Treating Phobia With Multivoxel Neuro-reinforcement: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Treating Phobia With Multivoxel Neuro-reinforcement (NCT03655262), a randomized na study. The primary outcome is Change in Amygdala Reactivity.

2. Clinical Question

The study also planned subjective fear and directed-attention assessments, alongside physiological measurement. Those assessments help describe different aspects of response to feared stimuli, but the posted primary outcome is the neural measure. The registry does not report inferential results establishing whether any session count differs from another. The available group summaries therefore describe observed changes, rather than answering the comparative question with a quantified estimate of uncertainty.

3. Trial Design

The registry describes a randomized, triple-masked, parallel-group study sponsored by the University of California, Los Angeles. It lists enrollment of 23 and three assigned groups: 1 Session, 3 Sessions, and 5 Sessions. The groups received the corresponding number of neuro-reinforcement sessions. The registry does not report the specific allocation procedure, the identities of the masked parties, or the operational details of masking.

The study started on 2018-10-01, had a primary completion date of 2022-06-30, and is listed as completed. Its stated objective was to test neuro-reinforcement based on decoded functional magnetic resonance imaging information in people with phobias. The description proposes that fear responses could be engaged unconsciously, without repeated direct exposure to feared stimuli.

The primary outcome was assessed over 10 days, with measurements at pre-treatment and post-treatment. It is a continuous measure reported as a mean and standard deviation. The registry lists 7 participants in the 1 Session group, 8 in the 3 Sessions group, and 8 in the 5 Sessions group for this outcome. It does not explain whether those counts represent all randomized participants, all participants with outcome data, or another analysis population.

The design includes multiple session-count groups, but the posted results do not give an inferential comparison among them. The registry reports no analyses for the outcome comparison, and notes that values are missing or not estimable for a two-arm comparison of this continuous endpoint. The planned sample-size justification and the detailed analysis plan are 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.

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.

6. Primary Endpoint

The registry lists the primary outcome as Change in Amygdala Reactivity, measured over 10 days (measured at pre-treatment and post-treatment).

The primary endpoint, Change in Amygdala Reactivity, measures the difference in functional magnetic resonance imaging response to target phobic animals compared with control phobic animals, from pre-treatment to post-treatment. It is expressed as a beta coefficient, a continuous neural-response measure. The registry specifies that lower, more negative values indicate a greater decrease in amygdala reactivity and a better outcome under the study's stated direction of interpretation.

Using a brain measure gives the study an outcome that is directly related to its proposed mechanism: neuro-reinforcement is intended to engage fear-related patterns of brain activity. Measuring the same contrast before and after treatment allows the reported endpoint to capture change rather than only a post-treatment level. The measurement timeframe is 10 days, with assessments at both time points.

This endpoint is not itself a direct measure of everyday functioning, subjective distress, or sustained response over a longer period. The registry also lists subjective fear and other measures among the study assessments, but their results are not included in the posted primary-outcome summaries. Therefore, the neural endpoint should be interpreted specifically as a measure of amygdala reactivity, not as a complete account of all possible participant experiences or outcomes.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous and is summarized by a mean and standard deviation. These statistics answer different questions. The mean describes the average observed change in a group. The standard deviation describes how spread out individual measurements are around that mean. Neither statistic, alone, quantifies the uncertainty in a difference between groups.

What does the mean describe?

For this endpoint, each group mean summarizes the average pre-treatment to post-treatment change in the fMRI contrast specified by the registry. The endpoint's direction matters: lower, more negative values are defined as greater decreases in amygdala reactivity. A group mean is therefore interpreted using that definition, but it does not show how many participants had a change in that direction or how consistently participants responded.

Group means can also conceal varied individual results. A similar average can arise when participant measurements cluster near the mean or when they are widely dispersed. The standard deviation gives some information about that dispersion. Here, the reported standard deviations are 0.796, 1.62, and 1.682 for the 1 Session, 3 Sessions, and 5 Sessions groups. Those values indicate that the observed measurements vary within groups; they do not determine whether group means differ statistically.

Why are confidence intervals useful?

In general, a confidence interval gives a range of values compatible with the data and the statistical method used. It conveys uncertainty around an estimate, such as a difference between group means. A narrow interval indicates greater precision than a wide interval, while an interval spanning values in opposing directions can leave the direction of an effect uncertain. No confidence intervals for the primary outcome are reported in the registry record. Without them, readers cannot use the posted results to assess the precision of a between-group contrast.

A p-value, in general, describes how incompatible observed data are with a specified null hypothesis under the assumptions of a test. It is not a measure of the probability that a treatment effect is real, and it does not measure the size or practical importance of an effect. The registry reports no primary-outcome analyses or p-values. A claim of statistical significance or lack of significance would therefore go beyond the available results.

