Investigating Fear Of Recurrence as a Modifiable Mechanism of Behavior Change: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Investigating Fear Of Recurrence as a Modifiable Mechanism of Behavior Change (NCT03853213), a randomized na study. The primary outcome is Change in Total Score for Concerns About Recurrence Scale [Adapted for Acute Coronary Syndrome].

2. Clinical Question

3. Trial Design

The registry describes a randomized, triple-masked, parallel-group behavioral study. Its status is completed. The listed phase is NA, and the enrollment is 26. The study was sponsored by Columbia University. The recorded start date is 2019-03-28, and the primary completion date is 2020-06-30. Which parties were masked is not reported in the registry record.

Participants were assigned to Cognitive Bias Modification Training or Attention Control Training. Both groups completed two tasks, each repeated 8 times over 4 weeks, with sessions twice per week. The intervention was designed to reinforce attention away from ACS threat-related stimuli. The control was described as a placebo version that was not intended to train attention toward or away from threatening or neutral material. Further details of the tasks are not reported in the registry record.

The primary endpoint was change in the adapted Concerns About Recurrence Scale total score. It was assessed at Pre-Training/Time 1 and Post-Training/Time 2, approximately 4 weeks apart. The registry lists a two-sided t-test for the between-group difference in mean change score. The primary analysis included only participants who completed both visits: 10 out of 11 in the intervention group and 7 out of 9 in the control group.

The study also listed outcomes related to self-reported medication nonadherence, physical activity, cue presence, future time perspective, and the percentage of adherent days for selected medications. The registry record does not report the rationale for the enrollment target, detailed eligibility criteria, or the statistical plan beyond the stated primary 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.

Before assignment: Twenty-six patients consented to the study. After consent, all participants were asked to complete a baseline questionnaire and a brief training with a demo tablet. Two participants decided to withdraw from the study, and four participants were administratively withdrawn by the PI as they were determined to be unable to comply with protocol (e.g., unable to complete tablet demo) and therefore not eligible for the study. These six participants were not assigned study groups.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The number of participants analyzed is the number of participants who completed Visit 1 and Visit 2 such that the change score outcome could be computed. Only the subjects that completed both visits were included in the analysis (10 out of 11 subjects in intervention group and 7 out of 9 subjects in placebo group).

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 Total Score for Concerns About Recurrence Scale [Adapted for Acute Coronary Syndrome], measured over Pre-Training/Time 1, Post-Training/Time 2 (approximately 4 weeks apart).

The primary endpoint was change in total score on an adapted 19-item Concerns About Recurrence Scale for ACS. The scale asks about fear of recurrence and has a total score range of 0 to 76; higher scores indicate greater fear. It includes health worries, role worries, and death worries. Each participant’s change score was calculated as the Time 2 score minus the Time 1 score, with the visits approximately 4 weeks apart.

The study used this measure because its central question concerned whether the intervention could alter fear of a future cardiac event. The registry identifies this as the sole primary outcome and says the trial was statistically powered to reduce fear of recurrence. That statement does not give a sample-size justification or further detail about the design assumptions; those are not reported in the registry record.

A negative change score means that the measured score was lower after training than before training. The group summaries are means of individual change scores, not a comparison of the separate average scores at each visit. The endpoint therefore describes change over the measurement interval, while the between-group test addresses whether the average change differed between the assigned groups.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is continuous: each participant has a numerical score, and the analysis summarizes change in that score. A mean change is the average of the individual post-training minus pre-training differences. A negative mean therefore denotes a lower score at the second measurement on average. This does not mean every participant’s score fell; individual-level results are not reported in the registry record.

What does the between-group estimate describe?

The central comparison is the difference between the groups’ mean change scores. It is distinct from asking whether either group changed from its own starting point. A within-group change can occur in both groups for reasons that include time, repeated measurement, and the experience common to both training schedules. The randomized comparison is intended to assess whether average change differed between the assigned groups.

