Dronabinol Interactions With Cognitive Enhancing Drug in Humans: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Dronabinol Interactions With Cognitive Enhancing Drug in Humans (NCT00842985), a registered early/phase 1 study. The primary outcome is CANTAB:CAmbridge Neuropsychological Test Automated Battery RVIP: Rapid Visual Information Processing.

2. Clinical Question

3. Trial Design

This was an early-phase 1 study with a sequential model and no masking. The registry lists enrollment as 53 and status as completed. The recorded start was 2008-09, and primary completion was 2009-10. The sponsor was Yale University. Cannabis and marijuana abuse are listed as conditions.

The trial description says participants received each drug condition in sequential order across 4 test days, with the order varying among participants. Allocation is not stated, and the registry flags that the comparison is not randomized and that the arms are not like-for-like controls. Those features matter when interpreting any contrast: condition differences could be entangled with the sequence or timing of sessions, and the record does not establish that groups were formed to support a controlled causal comparison.

The summary identifies modafinil as the pharmacotherapy under study and describes hypotheses about drug liking and cognitive measures. However, the intervention listing is only “drug condition.” The particular conditions, doses, and detailed schedule are not reported in the registry record. The registry also does not report the protocol details needed to determine how treatment order was assigned or how possible order effects were handled.

The primary-outcome population is described separately from the enrollment figure: it includes subjects who completed all four treatment sessions, with three of these 15 subjects excluded from analysis because of validity concerns for cognitive data from at least one session. The registry does not reconcile these outcome-population details with the enrollment figure. Enrollment therefore should not be read as the number contributing to the primary-outcome analysis.

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.

Screening

Reasons for not completing this period:

First Intervention

Reasons for not completing this period:

Washout Period of GE 4 Days

Reasons for not completing this period:

Second Intervention

Analysis population for the primary outcome, as defined in the registry: Subjects who completed all four treatment sessions. Three of these 15 subjects were not analyzed due to validity concerns of the cognitive data for at least one of the four testing sessions.

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 CANTAB:CAmbridge Neuropsychological Test Automated Battery RVIP: Rapid Visual Information Processing, measured over Once for each test session (4 total)..

The primary outcome is CANTAB RVIP, a component of the Cambridge Neuropsychological Test Automated Battery. The registry describes it as a measure of sustained attention with a working-memory component. It reports two subscales: RVP A', target sensitivity, which measures the ability to detect sequences; and RVP B'', response bias, which reflects the tendency to respond regardless of whether a target is present. The listed scales are probabilities expressed in units on a scale.

RVP A' ranges from 0-1, with lower values described as bad and higher values as good. RVP B'' ranges from -1 to +1, also described as bad to good. The two subscales represent different aspects of task performance: detecting target sequences and response tendency. A change in one should not automatically be interpreted as a change in the other.

The registry lists the outcome as a continuous measure summarized by a mean and standard deviation. It specifies measurement once for each test session, across 4 sessions. Such measures can be useful for describing cognitive task performance, but an endpoint definition alone does not show whether scores changed across conditions or whether any difference was precise. The registry record contains no outcome means, standard deviations, or comparison estimates, so the magnitude and uncertainty of performance differences are not reported.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help clarify what the registry record does and does not establish. These are general explanations, not descriptions of methods confirmed to have been used in this study. The record identifies an outcome format and a sequential design, but it does not report an analysis plan or analysis entries.

What does a mean and standard deviation describe?

For a continuous outcome, the mean describes the central level of the observed scores, while the standard deviation describes how spread out scores are around that mean. They answer different questions: a mean is about the average score, and a standard deviation is about variation among scores. Neither alone establishes a treatment effect. To assess a contrast, readers also need a clearly defined comparison and an estimate of its uncertainty. The registry specifies mean and standard deviation as the outcome summary, but posts no values for either RVIP subscale.

Why does the sequence matter?

When observations are collected across sessions in a sequence, the observations from a participant may be related rather than independent. General statistical methods for repeated measurements account for that dependence. Session order can also matter: task familiarity, changes over time, or effects carried forward from an earlier condition could influence later measurements. These are general possibilities, not claims that such effects occurred in this study. The record says condition order varied among participants but does not report how the order was assigned, whether sequence was incorporated into an analysis, or how potential carryover was assessed.

