Evaluating the Role of Expectations in Response to Caffeine Consumption: An RCT: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluating the Role of Expectations in Response to Caffeine Consumption: An RCT (NCT02461693), a randomized na study. The primary outcome is Mood State Score on POMS-2 Test.

2. Clinical Question

The design aims to study expectation as well as pill assignment. Participants were randomly assigned a probability of receiving caffeine versus placebo, and that probability was disclosed before pill consumption and subsequent cognitive testing. The pill itself was assigned according to predetermined randomization probabilities. The registry description states that neither participants nor staff administering the intervention would know for certain which pill a participant received at the time of consumption.

This setup creates a question beyond whether the caffeine and placebo groups differ. It also asks whether the expectation induced by the disclosed probability is related to mood and cognitive outcomes, and whether considering those expectations offers a different estimate of the caffeine pill’s effect. These are aims described in the registry summary, not conclusions established by the posted results. Arm-specific outcome estimates and an analysis of expectancy are not reported in the registry record, so the size and precision of any such relationship cannot be assessed from that record.

3. Trial Design

The registry describes a randomized, double-masked, parallel-group trial. It lists enrollment of 205 and a completed status. The recorded study dates are 2015-06 for study start and 2015-10 for primary completion. The phase is listed as NA. The sponsor is the University of Alabama at Birmingham.

Participants received either a one-time 200mg caffeine pill or a one-time lactose-based placebo pill. Before consumption, each participant was told a probability, ranging from 0-100%, of receiving caffeine rather than placebo. The study summary says pill assignment depended on predetermined randomization probabilities assigned by the study statistician. In this setup, a participant could receive a disclosed probability that altered expectations without knowing for certain which pill was actually administered.

The registry labels masking as double. Its description specifically says that participants and intervention-administering staff would not know for certain which pill a participant received at the time of consumption. It does not further describe masking for other study roles. The record’s description therefore supports a cautious account of who was masked rather than assumptions about every person involved in the trial.

The trial sought to estimate effects of expectancy on cognitive and affective outcomes and to examine caffeine effects under circumstances in which participants had information about their chance of receiving caffeine. The registry lists four outcome measures, but the posted results do not reduce the primary outcome to one comparison. Detailed randomization procedures, allocation concealment, a protocol, and a statistical analysis plan are not reported in the registry record. The enrollment figure alone does not establish whether recruitment matched a prespecified target or whether the study had adequate precision for any particular outcome.

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

Analysis population for the primary outcome, as defined in the registry: Results are of raw data as presented in our publication

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 Mood State Score on POMS-2 Test, measured over 30 minutes post pill consumption.

The listed primary outcome is mood state score on the Profile of Mood States, or POMS-2, measured 30 minutes after pill consumption. It is recorded as a continuous outcome summarized by a mean, with standard error as the dispersion measure. A continuous score can retain more information about variation in mood than a simple category such as improved or not improved.

The instrument description says participants rated 65 mood-related adjectives according to how they felt right now, using a scale from 0 (not at all) to 4 (extremely). The adjectives contribute to six mood subscales: Tension-Anxiety, Depression-Dejection, Anger-Hostility, Vigor-Activity, Fatigue-Inertia, and Confusion-Bewilderment. The description also includes Total Mood Disturbance, which aggregates the six subscales, and Friendliness. It reports raw-score ranges for these measures, but does not identify one of them as the sole reported primary comparison.

Direction matters when interpreting the scores. Higher Friendliness and Vigor-Activity scores indicate more positive feelings; higher scores for the other subscales and Total Mood Disturbance indicate more negative feelings. Because the registry lists multiple mood dimensions and the posted results do not resolve them into a single comparison, no specific direction or size of a caffeine-related mood difference can be stated from this record. The endpoint is relevant to the stated question because it measures self-reported mood shortly after the assigned pill, but it does not by itself establish how expectations affected cognitive performance.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies a randomized, parallel-group design and reports a mean as the parameter for the continuous primary outcome. Randomization is intended, in general, to make treatment assignment independent of participant characteristics at the point of assignment. That design feature can support comparisons between assigned groups, but it does not guarantee that groups will be identical in every respect or that the realized estimates will be precise. The registry does not report detailed allocation procedures, sequence generation, or concealment methods.

