The Effects of Smell on Mood and Physical Responses: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effects of Smell on Mood and Physical Responses (NCT00097253), a randomized phase 1 study. The primary outcome is Cortisol and Catecholamine Production.

2. Clinical Question

In statistical terms, the question concerns whether measured physiological responses differ across odor conditions and around the stressor. The record identifies the biomarkers and planned measurement times, but does not provide a single numerical estimate of an odor-related difference. It also does not report a specific primary contrast, such as lavender versus water, or explain how the several biomarkers and time points were combined or compared. Therefore, the clinical question can be described from the stated objective and endpoint, but the size and precision of any observed response cannot be assessed from the registry record.

3. Trial Design

This is a Phase 1, randomized, double-masked crossover study listed as completed. The registry reports enrollment of 56 participants and names the National Center for Complementary and Integrative Health as sponsor. The study’s stated purpose is to examine the body’s response to relaxing and stimulating fragrances commonly used in aromatherapy.

The intervention is behavioral exposure to relaxant and stimulant odors. Lavender, citrus, and water are identified in the record, with water described as the control condition. A crossover design generally means that participants receive more than one condition, so comparisons can use measurements from the same participant. That is a general explanation of the design label; the registry record does not report the actual treatment sequences, the order of exposure, or whether every participant completed every condition. It also does not report details of masking or how masking was maintained.

The primary outcome is cortisol and catecholamine production, expressed as a mean with standard deviation in pg/ml (log 10). The described biomarkers are cortisol, norepinephrine, and epinephrine. Measurements were scheduled before and after a physical cold pressor stressor that occurred at 11:00. The record lists three visits with at least 2 weeks between each, and an average time to complete all visits of 64.46 days (SD 48.4).

The registry lists four outcome measures and seven analyses. It does not report a numerical primary-outcome estimate or identify a single summary comparison in the information available for this analysis. The specific sample-size justification, randomization procedure, allocation sequence, and statistical analysis plan are not reported in the registry record. The dates listed are a study start of 2005-08 and primary completion of 2006-03.

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: We excluded individuals taking cardiovascular medications or medications with obvious immunological or endocrinological consequences. Additional exclusion criteria included perfume allergies, smoking, problems with smell/taste, respiratory problems, asthma, excessive alcohol/caffeine use.

Lavender

Lemon

Water (Control)

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 Cortisol and Catecholamine Production, measured over 3 Visits with at least 2 weeks between each. Average time to complete all visits was 64.46 days (SD 48.4). Cortisol: 9:05, 10:05, 10:55, 11:45, 12:15, 13:00. Nor/Epi: 9:05, 10:05, 10:55, 11:05, 11:45, 12:15.

The primary endpoint is titled “Cortisol and Catecholamine Production.” It concerns measurements of cortisol, norepinephrine, and epinephrine before and after a cold pressor stressor. The endpoint is reported as a mean with standard deviation in pg/ml (log 10). The stressor is described as physical and occurred at 11:00. The registry lists scheduled measurement times for cortisol and for norepinephrine and epinephrine, rather than presenting one value for each participant or one combined score.

These biomarkers are physiological measurements. In general, repeated measurements around a stressor can describe how a measured response varies over time and across conditions. A mean describes the average observed measurement for a group or condition; a standard deviation describes variation among measurements. Reporting the values on a log scale affects how differences and averages should be interpreted: a mean on that scale is not automatically the same as an arithmetic average on the original scale. Any transformation back to the original scale requires a specified method and suitable data.

The record does not report numerical endpoint values, a treatment contrast, or a confidence interval. It also does not explain whether the endpoint was intended to be interpreted separately for each biomarker and time point or through a prespecified summary. Thus, the stated measurement purpose is clear, but the magnitude and precision of any condition-related response are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, double-masked crossover study. In general, randomization assigns treatment conditions by chance according to a planned process, helping to reduce systematic differences in treatment assignment. Double masking generally means that two groups involved in the study are unaware of assigned conditions, but the registry record does not identify who was masked or how masking was maintained. The allocation procedure and sequence-generation details are not reported.

A crossover design generally exposes participants to more than one condition and compares responses within the same people. This can be useful when participants differ in stable characteristics that affect the outcome, because within-person comparisons can reduce the influence of those stable differences. The benefit depends on the conduct and analysis of the crossover. Treatment order, period effects, carryover effects, and the time allowed between conditions can matter. The registry reports three visits with at least 2 weeks between each, but does not report the sequence of conditions, the analysis of period or carryover effects, or whether any washout rationale was specified. These are general considerations, not claims about what occurred in the analysis.

The primary endpoint is a continuous measure summarized as a mean and standard deviation. In general, a mean is an average level and a standard deviation reflects variability around that average. The endpoint is expressed in pg/ml (log 10), which indicates a logarithmic measurement scale. Statistical comparisons on a log scale concern differences on that scale unless a model or transformation is used to express them otherwise. The analysis method, any back-transformation, and the resulting numerical estimates are not reported in the registry record.