The registry reports the estimated mean difference as 2.54. The record gives that estimate with a 95% confidence interval from -11.62 to 16.71. A confidence interval describes the uncertainty around an estimated quantity under the statistical model and its assumptions. In repeated-sampling terms, a method with 95% coverage would produce intervals containing the target quantity in 95% of repeated applications under those assumptions. It is not a statement that there is a 95% probability that this particular interval contains the true value.

Because the interval includes zero and reaches values on both sides, the data are compatible with a range of possible between-group differences, including little or no difference and differences in either direction. The interval is wide relative to the scale’s score range, reinforcing that the estimate is not precise. It should not be interpreted as a list of equally likely effects, and it does not establish that effects at the extremes are plausible without qualification.

What does the t-test contribute?

The registry names a two-sided t-test. In general, a t-test evaluates a null hypothesis of no difference in means against departures in either direction, using the observed group summaries and assumptions about the sampling distribution. A two-sided test does not limit the alternative to one prespecified direction. The reported p-value is 0.698. In general, a p-value is the probability, assuming the null hypothesis and the test’s assumptions, of obtaining a test statistic at least as inconsistent with the null as the observed statistic. It is not the probability that the null hypothesis is true, and it is not the probability that chance alone caused the result.

The t-test result and confidence interval are two ways of conveying evidence and uncertainty about the same comparison. A relatively large p-value does not demonstrate equivalence. An equivalence conclusion would require a prespecified margin defining differences small enough to count as equivalent and an analysis designed to test that question. Such a margin or equivalence analysis is not reported in the registry record.

Why does variability matter?

The standard deviation describes the spread of individual change scores around the group mean. The reported standard deviations are 19.55 for the intervention group and 3.93 for the control group. Greater variability makes a mean less precise when the analyzed group is small. Marked differences in spread can also matter to the assumptions and behavior of a conventional pooled-variance t-test. In general, unequal variances can make a pooled method less suitable; a method that does not assume equal variances is one possible alternative. The registry record does not say whether the reported t-test assumed equal variances or used an alternative variance estimate, so its exact implementation is not reported.

The analyzed counts were 10 and 7. These counts are the participants with both visits needed to calculate the change outcome, rather than necessarily everyone randomized. Small analyzed groups and uneven observed variability mean the estimate warrants cautious interpretation. They do not, by themselves, establish that the test assumptions failed.

How should planning calculations be read?

In general, a sample-size planning illustration asks how many participants might be needed under chosen assumptions to detect a specified effect with a chosen power and significance level. It is not a measurement of the power achieved by the completed trial. The registry describes its planning illustration as based on the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such calculations depend on their assumptions and are not a substitute for the observed confidence interval or the trial’s actual design rationale. The design rationale and sample-size justification are not reported in the registry record.

None of these general statistical concepts changes what the posted analysis estimates: a between-group difference in mean change on the adapted fear-of-recurrence scale among participants with both visits. The result remains an estimate with substantial uncertainty, and it does not answer the secondary adherence questions.

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("Cognitive Bias Modification Training", "Attention Control Training")
n   <- c(10, 7)
m   <- c(-5.4, -2.86)
sp  <- c(19.55, 3.93)
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  Cognitive Bias Modification Training  10  -5.400  19.550  -19.385  8.585
CARM  Attention Control Training  7  -2.860  3.930  -6.495  0.775

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a two-sided t-test of the mean difference in change score. The estimate is 2.54, its 95% confidence interval is -11.62 to 16.71, and the p-value is 0.698. These are the figures to use when describing the reported primary analysis.

A separate unadjusted recalculation is not estimable from the posted summaries. The registry record does not provide a recalculated between-group result for this continuous endpoint. Consequently, there is no recalculated estimate, interval, or p-value to compare with the sponsor-reported analysis, and no discrepancy between two sets of estimates can be assessed. The reported t-test should not be replaced with a result that is unavailable from these summaries.