What does randomization contribute?

Random allocation is intended to make assigned groups comparable, on average, with respect to factors known and unknown before treatment. It supports a causal comparison when the trial is analyzed according to a suitable prespecified plan and when the analysis handles missing observations appropriately. It does not guarantee that small groups will be balanced on every characteristic, and it does not itself produce an estimate of uncertainty. The registry identifies the study as randomized but does not report the allocation method or a primary comparative analysis.

The parallel design means that participants were assigned to one session-count group rather than being described as crossing between all groups. The intervention varied in session count. The registry calls the study triple-masked, but does not identify who was masked or how masking was maintained. These omissions limit evaluation of how assignment, expectations, or outcome assessment might have influenced the observed results.

How should the session groups be compared?

A comparative analysis would ordinarily estimate differences among the assigned groups, with uncertainty intervals and an analysis method appropriate for the continuous outcome and trial design. The posted record does not contain such estimates. It specifically notes that values are missing or not estimable for a two-arm comparison. Thus, the descriptive group summaries cannot be recast as a treatment contrast without doing a new calculation, and no new contrast is presented.

Several pairwise comparisons are possible in a study with three groups. If multiple comparisons are made, the chance of a false-positive finding can increase unless the analysis plan addresses multiplicity. The registry does not report a multiplicity strategy or an inferential plan. It is therefore not possible to determine how comparisons were intended to be prioritized or adjusted.

What do planning assumptions mean?

In general, sample-size planning estimates how many participants might be needed to detect an effect under stated assumptions. The registry's planning description refers to a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. Such assumptions are planning conventions, not evidence that a study had a particular probability of detecting an effect after it was conducted. The registry does not report its own sample-size justification, and no observed-power conclusion can be drawn from the group summaries.

More generally, over-dispersion means that observations vary more than a statistical model assumes. When present, it can make standard errors too small and uncertainty appear narrower than it is if the model does not accommodate the extra variation. The primary outcome is summarized with standard deviations, but the registry does not report a fitted model or a model-based analysis, so over-dispersion cannot be assessed as a feature of an actual analysis.

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("1 Session", "3 Sessions", "5 Sessions")
n   <- c(7, 8, 8)
m   <- c(-0.691, -0.827, -0.278)
sp  <- c(0.796, 1.62, 1.682)
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  1 Session  7  -0.691  0.796  -1.427  0.045
CARM  3 Sessions  8  -0.827  1.620  -2.181  0.527
CARM  5 Sessions  8  -0.278  1.682  -1.684  1.128

11. Multiplicity, Missing Data and Analysis Populations

The study includes a primary outcome and several other measurements. The registry names secondary measures for skin conductance response, the Fear Survey Schedule, and the Stroop Task; it also lists 5 outcome measures overall. When several outcomes or between-group comparisons are assessed, the analysis plan should ordinarily explain which results are primary and whether adjustment or another multiplicity strategy is used. A multiplicity procedure is not reported in the registry record.

The registry reports 0 analyses for the posted results. It therefore does not provide a primary inferential analysis, a set of secondary-outcome estimates, or details of how multiple comparisons were handled. The absence of posted analyses is not evidence that no analyses were conducted elsewhere; it means the record does not give the information needed to evaluate them.

Missing-data methods are also not reported in the registry record. The primary outcome counts are 7, 8, and 8 across the 1 Session, 3 Sessions, and 5 Sessions groups. The registry does not clarify whether these are numbers randomized, numbers assessed, or numbers included in an analysis. It does not report reasons for missing outcome measurements, imputation, sensitivity analyses, or assumptions about missingness.

The analysis population is not reported in the registry record. In particular, the record does not specify whether results were intended to follow randomized assignment for all participants or were restricted to participants with complete measurements. Without that information, readers cannot determine how exclusions or incomplete observations might affect the group summaries. These gaps are especially relevant when interpreting a small randomized study, because the choice of population and missing-data method can influence both the estimated effect and its uncertainty.

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.

The registry reports no serious adverse events in any of the three groups. It lists 0 serious events among 7 participants at risk in the 1 Session group, 0 among 8 at risk in the 3 Sessions group, and 0 among 8 at risk in the 5 Sessions group. It also reports 0 other adverse events in each group and 0 deaths across the groups.

These entries describe what was recorded among participants counted as at risk. They should be read as zero events reported in this study, not as proof that an event is impossible or that the intervention has no safety concerns. A study with limited enrollment cannot characterize uncommon events precisely, and the registry does not give additional details about event collection, definitions, observation time for safety, or attribution.

The registry lists no serious adverse-event terms. It does not provide narrative descriptions of possible events or explain whether there were events considered unrelated to the intervention. Accordingly, the safety information supports only a factual statement about the recorded counts; it does not support broader conclusions about safety in other settings or populations.

