience. Differences in the raw counts should not be interpreted as comparative risk estimates without a suitable analysis and fuller information on event definitions and follow-up. The registry record does not report those details.

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

The posted primary result is a statistically notable rank-based comparison of delirium-free days: the acupuncture group’s reported median is 5.5 days and the standard-care group’s is 0.0 days, with p < 0.001. This describes the recorded trial result under the sponsor-reported test. It does not, by itself, give a precise estimate of the size of the between-group effect, because no primary effect estimate or confidence interval is reported.

The result does not establish what would happen in every clinical setting or for every person with delirium. The registered standard-care description allows care at the discretion of department medical staff, and the record does not give detailed eligibility criteria or treatment procedures. The finding should therefore be understood in the context of the groups and methods recorded, rather than generalized beyond them without further evidence.

The endpoint reflects delirium-free days during the evaluation period, based on daily Confusion Assessment Method assessments. It does not directly answer every question about hospital stay, function, discomfort, sleep, medication exposure, complications, or longer-term outcomes. Those matters are represented among the listed secondary outcomes, but numerical results for most are not reported in the registry record. The available length-of-stay recalculation is imprecise and unadjusted, and it should not be treated as a definitive confirmation or refutation of the primary result.

The registry reports no serious adverse events and no deaths in either group, alongside other adverse-event counts. Those observations describe the recorded trial experience, not a guarantee of safety or a full characterization of possible harms. The record does not give detailed event terms or relatedness assessments.

Overall, the available data support a difference in the reported primary test, while leaving important questions about effect size, precision, missing assessments, analysis populations, and secondary outcomes unresolved. Statistical significance and clinical importance are distinct concepts; the registry record does not define a threshold for clinical importance or provide enough information to determine it.

14. Why This Trial Matters Statistically

This trial illustrates the distinction between a p-value and an effect estimate. The primary analysis reports p < 0.001, while the registry does not report a primary between-group estimate or confidence interval. Readers can identify evidence of a group difference under the reported test, but cannot assess the magnitude and precision of that difference from the primary result alone.

It also illustrates why outcome summaries and analysis methods need to be read together. The primary measure is a count of delirium-free days, summarized by medians and full ranges and compared with a rank-based test. Those summaries offer a useful description of each group’s central value and endpoints, but they do not show the complete distribution or quantify the contrast with uncertainty.

The secondary length-of-stay result highlights a different issue: a recalculated estimate can be informative while remaining unadjusted and imprecise. Its confidence interval spans negative and positive mean differences, so the point estimate should not be treated as a settled direction of effect. Planning illustrations based on that observed estimate are scenarios, not evidence of the trial’s original design adequacy or observed power.

Finally, the record leaves important reporting gaps. It does not describe missing-data handling, analysis populations, multiplicity procedures, or a sample-size justification. The majority of listed secondary outcomes lack numerical results. Recognizing these limits is part of sound interpretation: a statistically notable primary test can coexist with uncertainty about effect size, robustness, and how the finding relates to outcomes beyond the primary endpoint.

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 registry record provides limited detail about the enrolled population. It identifies the condition as delirium in old age and describes hospitalized older people, but detailed eligibility criteria, age thresholds, baseline characteristics, and recruitment information are not reported in the registry record. This makes it difficult to judge how closely the enrolled group represents other populations or care settings.

The record identifies the study as quadruple-masked, but does not state which groups of people were masked or how masking was maintained. Although the official title uses sham-controlled wording, the registered arm descriptions compare acupuncture with standard care, and no sham arm description is given. The record does not explain the difference. The standard-care arm is described as conventional delirium care at the discretion of department medical staff, which leaves the comparator’s content open to variation.

The primary outcome is described, but no primary effect estimate or confidence interval is reported. The registry also does not report details on missing daily assessments, the analysis population, post-randomization exclusions, or sensitivity analyses. These omissions limit assessment of how robust the primary result is to incomplete data or analytic choices.

Numerical results for most secondary outcomes are not reported in the registry record. The available length-of-stay recalculation is unadjusted and has a broad confidence interval. The record does not report a multiplicity strategy or a detailed analysis plan, and it does not give a sample-size justification. These limitations constrain both the strength of secondary conclusions and the ability to evaluate how the trial’s analyses were planned and interpreted.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Delirium Treatment With Acupuncture in Internal Medicine Departments — NCT03398928.
• Registry record last updated 2020-09-18; status last verified 2017-12.
• Results first posted 2020-09-18.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes NCT03398928, titled Delirium Treatment With Acupuncture in Internal Medicine Departments. It lists a randomized, parallel-group, quadruple-masked study with 81 participants and completed status. The recorded groups are acupuncture, with 50 participants, and standard care, with 31 participants.

The primary endpoint is the number of delirium-free days during the 7 days of evaluation, assessed with the Confusion Assessment Method. The group medians are 5.5 days for acupuncture and 0.0 days for standard care; both groups have a full range of 0.0 to 6.0. The reported Wilcoxon (Mann-Whitney) superiority analysis has p < 0.001. No primary effect estimate or confidence interval is reported.

Length of hospital stay is the secondary outcome with a numerical recalculation: the unadjusted mean difference is -1.9 days, with a 95% confidence interval from -8.57 to 4.77 and p = 0.577. Numerical results for most other listed secondary outcomes are not reported in the registry record. The record also reports zero serious adverse events and zero deaths in each group, with other adverse events reported for 9 participants in acupuncture and 7 in standard care.