The outcome description includes cortisol, norepinephrine, and epinephrine, with measurements before and after a cold pressor stressor. It lists multiple scheduled times for each biomarker. Repeated observations from the same participant are correlated in general: measurements from one person tend to share features that measurements from different people do not. An analysis of repeated measurements should account for this dependence, rather than treating all observations as independent. The registry record does not report the model used, the covariance structure, or whether measurements were analyzed separately or jointly.

In general, a statistical model for these data could compare conditions, measurement times, and their interaction, but the registry record does not identify such a model. It is not appropriate to infer that a particular model was used merely because it could suit the design. The exact estimand—the effect or contrast the analysis aimed to estimate—is also not reported. Without that definition, a numerical result could not be interpreted unambiguously as a particular comparison among odor conditions or times.

Variation in continuous outcomes can be more complex than a single mean and standard deviation suggest. In general, over-dispersion refers to variability greater than a model assumes; when present, standard errors may be too small if the model fails to account for it. This is a general modeling issue, not a reported feature of this study. The registry does not report diagnostic checks, distributional assumptions beyond the stated mean and standard deviation, or any adjustment for unusual observations.

Multiple outcomes and comparisons raise multiplicity concerns. The registry lists four outcome measures and seven analyses, and the primary outcome description includes several biomarkers and time points. In general, testing many hypotheses can increase the chance of observing at least one apparently unusual result by chance. Researchers may address this through a prespecified primary contrast, a correction procedure, or a hierarchy of outcomes. The specific strategy, if any, is not reported in the registry record. Neither the number of analyses nor the number of outcome measures alone reveals whether an adjustment was needed or applied.

The sample-size assumptions describe a planning illustration based on the observed effect, rather than the trial’s own design justification. The illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is a stated fraction of the observed effect. These are planning assumptions, not evidence that the study had a particular probability of detecting an effect. Any sample-size values from such an illustration are planning quantities, never observed power. The illustration’s two-arm equal-allocation assumption also differs from the recorded three odor conditions and crossover design, so it should not be read as a description of the study’s actual allocation. The trial’s own sample-size justification is not reported in the registry record.

Finally, a confidence interval and a p-value answer different questions. In general, a confidence interval conveys estimate precision and a range of values compatible with the data and model. A p-value measures how unusual the data would be under a specified null hypothesis and model; it does not describe the size or importance of an effect. Neither is reported for the primary endpoint. The available design details cannot substitute for these results or reproduce them.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and seven analyses. The primary outcome description also includes multiple biomarkers and repeated measurement times. These features create potential for multiple comparisons, but the counts alone do not identify which results were confirmatory, whether the analyses were prespecified, or how the family of comparisons was defined. Any multiplicity adjustment or testing hierarchy is not reported in the registry record.

In general, multiplicity matters because conducting many statistical tests can raise the chance of at least one apparently notable result under a collection of null hypotheses. A clearly specified primary contrast can help readers distinguish the main question from exploratory comparisons. The record does not report a single reduced comparison for the primary outcome, and the available information does not establish the intended testing hierarchy.

Missing-data handling is also not reported in the registry record. It does not describe how incomplete visits, unavailable biomarker measurements, or withdrawals were handled, whether any imputation was used, or what assumptions supported the analysis. The analysis populations are not reported either; the record does not identify whether analyses used all enrolled participants, those completing particular visits, or another defined group.

These omissions matter because repeated measurements and crossover comparisons can be affected by incomplete observations. In general, methods that use only complete cases and methods that use available repeated observations can rely on different assumptions. The record does not permit an assessment of whether missingness occurred, how much data were missing, or whether the analysis was sensitive to missing-data assumptions. No conclusion about the impact of missing data can be drawn from the registry record.

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 safety entry lists one combined group titled “Lavender Oil / Lemon Oil / Water Control.” It reports 56 participants at risk for serious adverse events and 0 affected, and 56 at risk for other adverse events and 0 affected. These are the counts recorded for that group; they are not separate event counts for each odor condition.

The record does not list any serious adverse-event terms. The deaths field is null, so deaths are not reported in the registry record; a null entry should not be read as a reported count of zero. The record also does not provide event definitions, follow-up detail for safety reporting, or a breakdown by visit or exposure condition.

In general, a count of zero reported events does not establish that a condition has no risk. The interpretation depends on how events were defined, collected, and attributed, and on the amount of observation. These figures describe the registry’s reported counts for the combined group; they do not establish a causal safety comparison among lavender, citrus, and water. No numerical comparison of safety between conditions is reported.

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

The registry record establishes that a randomized, double-masked crossover study was conducted to examine physiological responses to relaxant and stimulant odors. It identifies lavender, citrus, and water, and describes cortisol, norepinephrine, and epinephrine measurements around a cold pressor stressor. It also reports the enrollment and the visit schedule. These details define the study question and the types of measurements collected.

