Stress Reduction Techniques and Anxiety: Therapeutic and Neuroendocrine Effects: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Stress Reduction Techniques and Anxiety: Therapeutic and Neuroendocrine Effects (NCT01033851), a randomized phase 4 study. The primary outcome is Active Symptoms of Generalized Anxiety Disorder.

2. Clinical Question

3. Trial Design

This was a Phase 4, randomized, double-masked, parallel-group trial sponsored by Massachusetts General Hospital. The registry lists 89 participants enrolled and identifies Generalized Anxiety Disorder as the condition. The study began in 2008-03, reached primary completion in 2012-01, and is listed as completed. These dates describe the study record; they do not establish when individual participants were assessed.

Participants were assigned to one of two behavioral interventions. MBSR was described as an 8-week group intervention in mindfulness meditation techniques. SME was an 8-week group intervention educating participants about stress physiology and health lifestyle changes. The study description characterizes the approaches as ways of reducing anxiety and stress and states a hypothesis that the approaches would reduce anxiety in different ways. It also describes measurement of stress hormones, but no hormone results are included in the posted outcome details.

The primary outcome was active symptoms of GAD measured with HAMA at 2 months. The primary analysis population included participants who attended at least one class of an intervention. The registry provides outcome summaries for 48 MBSR participants and 41 SME participants. Those counts sum to the listed enrollment, but the record does not provide a participant flow account or details of attendance beyond the population definition.

Masking is recorded as double, but the registry record does not report who was masked or how masking was implemented. The randomization method, allocation concealment procedures, original sample-size justification, and detailed statistical plan are not reported in the registry record. The two groups both received structured interventions, so the comparison addresses differences between those programs rather than either program against no intervention.

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:

Analysis population for the primary outcome, as defined in the registry: The population analyzed included all participants who attended at least one class of an intervention.

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 Active Symptoms of Generalized Anxiety Disorder, measured over 2 months.

The primary endpoint was the mean score on the Hamilton Anxiety Scale (HAMA) at 2 months, reported in units on a scale. HAMA was defined as a 14-item measure of anxiety symptoms. Each item is answered on a 0-4 scale; the description says that 0 for a single question generally represents no symptoms and 4 represents severe levels of that symptom. Item scores are added to produce a total score with a possible range of 0 to 56. Thus the outcome summarizes reported symptom severity across several anxiety features rather than focusing on one symptom alone.

A mean is a measure of the group’s average score. The standard deviation describes how dispersed individual scores are around that average. Reporting both helps readers distinguish the group’s typical level from the variation among participants. The endpoint was designated as the primary anxiety outcome, making it the main measure for the study’s stated comparison of the two stress-reduction approaches.

The record defines the analysis population as participants who attended at least one class of an intervention. It does not report how outcomes from participants without an observed 2-month assessment were handled, nor does it provide additional details about measurement procedures or score interpretation beyond the scale description. The endpoint therefore supports interpretation of the posted group summaries, but those summaries alone do not show the distribution of individual responses or the course of symptoms at other time points.

7. Results Posted in the Registry

• Stress Management Education: mean 16.27 versus 13.65 on control; difference 2.62 (95% CI -0.36 to 5.60), standardized effect d = 0.37; z-test p = 0.085.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Clinical Global Impression of Severity (CGIS) of Anxiety Symptoms.

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Stress Management Education: mean 3.58 versus 3.15 on control; difference 0.43 (95% CI -0.07 to 0.93), standardized effect d = 0.36; z-test p = 0.093.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Stress Management Education and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists one secondary outcome, Clinical Global Impression of Severity (CGIS) of Anxiety Symptoms, measured at 2 months. Its group summaries and comparison follow a broad pattern similar to the primary outcome: the unadjusted point estimate is in the direction of a higher mean in SME than MBSR, while its interval includes zero. Separately, the record reports an ANOVA p value for this outcome without an effect estimate or confidence interval. These are different forms of evidence and should remain distinguishable when assessing the overall results.