The arm-level means and standard deviations describe the observed change scores, while the sponsor-reported test supplies the stated inferential comparison. The record does not report further details about the t-test implementation, including any variance adjustment or covariate adjustment. The sponsor result therefore remains the appropriate account of the registry’s reported analysis, interpreted with the uncertainty in its confidence interval and the limited primary analysis population.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and five secondary outcome measures. The primary endpoint is change in Concerns About Recurrence Scale total score. The secondary outcomes concern self-reported extent of medication nonadherence, physical activity, cue presence, future time perspective, and percentage of adherent days for aspirin, a beta-blocker, or a statin. Results for these secondary outcomes are not reported in the registry record.

When a study tests multiple outcomes or multiple hypotheses, the chance of at least one apparently unusual result can increase if each test is considered separately. In general, multiplicity procedures can control a specified error rate across a family of tests, but the choice depends on the study’s objectives and analysis plan. Whether any adjustment was planned or used is not reported in the registry record. The record lists six outcome measures and six analyses, but does not explain how the analyses map to the outcomes or whether the number of tests reflects all analyses performed.

The primary population is defined by completion of both visits so that a change score could be calculated. The analysis included 10 out of 11 participants in the intervention group and 7 out of 9 in the control group. This is a complete-case approach for the primary endpoint in the sense that both measurements were required for inclusion. Reasons for noncompletion, the pattern of missing measurements, and any comparison of participants with and without complete data are not reported in the registry record.

Methods for imputing missing data, sensitivity analyses for missingness, and an intention-to-treat analysis are not reported in the registry record. The record also does not state whether all randomized participants were retained in analyses of outcomes that did not require both visits. It is therefore not possible to determine from the registry entry how robust the primary estimate is to assumptions about missing outcomes or whether analysis populations were defined differently for other endpoints.

13. 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 zero serious adverse events in each group. Serious adverse events were assessed among 11 participants in Cognitive Bias Modification Training and 9 participants in Attention Control Training. The record also reports zero other adverse events in each group and zero deaths in each group. No top serious-event terms are listed.

These counts describe events recorded in the registry for the stated groups and at-risk counts. They do not establish that no adverse events of any kind occurred outside the registry’s reporting definitions or observation period. The timing and duration of safety monitoring, event definitions, and methods used to collect events are not reported in the registry record.

With no serious events recorded in either group, the registry figures do not provide a comparative estimate of serious-event risk. In general, the absence of observed events in a small study does not establish that the underlying risk is zero; it limits what can be learned about uncommon events. The reported safety information should be read as a description of recorded counts, not as a comprehensive safety conclusion.

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

The primary result estimates the between-group difference in mean change on an adapted scale measuring concerns about recurrence of ACS. The two group means were negative, and the registry’s comparison had a confidence interval extending from -11.62 to 16.71 with a p-value of 0.698. The estimate is uncertain, and the posted result does not tightly distinguish among possible between-group differences.

This result does not establish that the intervention caused a reduction in fear, nor does it establish that the intervention had no effect. A nonsignificant p-value is not proof of equality. The confidence interval permits a range of values on both sides of zero, so the registry result cannot be read as a precise estimate of the direction or size of a between-group difference.

The study’s primary measure is fear of recurrence, not medication adherence. The registry describes adherence as a secondary focus and lists measures of self-reported nonadherence and adherent days for selected medications, but their results are not reported in the registry record. The primary finding therefore cannot be used to claim that medication adherence improved, worsened, or was unchanged. Nor does it test whether a change in fear explains a change in adherence.

The inference is also limited to the participants included in the primary analysis, who had completed both visits. The registry reports 10 analyzed participants in the intervention group and 7 in the control group. Missing outcome data and the comparatively large standard deviation in the intervention group contribute to uncertainty in interpreting the group comparison. The record does not provide enough information to evaluate the sensitivity of the result to missingness or alternative analysis choices.