A randomized, like-for-like comparison can help separate treatment effects from other differences between comparison groups. This registry record does not describe a randomized comparison and explicitly warns that the arms are not like-for-like controls. That does not make descriptive statistics meaningless, but it does limit the strength of causal conclusions that could be drawn from a contrast. The record does not provide an adjusted or unadjusted estimate with which to examine this issue.

How should uncertainty be assessed?

In general, a confidence interval describes a range of values compatible with the data and the method used to calculate it. Its width conveys precision: a wider interval indicates less precision than a narrower interval. A p-value, when reported, addresses how compatible the data are with a specified null model; it does not measure the size or clinical importance of an effect. These measures complement an effect estimate rather than replace it. Neither confidence intervals nor p-values are reported for the primary outcome in the registry record, so neither statistical precision nor evidence against a null hypothesis can be evaluated.

For repeated cognitive scores, the choice of analysis may also depend on the structure of the data, the treatment sequence, and the question of interest. A model could target within-person changes or contrasts between conditions, but the appropriate formulation depends on design details and assumptions. The registry does not report a model, covariates, a contrast definition, or an estimation method. It would be inappropriate to infer that a particular repeated-measures, paired, or adjusted analysis was used.

What do the planning illustrations mean?

The 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 effect to detect is described as a stated fraction of the observed effect. For non-time-to-event outcomes, the stated convention is participants per arm; time-to-event calculations are described differently as total events. These assumptions explain a planning calculation, not the trial’s own design. Such figures are not observed power and do not demonstrate that the study had a particular probability of detecting an effect. The registry record does not provide a sample-size justification for the trial or analysis results that could substantiate a trial-specific power claim.

Over-dispersion is a general concern mainly for count outcomes when observed variability exceeds what a simple model assumes. The primary outcome is continuous and reported as a mean with standard deviation, so count-model over-dispersion is not the central issue identified by the record. More relevant questions include within-participant dependence across sessions, validity of cognitive scores, session order, and the handling of missing measurements. The registry does not explain how those issues were modeled or checked.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("THC+Modafinil", "THC+Placebo", "Pla+Modafinil", "Placebo+Placebo")
n   <- c(12, 12, 12, 12)
m   <- c(0.91, 0.9, 0.92, 0.9)
sp  <- c(0.1, 0.1, 0.1, 0.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  THC+Modafinil  12  0.910  0.100  0.846  0.974
CARM  THC+Placebo  12  0.900  0.100  0.836  0.964
CARM  Pla+Modafinil  12  0.920  0.100  0.856  0.984
CARM  Placebo+Placebo  12  0.900  0.100  0.836  0.964

11. Multiplicity, Missing Data and Analysis Populations

The primary-outcome description includes two RVIP subscales, RVP A' and RVP B''. The registry lists one outcome measure, but it does not explain whether the subscales were treated as separate inferential tests, whether one was designated as more important, or whether any adjustment for multiple testing was planned. It also does not report an analysis plan or analysis entries. Therefore, the approach to multiplicity is not reported in the registry record. In general, testing several outcomes or contrasts can increase the chance of at least one apparently notable result; an analysis plan can specify how that issue is handled.

The recorded population includes subjects who completed all four treatment sessions. The registry states that three of these 15 subjects were not analyzed because validity concerns affected cognitive data for at least one session. This description identifies a reason for exclusion but does not say how invalid measurements were defined, whether exclusions were decided without knowledge of condition results, or whether any sensitivity analyses were performed.

Handling of missing or unusable measurements is not reported in the registry record. There is no stated imputation method, missing-data assumption, or analysis including incomplete sessions. The record also does not define a formal intention-to-treat or per-protocol analysis population. The enrollment figure of 53 and the outcome-population description are distinct registry statements, and the record does not reconcile them. The number contributing to any analysis should not be inferred from the enrollment figure or by subtracting the listed exclusions.

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.

For serious adverse events, the registry lists the group as All Participants, with 26 at risk and 0 affected. This is the posted serious-event count for that group. It describes the registry’s recorded count and denominator; it does not establish that serious events were impossible or that the data capture was complete.

No serious adverse-event terms are listed. The separate fields for other adverse events show 0 at risk and 0 affected. Those fields do not provide a usable denominator for assessing the frequency of other adverse events, and they should not be interpreted as evidence that no non-serious adverse events occurred among all participants. The death field is null, rather than a reported zero, so it does not supply a numerical death count.

Safety counts are descriptive. They do not by themselves establish comparative safety, particularly when the registry does not report a randomized, like-for-like control comparison. The record does not give event timing, severity details beyond the serious-event category, or an explanation of how events were collected. These limitations constrain interpretation of the posted figures.