What does the mean represent?

For a continuous outcome such as a mood score, a group mean is the average score among the participants included in the analysis. A difference between group means can describe an unadjusted contrast, while a model-based estimate may account for other variables or the trial’s expectancy structure. The registry record does not report the group means, a contrast, or a model-based estimate. It also does not specify whether the primary analysis was adjusted or unadjusted. Accordingly, no treatment contrast can be interpreted from the listed endpoint definition alone.

Standard error is the dispersion measure named for the primary outcome. In general, it describes the sampling uncertainty of an estimated mean, rather than the spread of individual scores. A standard deviation, by contrast, describes variability among individual observations. These quantities answer different questions. The record names standard error but does not report an actual standard error for any POMS-2 score, so the precision of the mood estimate cannot be evaluated.

How should uncertainty be assessed?

As a general statistical point, a confidence interval gives a range of estimates compatible with the data under the specified analysis and its assumptions. Its width reflects uncertainty: a wider interval indicates less precision than a narrower interval. An interval also helps readers assess which effect sizes remain plausible, instead of relying only on whether a test crosses a significance threshold. No confidence interval is reported for the primary outcome in the registry record, so neither the range of plausible effects nor the precision of an estimate is available.

A p-value, when reported, assesses how compatible data are with a stated null hypothesis under the test’s assumptions. It is not the probability that the null hypothesis is true, nor does it communicate the size or practical importance of an effect. The registry does not report an inferential test or p-value for the primary outcome. Statistical significance and practical meaning therefore cannot be judged from the available outcome record.

Why do expectations complicate the comparison?

The distinctive feature of the study is that a probability of caffeine receipt was disclosed before the pill was taken. In general, the disclosed probability may influence a participant’s expectation, while the assigned pill represents the actual treatment received. Estimating the separate contribution of expectation and pill assignment requires an analysis that makes those quantities explicit. For example, an analysis might compare outcomes across expectation conditions, pill assignments, or their combination; the appropriate specification depends on the design and prespecified question. The registry describes the manipulation but does not report an analysis model or estimates for these contrasts. No particular model or causal estimate should be attributed to the trial based only on the design description.

The record says assignment depended on predetermined randomization probabilities and that participants were informed of their individual probability. This is not the same description as a conventional trial in which participants are simply told nothing about their chance of receiving the active pill. At the same time, the record says participants and administering staff did not know for certain which pill was given. That uncertainty is part of the described design, but the exact probability assignments, allocation proportions, and operational details are not reported in the registry record.

What do the planning assumptions mean?

The sample-size assumptions are explicitly described as a planning illustration based on the observed effect, rather than the trial’s own design. They specify a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is expressed as a stated fraction of the observed effect. Such illustrations estimate how many participants per arm would be needed under the stated assumptions for a chosen power level. They are conditional planning exercises, not evidence that the trial had that power, and they are not observed power. Numerical planning results and a detailed sample-size justification are not reported in the registry record. They should not be treated as the trial’s prespecified enrollment target.

For a continuous outcome, an illustrative sample-size calculation generally depends on the effect size to detect, the outcome’s variability, the significance level, desired power, and allocation assumptions. If the underlying effect estimate is uncertain, planning values based on it are also uncertain. The registry’s stated assumptions do not report a complete design rationale, nor do they establish that the trial was planned around the listed primary endpoint. No further assumptions should be inferred.