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

The posted results show three group-level mean changes in the primary neural measure. The mean is -0.691 after 1 Session, -0.827 after 3 Sessions, and -0.278 after 5 Sessions. Because the registry defines more negative values as greater decreases in amygdala reactivity, the 3 Sessions group has the most negative reported mean and the 5 Sessions group the least negative. These are descriptive comparisons of group averages.

The results do not establish that the session groups differ. The registry contains no inferential analysis, confidence intervals, or p-values for the primary endpoint, and it notes that values are missing or not estimable for a two-arm comparison. The within-group standard deviations are 0.796, 1.62, and 1.682. This variability is relevant to precision, but it cannot substitute for a confidence interval or a direct comparison.

The study's mechanistic focus is on amygdala reactivity measured by fMRI. The primary outcome therefore speaks to that specified neural contrast and its measured change over the stated timeframe. It does not by itself establish changes in subjective fear, attention, day-to-day functioning, or lasting effects. Although other outcomes are listed in the registry, their numerical results are not reported in the posted record.

Likewise, these findings should not be treated as evidence that a particular number of sessions is superior. The design randomized participants, but the registry's descriptive results do not include the comparative analysis needed to support that conclusion. No advice about treatment choice follows from these summaries. The most defensible reading is that the registry reports variable group means and does not report enough inferential information to determine the precision or statistical evidence for differences among groups.

14. Why This Trial Matters Statistically

This trial is statistically interesting because it compares different session counts while measuring a neural response before and after an intervention. Such a design raises a clear estimation question: how much does the change differ across assigned groups, and how uncertain is that difference? The registry supplies group means and standard deviations, but not the comparative estimates needed to answer that question.

The record also illustrates why descriptive summaries and inferential results are not interchangeable. Means can show the direction and scale of observed group averages, while standard deviations show participant-level spread. Neither provides the uncertainty around a between-group effect. With outcome counts of 7, 8, and 8 and considerable within-group variation, careful reporting of effect estimates and confidence intervals would be important for judging precision.

Multiple groups and multiple outcomes create additional statistical choices. A complete account would explain which contrasts were prioritized, how multiplicity was handled, which participants were included, and how missing data were addressed. Those details are not reported in the registry record. Their absence makes it harder to distinguish a planned confirmatory analysis from descriptive or exploratory analyses.

The study therefore offers a useful example of how a randomized design can be paired with an incomplete statistical record. The posted summaries give readers an initial view of observed responses, but without an inferential comparison they cannot quantify evidence for differences among session counts. Transparent reporting of analysis methods and uncertainty is essential for interpreting results, especially when the outcome is variable and the groups are small.

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 primary limitation for interpretation is that the registry does not report a comparative analysis of the primary outcome. It provides group means and standard deviations but no treatment-effect estimates, confidence intervals, or p-values. The record also identifies the two-arm continuous comparison as not estimable from the posted values. As a result, the relative effects of 1, 3, and 5 sessions cannot be evaluated inferentially from these results.

The primary outcome counts are 7, 8, and 8, while overall enrollment is listed as 23. The registry does not explain the relationship between enrollment and outcome counts, or whether any participants lacked post-treatment measures. Reasons for missing data, missing-data methods, and the analysis population are not reported in the registry record. This leaves uncertainty about who contributed to the reported summaries and how incomplete observations were handled.

The record does not report the sample-size justification, detailed analysis plan, or multiplicity strategy. It also does not state who was masked, how masking was maintained, or the randomization procedure. These details matter when assessing whether the posted analyses correspond to a prespecified plan and how susceptible the measurements might be to bias.

The primary measure is a neural response to a specified stimulus contrast over 10 days. It is a focused measure and does not alone describe subjective fear or broader outcomes. Although secondary measures are named, their numerical results are not reported. These limitations mean the posted results should be regarded as incomplete for judging comparative effects, precision, and the broader relevance of the measured changes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Treating Phobia With Multivoxel Neuro-reinforcement — NCT03655262.
• Registry record last updated 2024-02-02; status last verified 2024-01.
• Results first posted 2023-12-27.
• 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 describes a completed randomized, triple-masked, parallel-group behavioral study of unconscious neuro-reinforcement for phobia. Participants were assigned to 1, 3, or 5 sessions. Enrollment is listed as 23, and the primary outcome is change in amygdala reactivity measured by fMRI over 10 days.

Reported primary-outcome means are -0.691, -0.827, and -0.278 for the 1 Session, 3 Sessions, and 5 Sessions groups, respectively, with standard deviations of 0.796, 1.62, and 1.682. More negative values are defined as greater decreases in reactivity. No inferential analyses are reported, and a two-arm comparison is described as not estimable. The record also reports 0 serious adverse events, 0 other adverse events, and 0 deaths in each group.