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

The registry record establishes that the study was completed, used sequential drug conditions, and identified CANTAB RVIP as its primary cognitive outcome. It also documents the intended hypothesis concerning modafinil, drug liking, and cognitive measures. These are useful descriptions of the study’s question and structure, but they are not evidence of an observed benefit or lack of benefit.

No numerical primary-outcome results or analysis entries appear in the registry record. It therefore cannot show whether RVIP scores differed across conditions, how large any difference was, or how precisely it was estimated. The absence of a posted estimate is not evidence that the estimate was zero. Likewise, a stated hypothesis does not establish that the hypothesized result occurred.

The design description also does not support treating any potential condition contrast as a randomized treatment effect. The order was sequential and varied, allocation is not stated, and the registry warns that the arms are not like-for-like controls. General concerns about order and time effects would need to be addressed in a reported analysis, but the record does not say whether they were evaluated.

The serious adverse-event entry records 0 affected among 26 at risk. This should be read as a registry count, not as a guarantee of safety or a comparative safety conclusion. The other adverse-event fields and null death field have different meanings and should not be converted into claims that no other events or deaths occurred.

Overall, the record supports a description of the study’s registered question, outcome, and design, but not a conclusion about the effect of the drug conditions on cognitive performance. It also does not permit an assessment of statistical significance, precision, or the clinical importance of any difference.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of why design and results need to be considered together. A cognitive endpoint may be clearly defined, with subscales and score ranges, yet still provide no basis for evaluating an effect if outcome estimates and analyses are absent. The RVIP description tells readers what was intended to be measured; it does not reveal how participants performed.

The sequential structure is also important. When people complete conditions across sessions, observations may be related within participant, and their order may matter. These are general design considerations, not evidence of bias in this particular study. Because allocation and the handling of order are not reported, the registry record cannot show whether a comparison separates condition effects from timing or sequence effects.

The record also distinguishes enrollment from the population defined for the outcome. It names a completed-session population and reports validity-related exclusions within that group, but it does not reconcile those details with enrollment. That makes clear why analysis populations and exclusions should be documented explicitly: readers need to know which observations contributed to each result and how that set was selected.

Finally, the planning assumptions are not substitutes for a reported analysis or a trial-specific sample-size rationale. Planning illustrations can describe how a future study might be designed under assumptions, but they cannot reveal what this study observed. The registry record demonstrates the value of separating a study’s hypothesis, measurement plan, analytic evidence, and safety reporting rather than treating them as interchangeable.

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 record provides little detail about the drug conditions. Although the summary names modafinil as the pharmacotherapy under study, the intervention listing is only “drug condition.” Specific conditions, doses, and treatment schedules are not reported in the registry record. This limits the ability to understand precisely what was compared.

Allocation is not stated, masking was none, and the record flags that the comparison was not randomized and the arms were not like-for-like controls. The sequential order varied among participants, but the method for assigning order and the handling of possible period or carryover effects are not described. These design features limit causal interpretation of any condition contrast.

The primary-outcome population consists of session completers, and three of these 15 subjects were excluded for validity concerns affecting cognitive data from at least one session. The criteria for those concerns, the treatment of incomplete measurements, and the relationship between this population and the enrollment figure of 53 are not explained. No formal analysis population or missing-data method is reported.

The record identifies two RVIP subscales but does not report how multiplicity was handled. It also contains no analysis entries and no numerical endpoint results, estimates, or uncertainty measures. Therefore, neither the size nor precision of any cognitive difference can be assessed. The serious adverse-event entry is limited to its posted count and denominator; event details and the meaning of the null death field are not expanded upon.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Dronabinol Interactions With Cognitive Enhancing Drug in Humans — NCT00842985.
• Registry record last updated 2017-05-30; status last verified 2017-04.
• Results first posted 2012-12-17.
• 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

NCT00842985 is a completed early-phase 1 study sponsored by Yale University. It used a sequential model without masking; participants received drug conditions across 4 test days, with order varying among participants. Enrollment is listed as 53, while the primary-outcome population is described as session completers, with validity-related exclusions noted.

The primary outcome is CANTAB RVIP, summarized as a continuous measure using means and standard deviations. It includes RVP A' and RVP B'' subscales. The registry record contains no analysis entries or numerical outcome results. It reports 0 serious adverse events among 26 at risk, with no serious-event terms listed; the death field is null.