In general, a randomized comparison can help distinguish outcomes between assigned groups, but interpretation still depends on outcome measurement, follow-up, missing data, and the statistical model. The result is best understood as an imprecise estimate for a specific scale and measurement interval, rather than a definitive answer about the broader mechanism or clinical behavior.

15. Why This Trial Matters Statistically

This trial is statistically notable for asking about a proposed behavioral mechanism rather than making medication adherence the primary endpoint. It measured fear of recurrence before and after training and treated adherence-related measures as secondary outcomes. That distinction matters: evidence about a change in a psychological scale is not automatically evidence about a later behavior, and the mechanism proposed in the study requires its own empirical assessment.

The design used a randomized parallel-group comparison and a placebo-style attention control. Comparing change between groups is more informative about the intervention contrast than interpreting a within-group change alone. Both group means were negative, but the primary question is whether the average changes differed. The reported confidence interval is broad, making precision central to the interpretation.

The study also illustrates why enrollment and analysis counts should not be treated as interchangeable. The primary change score required two completed visits, and the analyzed groups included fewer participants than the recorded at-risk counts. Reporting who contributes to each endpoint and how missing outcomes are handled is essential for judging the meaning of an estimate.

Finally, the record makes a useful distinction between a measured primary outcome and a broader research rationale involving medication adherence. The primary result cannot answer the secondary adherence questions when those results are not reported. More generally, a well-defined endpoint, clear analysis population, transparent missing-data methods, and uncertainty intervals are all important to interpreting behavioral trials, especially when the analyzed groups are small and individual change scores vary substantially.

16. 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 analysis is based on participants who completed both visits, rather than all participants listed as at risk in the two groups. The registry reports 10 out of 11 in the intervention group and 7 out of 9 in the control group for the primary change-score analysis. It does not report why participants lacked the required visits or whether noncompletion was related to group assignment or outcome values. This limits assessment of potential bias from missing data.

The observed standard deviations differ between groups: 19.55 in the intervention group and 3.93 in the control group. In addition, the confidence interval for the reported mean difference, -11.62 to 16.71, is wide. These features indicate imprecision and make the result sensitive to the characteristics of the analyzed participants. The registry does not describe whether the t-test accounted for unequal variances.

The registry does not report a sample-size justification, detailed statistical plan, prespecified analysis decisions beyond the named t-test, or sensitivity analyses. It also does not report adjustment for multiple testing, methods for handling missing outcomes, or results for the listed secondary endpoints. This prevents a fuller evaluation of how the primary analysis fits within the study’s broader set of questions.

The masking description is triple, but which study roles were masked is not reported in the registry record. Further eligibility details, task implementation details, and safety-monitoring duration are also not reported. The primary measure is self-reported and adapted for ACS; the record describes its items and score range but does not report additional validation information for this use. These limitations leave uncertainty about generalizability, measurement properties, and the robustness of the observed comparison.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Investigating Fear Of Recurrence as a Modifiable Mechanism of Behavior Change — NCT03853213.
• Registry record last updated 2021-09-17; status last verified 2021-08.
• Results first posted 2021-07-26.
• 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.

20. Record Summary

NCT03853213 is a completed randomized, triple-masked, parallel-group behavioral study sponsored by Columbia University. It compared Cognitive Bias Modification Training with Attention Control Training in participants with suspected ACS. The intervention aimed to reinforce attention away from ACS threat-related stimuli, while the attention control was described as a placebo control.

The sole primary outcome was change in total score on an adapted Concerns About Recurrence Scale, measured approximately 4 weeks apart. The primary analysis included participants who completed both visits. The sponsor-reported two-sided t-test gave a mean difference in change score of 2.54, a 95% confidence interval from -11.62 to 16.71, and a p-value of 0.698. The interval is broad, and the estimate is imprecise.

Five secondary outcome measures are listed, including medication-adherence measures, but their results are not reported in the registry record. The record reports zero serious adverse events, zero other adverse events, and zero deaths in each group. Details of the sample-size rationale, missing-data methods, multiplicity handling, and further statistical procedures are not reported.