The secondary result can add context to the primary HAMA comparison because it addresses anxiety severity using another measure. Agreement in direction across measures can be informative, but it does not make one outcome a substitute for the other. The registry does not report the correlation between measures, their relative clinical importance, or a prespecified hierarchy for interpreting them. Nor does it report enough analysis detail to reconcile the ANOVA result with the unadjusted recalculation.

Unadjusted secondary results generally carry less weight than a prespecified primary outcome because they may be one among multiple opportunities to observe a seemingly notable result. When several outcomes or analyses are examined, the chance of at least one result appearing unusual by chance can increase. This is the general statistical reason multiplicity matters. The registry record does not report a multiplicity adjustment or a strategy for controlling interpretation across outcomes.

For that reason, the secondary findings are most appropriately treated as supportive or exploratory context rather than conclusive confirmation of the primary result. The posted record contains only one listed secondary outcome, but it does not establish whether other outcomes were measured, analyzed, or omitted from the record. The absence of additional posted results is not evidence that no other analyses took place.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Clinical Global Impression of Severity (CGIS) of Anxiety Symptoms.

9. Statistical Interpretation

10. Statistical Methods Explained

Continuous outcomes such as HAMA and CGIS are summarized by means and standard deviations. The mean gives the arithmetic average of the observed scores in a group. The standard deviation describes the spread of those scores around that group average; it is not a confidence interval and does not directly describe uncertainty in the mean. In this trial, the reported group sizes accompany the means and standard deviations, allowing a simple comparison of the two group averages.

What does the mean difference represent?

The primary recalculation compares the SME mean with the MBSR mean and reports the difference as SME minus MBSR. Its estimate is 2.62 units. This direction matters: a positive value means the observed mean was higher in SME, while a negative value would mean it was higher in MBSR. This is a difference in group averages, not a difference for every participant. It does not show that each person’s score changed by that amount, and the record does not provide individual before-and-after scores.

The recalculation is described as unadjusted. In general, an unadjusted comparison uses the posted group summaries without controlling for other participant characteristics or model terms. That makes the result relatively transparent, but it also means the comparison does not account for any covariates. The registry record does not report whether the trial’s own primary analysis used adjustment, repeated measurements, or another model, so the recalculation should not be mistaken for a full account of the planned or conducted analysis.

How should a confidence interval be understood?

A confidence interval expresses the uncertainty around an estimate under the assumptions of the method used to calculate it. A narrower interval indicates greater precision; a wider interval indicates less precision. For the primary mean difference, the 95% interval runs from -0.358 to 5.598. Its inclusion of zero means that a zero difference is among the values compatible with this interval-based analysis. The endpoints also indicate that the data remain compatible with differences in either direction. A confidence interval is not a range containing every participant’s score and does not assign a probability to a particular fixed value after the analysis.

The p value is another way to describe how compatible the observed result is with a specified null model, typically one in which there is no group difference. The primary recalculated unadjusted p value is 0.085. It does not measure the size or clinical importance of a difference, and it is not the probability that the null hypothesis is true. It should be interpreted alongside the estimate and interval rather than on its own. The record does not state a primary-outcome significance threshold for the trial’s original analysis.

What role does variability play?

Participant scores vary, even when people are in the same assigned group. Greater variability makes a group mean less precisely estimated for a given group size. The reported standard deviations are 7.01 for MBSR and 7.26 for SME on HAMA. The recalculation also reports a pooled standard deviation of 7.126 and a standardized effect estimate of 0.368. A standardized effect expresses the mean difference relative to a measure of score variability; it places the difference on a scale that is not tied to the original outcome units. It does not replace the original-unit difference, and it should not be read as a clinical judgment.

When observed variability is greater than anticipated, a study may have less precision than planned. In count outcomes, extra variability relative to a basic statistical model is often called over-dispersion; allowing for it can prevent standard errors from being too small. That general issue is not the central one for these continuous outcomes. The registry record does not report a variance model, an over-dispersion adjustment, or a reason to use one for HAMA or CGIS.