The record does not report numerical results for the primary endpoint. Consequently, it does not show the direction or size of any difference between odor conditions, the amount of uncertainty around an estimate, or whether any contrast met a statistical criterion. It cannot support a conclusion that an odor changed a biomarker, that conditions were equivalent, or that an observed response was important. Missing numerical results should not be interpreted as a null finding.

In general, statistical significance and clinical meaning are distinct. A small effect can be estimated precisely, while a larger estimate can be uncertain; whether a physiological difference matters depends on context and a prespecified interpretation, neither of which is reported in numerical form. A confidence interval would help communicate precision, but one is not reported in the registry record.

The study’s stated focus is a physiological response to fragrances, not a reported clinical treatment outcome. The record lists stress, anxiety, and depression as conditions, but does not provide evidence about symptom change or patient benefit. The available information therefore supports a description of the research question and design, not medical advice or a claim about treatment effects.

The findings cannot be generalized from the registry description alone. The participant characteristics, eligibility criteria, analysis population, and numerical outcome results are not reported in the information available for interpretation. Any broader inference would require those details and the study’s actual estimates, alongside an assessment of their precision and limitations.

13. Why This Trial Matters Statistically

Statistically, this study brings together a crossover design, repeated biomarker measurements, and a physical stressor. Each feature affects how results should be summarized. A crossover can support within-person comparisons in general, while repeated measurements require attention to their dependence. Measurements taken around a stressor also require a clearly defined time-based comparison. The registry identifies these design elements but does not report the statistical model used to connect them.

The endpoint is not one simple measurement: it names three biomarkers and gives multiple scheduled times. That makes the definition of a primary contrast important. A reader needs to know whether the intended question concerns an average across times, a response at a particular time, change around the stressor, or a condition-by-time pattern. The registry record does not report such a summary contrast, so the intended estimand cannot be reconstructed from the endpoint title alone.

The record also illustrates why outcome counts are not enough to judge multiplicity. Four outcome measures and seven analyses are listed, but the number of tests, their relationships, and any adjustment strategy are not described. Without a prespecified analysis hierarchy, readers cannot distinguish confirmatory comparisons from other analyses based only on these counts.

Finally, a design description is not a result. Enrollment, randomization, masking, and scheduled measurements help explain what the study set out to do; they do not reveal the estimated response or its precision. Because numerical primary-outcome results are not reported in the registry record, the statistical contribution that can be assessed is chiefly the study structure and the information needed for a rigorous interpretation.

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 principal limitation for interpreting this trial is that numerical results for the primary outcome are not reported in the registry record. There are no posted means, standard deviations, treatment contrasts, confidence intervals, or p-values for cortisol, norepinephrine, or epinephrine. The magnitude and precision of an observed response therefore cannot be evaluated.

The primary outcome description spans multiple biomarkers and measurement times, but does not identify a single reduced comparison. The record does not report the exact estimand, the statistical model, how repeated observations were handled, or whether condition order and period effects were considered. These details are particularly relevant in general to crossover and repeated-measures designs, but their handling in this study is not reported.

The record identifies three visits with at least 2 weeks between each and an average time to complete all visits of 64.46 days (SD 48.4). It does not describe visit completion by participant, the exact treatment sequence, or whether all measurements were obtained at every scheduled time. The potential effect of incomplete or unevenly spaced observations cannot be assessed from the reported information.

Further limitations are that the registry does not report participant characteristics, detailed eligibility criteria, analysis populations, missing-data methods, or a multiplicity strategy. It also does not provide detailed descriptions of the lavender, citrus, and water conditions or the procedures used to maintain double masking. The sample-size justification for the trial itself is not reported in the registry record. The planning assumptions described separately are not a substitute for that justification or for numerical results.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: The Effects of Smell on Mood and Physical Responses — NCT00097253.
• Registry record last updated 2010-02-18; status last verified 2010-02.
• Results first posted 2010-02-04.
• 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 Phase 1, randomized, double-masked crossover study sponsored by the National Center for Complementary and Integrative Health. Enrollment was 56. The behavioral intervention involved exposure to relaxant and stimulant odors, with lavender, citrus, and water identified as conditions. The study’s stated purpose was to examine bodily responses to fragrances commonly used in aromatherapy.

The primary outcome was titled “Cortisol and Catecholamine Production” and concerned cortisol, norepinephrine, and epinephrine measurements before and after a cold pressor stressor. The endpoint was expressed as a mean with standard deviation in pg/ml (log 10). The record describes three visits with at least 2 weeks between each and an average completion time of 64.46 days (SD 48.4).

Four outcome measures and seven analyses are listed. Numerical results for the primary outcome, the analysis plan, missing-data methods, analysis populations, and multiplicity procedures are not reported in the registry record. Safety figures list 0 serious adverse events and 0 other adverse events among 56 at risk in a combined group; deaths are not reported.