Finally, the registry’s primary-outcome entry includes multiple mood dimensions and indicates that the outcome has several categories or time points rather than one reduced comparison. As a general matter, testing multiple outcomes or contrasts raises the possibility of chance-positive findings across the analysis family. A prespecified multiplicity strategy can address that issue, but no strategy is reported in the registry record. Without outcome estimates, a model, uncertainty intervals, and analysis details, statistical interpretation remains limited to the design and outcome definitions recorded.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures, while the primary outcome description includes several mood subscales and a Total Mood Disturbance score. The record flags that the primary outcome has several categories or time points and is not reduced to one comparison. The number of analyses is listed as zero. This does not establish whether no analyses were performed; it indicates that analysis results are not reported in the registry record.

When multiple outcomes, subscales, time points, or contrasts are examined, there are more opportunities for chance findings. As a general statistical principle, a prespecified hierarchy or multiplicity adjustment can help control how results are interpreted across an analysis family. The registry record does not report whether the investigators used such a method, which outcome was prioritized among the POMS-2 dimensions, or whether the analysis plan changed. It is therefore not possible to assess how selective reporting or multiple testing may affect interpretation.

Missing-data details are also not reported in the registry record. The record does not give the number of participants with observed POMS-2 data, reasons for missing measurements, or methods used to handle missing outcomes. It does not specify whether participants were analyzed according to their randomized assignment, only if they completed testing, or under another analysis population definition. There is no reported intention-to-treat, per-protocol, or safety population description for the mood analysis.

The primary outcome’s population description says results are raw data as presented in a publication, but the registry record does not provide the numerical primary-outcome estimates or a full account of the analysis population. The adverse-event risk counts do not resolve missingness for the mood or cognitive outcomes. Without these details, readers cannot determine whether outcome availability differed across groups or how any such differences were addressed.

11. 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.

The registry records zero serious adverse events for the Caffeine group, with 103 participants at risk, and zero for the Placebo group, with 102 participants at risk. It also records zero other adverse events for each group. These are the adverse-event counts entered in the registry record; they should be read as reported counts rather than as proof that no symptoms or health events occurred.

The death counts are null and therefore not reported in the registry record. The entry does not provide descriptions of non-serious events, event definitions, severity grading, assessment procedures, or the period over which adverse events were collected. It also does not explain whether counts refer to participants with events or event episodes beyond the displayed categories. No further safety profile can be reconstructed from these entries.

Zero recorded serious adverse events does not establish comparative safety in a broader population, and it does not address the trial’s mood or cognitive outcomes. The group counts are safety information, not an estimate of the effect of caffeine on POMS-2 scores. The registry’s safety record should therefore be interpreted narrowly: it records zero serious and zero other adverse events in each group, with death information not reported.

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

The record establishes that a randomized, double-masked, parallel-group trial was conducted to examine caffeine, placebo, and expectations about receiving caffeine. It documents a one-time 200mg caffeine pill and a lactose-based placebo pill, along with disclosure of an individual probability of caffeine receipt before pill consumption. It also identifies a mood measure at 30 minutes after consumption as the primary outcome. These details describe the study question and its structure.

The record does not provide a consolidated primary-outcome comparison, arm-specific mood means, standard errors, confidence intervals, or an inferential test. Therefore, it does not show whether mood scores differed between the assigned groups, how large any difference was, or how precisely an effect was estimated. Nor does it establish whether disclosed expectations were associated with mood outcomes or altered a caffeine estimate. Those are stated aims, not results demonstrated by the available outcome entries.

The POMS-2 description covers several mood dimensions, and score direction differs by measure. Higher Vigor-Activity and Friendliness scores indicate more positive feelings, while higher scores on the other listed dimensions indicate more negative feelings. Without the actual estimates and a clearly specified comparison, a statement that scores rose or fell would not have an interpretable meaning across all subscales.

The recorded absence of adverse events should also be kept distinct from efficacy or mood interpretation. It describes adverse-event entries, not an outcome comparison, and death counts are not reported in the registry record. The trial’s enrollment figure and completion status do not by themselves establish the precision of a result or the completeness of outcome reporting.