Why are there two secondary-outcome results?

For CGIS, the posted unadjusted recalculation gives a mean difference of 0.43 and a 95% confidence interval from -0.072 to 0.932, with an unadjusted p value of 0.093. The registry separately reports an ANOVA result with p = 0.037 and no corresponding estimate or confidence interval. These results need not be identical because one is an unadjusted calculation from the group summaries and the other is an analysis reported in the registry. However, the details needed to explain the difference are not reported in the registry record. It is therefore important to identify which result is being discussed and not to combine their estimates or p values.

An analysis of variance (ANOVA) is a general method for comparing group means. Its use in the secondary result does not, by itself, reveal the exact model specification, adjustment, or assumptions. The registry lists the method but does not give additional details for this result. A p value without an effect estimate and interval gives less information about the size and precision of the group contrast.

What assumptions underlie the planning illustrations?

The sample-size planning illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It sets the effect to detect as a stated fraction of the observed effect. These are the assumptions of the posted planning illustration, not reported assumptions from the trial’s original design. The record does not report the original sample-size justification or its planned detectable effect. The planning values therefore describe hypothetical future-study sizes under stated assumptions, rather than a retrospective assessment of whether this trial had adequate power.

11. Confidence Intervals and Precision

Precision is the degree to which an estimate is tightly determined by the observed data and the method used. The primary mean difference is 2.62 units, and its 95% confidence interval extends from -0.358 to 5.598. The interval crosses zero and includes both a small negative difference and a positive difference. This means the posted summary-statistic comparison does not pinpoint the direction of the underlying mean difference with high certainty. It also leaves uncertainty about its magnitude.

The group means and standard deviations give context for that interval: MBSR had a mean of 13.65 and standard deviation of 7.01 among 48 participants; SME had a mean of 16.27 and standard deviation of 7.26 among 41 participants. The standard deviations indicate notable spread of individual scores within each group relative to the mean difference. They describe participant variability, while the confidence interval describes uncertainty in the estimated group contrast. Those are related but distinct quantities.

A confidence interval should not be read as a set of likely individual outcomes, nor as proof that every value inside it is equally plausible. It is an interval produced by a statistical procedure and depends on assumptions. The record does not provide all details of the trial’s primary inferential method, so the recalculated interval is best understood as the precision statement for the posted unadjusted comparison.

The secondary unadjusted difference is 0.43, with a 95% confidence interval from -0.072 to 0.932. The separate registry ANOVA result does not include a confidence interval or effect estimate. Without those quantities, its precision and magnitude cannot be assessed from that reported result alone.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size planning figures are participants per arm, not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; the target effect is the stated fraction of the observed effect. The registry record does not report the trial’s original sample-size assumptions or justification, so these figures are planning illustrations for a hypothetical study rather than a reconstruction of the original design.

The 80% power column gives the number of participants per arm that the planning calculation indicates would be needed to have an 80% chance of detecting the specified effect under its assumptions, if the effect and model assumptions were correct. The 90% power column gives the corresponding per-arm planning number for a higher chance of detection under those same assumptions. These are prospective design concepts: power concerns the probability of detecting a specified effect under a statistical model, not the probability that a result is true.

The planning rows consider 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. As the target effect becomes smaller, the required number of participants per arm increases. That pattern follows because a smaller signal is harder to distinguish from random variation unless the study includes more information, often through a larger sample.

These values are not observed power. They do not show whether the completed trial succeeded or failed to detect an effect, and they do not replace the observed estimates and confidence intervals. The illustrated sample sizes are conditional on the listed assumptions and on using the observed effect as the reference for planning; they should not be interpreted as the trial’s actual enrollment target.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 2.620; pooled standard deviation σ = 7.126.
2. n = 2 × (1.96 + 0.842)² × 7.126² / 2.620² = 116.1, rounded up to 117 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 2.62                      # observed difference in means
sigma <- 7.12603184704205                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1033851)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 117 (80% power), 156 (90% power)
 75% of observed effect: n per arm = 207 (80% power), 277 (90% power)
 50% of observed effect: n per arm = 465 (80% power), 622 (90% power)
power.t.test: 117.1 
Simulated power at n = 117 per arm: 0.803 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when investigators examine more than one outcome, time point, subgroup, or statistical model. Each additional analysis can create another opportunity for a result to appear statistically notable by chance. One common response is to prespecify a primary outcome and analysis, and to use an adjustment or ordered testing strategy for additional comparisons. The registry identifies HAMA as the primary outcome and lists CGIS as a secondary outcome. It does not report a multiplicity adjustment, a testing hierarchy, or how results across outcomes were intended to be interpreted.

The registry record also does not report a detailed primary statistical analysis plan, including whether covariates were to be adjusted for, how repeated observations would be handled, or what model would be used for the primary comparison. It does list an ANOVA for the secondary result, but not its model details. These omissions limit the ability to distinguish prespecified confirmatory analyses from analyses chosen after examining the data.

The analysis population is described as all participants who attended at least one class of an intervention. This is not the same as a clear statement that all randomized participants were analyzed as assigned, regardless of attendance. The record does not report an intention-to-treat analysis population or explain how participants who attended no classes were handled. It also does not give a participant flow account.

Missing outcome data can affect both precision and the validity of comparisons, depending on how much is missing and why. The registry record does not report the number of missing HAMA or CGIS assessments, the reasons for missingness, or the method used to handle missing values. It does not describe imputation, complete-case analysis, or sensitivity analyses. Without these details, it is not possible to judge whether missing data could have influenced the posted comparisons.

These are reporting limits rather than evidence that a particular analytic choice was inappropriate. They mean that interpretation should stay close to the quantities actually posted, while recognizing that the full conduct and analysis of the trial cannot be reconstructed from the registry record alone.

15. 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 1% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists serious adverse events for both intervention groups. There were 0 serious events among 48 participants at risk in the MBSR group and 0 among 41 participants at risk in the SME group. The serious adverse event terms list is empty. These figures describe what is recorded for the groups and should not be taken as proof that no event could occur outside the period or scope captured by the record.

For other adverse events, the registry lists 1 affected participant among 48 at risk in MBSR and 1 among 41 at risk in SME. The record does not give event descriptions, severity details, timing, or an assessment of whether the events were related to the interventions. It also does not report a death count: deaths are listed as null for both groups, which is not the same as an explicit count of zero.

Adverse event counts are descriptive. The registry record does not provide enough information to assess differences in risk, event types, or causal attribution. The figures should therefore be read as the recorded counts, not as a complete safety profile or as evidence comparing the interventions’ safety.

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

The posted primary comparison describes average HAMA scores at 2 months among participants who attended at least one class. The mean was 13.65 in MBSR and 16.27 in SME. The unadjusted difference, expressed as SME minus MBSR, was 2.62 units, with a 95% confidence interval from -0.358 to 5.598 and an unadjusted p value of 0.085. The estimate points toward a higher average score in SME, but the interval includes zero and allows for differences in either direction. The result therefore has uncertainty about both direction and size.

These data do not establish that one intervention causes better outcomes than the other. Both groups received an active program, and the posted primary comparison is an unadjusted recalculation from summary statistics. The registry does not report sufficient primary-analysis detail to determine whether the trial’s own analysis adjusted for other factors or used a different model. The record’s statement of a hypothesis is not a result, and the summary statistics alone do not reveal how individual participants’ scores changed.

The results also do not establish whether either intervention is effective compared with no treatment, usual care, or another form of care, because no such group is listed. Nor do they establish effects on stress hormones: although the study description says stress hormones would be measured, the outcome details listed do not include hormone results. Similarly, the recorded adverse event counts do not by themselves establish comparative safety.