In short, the registry record supports an account of what the trial set out to investigate and how the pills and expectation information were described. It does not support a conclusion about the direction, magnitude, statistical significance, or clinical importance of the primary mood result. Those matters remain unresolved in the posted results.

13. Why This Trial Matters Statistically

This trial is statistically notable because its design treats expectation as part of the research question rather than only as a possible influence to eliminate. Participants were told their individual probability of receiving caffeine before the pill was taken, while neither participants nor staff administering the intervention knew for certain which pill was given. The design therefore distinguishes information about the chance of treatment from certainty about the actual pill.

That distinction makes the target comparison important. A comparison based only on pill assignment would address a different question from one that also examines the disclosed expectation. In general, interpreting a study with both elements requires a clear account of which contrast estimates the effect of assignment, which addresses expectation, and how those factors are handled together. The registry describes the purpose of estimating expectancy effects but does not report those analytical contrasts or their estimates.

The trial also illustrates why outcome definitions need to be specific. POMS-2 is represented by multiple subscales and an aggregate mood-disturbance score, with different directions of interpretation for positive and negative dimensions. A single phrase such as mood change is not enough to identify which score is being compared or what a higher value means. The registry’s failure to reduce the primary outcome to one comparison limits the interpretability of the posted results.

More broadly, the record shows the difference between a well-described research aim and an evaluable result. The trial design and endpoint timing are documented, but numerical outcome estimates and analysis details are absent. Statistical interpretation depends not only on randomization and masking but also on transparent estimates, uncertainty, outcome selection, and analysis populations. Without those pieces, readers can understand the question but cannot assess the answer.

14. 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 central limitation for interpretation is the absence of a single primary-outcome comparison. The registry lists mood state score on POMS-2 at 30 minutes post pill consumption, but also identifies multiple subscales and score categories and does not provide arm-specific numerical estimates. The record therefore does not permit assessment of effect direction, magnitude, or precision.

Outcome reporting is incomplete for statistical appraisal. The registry lists four outcome measures and zero analyses, but does not report a statistical model, confidence intervals, p-values, or an approach to multiplicity. It does not specify how the expectancy information and pill assignment were jointly analyzed. These omissions leave the stated expectancy objective without a reported numerical estimate in the registry record.

The record does not report demographic characteristics, eligibility criteria, detailed recruitment information, or participant flow. It gives enrollment, but not the analysis population for mood outcomes or the number with complete measurements. Missing-data extent, reasons, and handling are not reported in the registry record. These gaps limit assessment of how representative the analyzed group was and whether outcome availability could have affected comparisons.

The registry also does not report a protocol, statistical analysis plan, sample-size justification, or multiplicity plan. The sample-size assumptions describe a planning illustration based on the observed effect, not the trial’s own design, and do not establish observed power. The record’s adverse-event counts are limited: deaths are not reported, and event definitions and collection details are absent.

Finally, the intervention is described as a one-time pill and the primary mood assessment occurs 30 minutes after consumption. The record does not provide longer-term outcomes or detailed information about the setting and participant characteristics. Conclusions should remain tied to the documented study question and timing, without extrapolating beyond the reported results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Evaluating the Role of Expectations in Response to Caffeine Consumption: An RCT — NCT02461693.
• Registry record last updated 2017-07-12; status last verified 2017-06.
• Results first posted 2017-07-12.
• 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.

18. Record Summary

The registry record describes a completed randomized, double-masked, parallel-group trial sponsored by the University of Alabama at Birmingham, with enrollment of 205. Participants were assigned a disclosed probability of receiving caffeine and then received a one-time 200mg caffeine pill or a lactose-based placebo pill. The primary outcome is POMS-2 mood state score 30 minutes after pill consumption.

The trial’s stated aims include examining expectancy effects on mood and cognitive outcomes. The registry describes multiple POMS-2 dimensions but does not reduce the primary outcome to one comparison. Numerical mood results, analysis details, and missing-data methods are not reported in the registry record. The registry records zero serious and zero other adverse events in each group; death counts are not reported.