Interpretation is limited to the defined analysis population and the posted 2-month outcomes. The record does not report missing-data handling, detailed participant flow, original design assumptions, or a full statistical plan. Those omissions matter because they affect how confidently the summaries can be generalized and how the uncertainty should be understood. The appropriate statistical reading is therefore descriptive and cautious: the record estimates a group contrast, but leaves substantial uncertainty about its magnitude and does not settle a causal or clinical conclusion.

17. Why This Trial Matters Statistically

This trial illustrates why a group mean difference should be read together with its uncertainty. The primary estimate is 2.62 HAMA units, but the 95% confidence interval extends from -0.358 to 5.598. A point estimate alone could give an overly definite impression; the interval makes clear that the posted comparison does not tightly determine either the direction or the size of the difference.

It also illustrates the distinction between an outcome summary and a trial’s own statistical analysis. The primary outcome is reported through group means and standard deviations, while the comparison is an unadjusted recalculation. For the secondary outcome, the registry includes both an unadjusted recalculation and a reported ANOVA p value, without enough analysis detail to explain their difference. Identifying the source and nature of each statistic is essential to avoid treating different analyses as interchangeable.

The design also provides an example of an active-comparator question. MBSR and SME are both structured behavioral programs, so the comparison is about outcomes across these two approaches rather than about intervention versus no intervention. The analysis population further matters: the record defines it by attendance at at least one class, and does not report an intention-to-treat analysis.

Finally, the sample-size illustration shows the relationship between the target effect and planning needs. Smaller effects require more participants under the listed assumptions. Those illustrations are distinct from the completed trial’s results and from observed power. Keeping those distinctions clear helps readers assess what the registry can establish, what remains uncertain, and which design details are not reported.

18. 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 information is limited. The registry provides group means, standard deviations, and an unadjusted recalculation, but does not report a sponsor-reported primary estimate or a detailed primary model. It is not possible to determine from the record whether the original analysis adjusted for participant characteristics, used additional measurements, or followed a prespecified analysis plan.

The analysis population included participants who attended at least one class. The record does not clarify how people who attended no classes were treated, whether all randomized participants were otherwise included, or how many participants lacked a 2-month outcome. Missing-data methods and sensitivity analyses are not reported. These gaps limit assessment of selection and missingness effects.

The registry reports only the 2-month primary and secondary outcome summaries included in the registry entries. It does not provide individual trajectories, distributions, baseline scores, subgroup results, or stress hormone results. The brief description mentions hormone measurement, but those results are not listed. The record also does not describe the specific masking procedures or the randomization and allocation concealment methods.

The comparison involves two active interventions, so the results do not answer how either approach compares with no intervention or usual care. Both programs are described as 8-week group interventions, but the record does not provide information about attendance patterns, adherence beyond the minimum population definition, or delivery fidelity. Finally, the secondary result includes an ANOVA p value without an estimate or interval, and the registry does not provide enough detail to reconcile it with the unadjusted summary-statistic result.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Stress Reduction Techniques and Anxiety: Therapeutic and Neuroendocrine Effects — NCT01033851.
• Registry record last updated 2014-06-04; status last verified 2014-05.
• Results first posted 2013-06-11.
• 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.

22. Record Summary

The registry lists a completed Phase 4, randomized, double-masked, parallel-group study sponsored by Massachusetts General Hospital, with 89 participants enrolled. It compared two 8-week behavioral group interventions for Generalized Anxiety Disorder: Mindfulness Based Stress Reduction and Stress Management Education.

The primary endpoint was mean HAMA score at 2 months among participants who attended at least one intervention class. The group means were 13.65 for MBSR and 16.27 for SME. The posted unadjusted recalculation for SME minus MBSR was 2.62 units, with a 95% confidence interval from -0.358 to 5.598 and an unadjusted p value of 0.085.

The record also lists CGIS as a secondary outcome and reports an ANOVA p value of 0.037 without an estimate or interval. Details of the original sample-size justification, primary analysis plan, missing-data handling, and multiplicity strategy are not reported in the registry record. The posted findings describe the recorded outcomes but do not resolve those reporting limitations.