Loving-Kindness Meditation for PTSD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Loving-Kindness Meditation for PTSD (NCT01962714), a randomized na study. The primary outcome is Clinician Administered PTSD Scale (CAPS-5) Score.

2. Clinical Question

The registry describes a randomized controlled trial and identifies PTSD and depression as conditions of interest. Its summary says the investigators expected LKM reductions in PTSD and depressive symptom severity would not be meaningfully worse than reductions with CPT-C. It also states a hypothesis about posttraumatic maladaptive beliefs mediating symptom improvement, with a stronger role expected for CPT-C. The record does not report a mediation analysis, so that hypothesis cannot be assessed from the posted results.

The principal endpoint reported in the results is CAPS-5 score at 6 months post-intervention. The endpoint is a continuous symptom-severity measure, with higher scores indicating worse PTSD. Although the registry describes the intended non-inferiority question in terms of change over time, the reported comparison is identified as a mean difference in final values. The distinction between the intended question and the labeled reported contrast is important for interpretation.

3. Trial Design

The registry identifies this as a randomized, single-masked, parallel-group trial. Participants were assigned to LKM or CPT-C. The registry lists enrollment as 184, while its brief summary says 170 male and female Veterans with current PTSD would be randomized. These figures describe different entries in the record; it does not explain their relationship, and they should not be treated as interchangeable measures of the analysis population.

Both interventions were behavioral group courses described as lasting 12 weeks, with 90-minute sessions and groups of 10 participants. The comparator was CPT in its cognitive-only version. This active-comparator design addresses how the two group treatments compare with one another, rather than estimating either treatment’s effect against an untreated or placebo group. The registry does not report details of treatment adherence, attendance, or treatment fidelity.

The trial is marked completed. The listed start date is 2014-12-01, and the primary completion date is 2018-11-29. The record gives no phase beyond “NA.” The sponsor is the VA Office of Research and Development. These administrative details identify the registered study but do not establish how participants were recruited, how randomization was implemented, or how masking was maintained; those details are not reported in the registry record.

The primary outcome was CAPS-5 score at 6 months post-intervention. The registry describes linear mixed-effects models for continuous outcomes, including time and a time-by-treatment interaction as fixed effects. It describes the non-inferiority assessment as comparing the rate of change from baseline to the 6-month follow-up and sets the CAPS-5 margin at 5 points. The analysis reported in the results is labeled “Mean Difference (Final Values),” so the record does not make clear that the posted estimate is the specified change-rate contrast.

Although the record lists two outcome measures and two analyses, only the CAPS-5 analysis is detailed in the posted results. A result for depressive symptoms and an analysis of the mediation hypothesis are not reported in the registry record. The record also does not describe a sample-size justification, allocation concealment, the handling of deviations from assigned treatment, or the analysis population used for the posted estimate.

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

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 Clinician Administered PTSD Scale (CAPS-5) Score, measured over 6 months post-intervention.

The primary endpoint is the Clinician Administered PTSD Scale (CAPS-5) score at 6 months post-intervention. The registry describes this as a structured interview with 30 items and a score range of 0-80. Higher scores indicate worse PTSD. The endpoint is continuous, so the reported arm summaries are means with 95% confidence intervals.

A clinician-administered symptom measure provides a structured way to assess PTSD severity rather than relying only on a participant’s general impression of change. A continuous score can also represent degrees of symptom severity across the scale, which makes comparisons of mean scores possible. These are general reasons for using such an endpoint; the registry does not report additional rationale for selecting CAPS-5 as the primary outcome.

The timing is relevant: the listed endpoint is measured 6 months after intervention, while the registry’s non-inferiority description refers to change from baseline through that follow-up. A final score and a change score answer related but distinct questions. A final-score comparison describes the groups’ endpoint values; a change comparison incorporates the baseline-to-follow-up trajectory specified in the registry description. The posted analysis is labeled as a difference in final values, and baseline values are not reported in the results.

Although depressive symptoms are part of the trial’s stated aims, the depression measure and result are not reported in the registry record. Therefore, the CAPS-5 result should not be read as a result about depression, functioning, durability beyond the stated follow-up, or the proposed mediation of symptom improvement.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The trial is described as randomized, single-masked, and parallel-group. Randomization is intended, in general, to make treatment groups comparable on average at assignment, including on factors that are measured and those that are not. That protection depends on the conduct of randomization and on how outcomes are handled after assignment. The registry identifies the design but does not report the sequence-generation or allocation-concealment procedures, so those implementation details cannot be evaluated from the record.

What does the confidence interval express?

The arm means are accompanied by 95% confidence intervals. In general, a confidence interval expresses sampling uncertainty around an estimate under the assumptions of the statistical method. It is not a range containing 95% of individual participants’ scores, and it is not a statement that there is a 95% probability that the particular interval contains the true value. For a treatment comparison, the interval for the contrast is more directly informative about uncertainty in the difference than the separate arm intervals.

The reported contrast is 2.09, with a 95% confidence interval from -2.59 to 6.78. The registry labels it a mean difference in final values. Because the interval includes zero, it is compatible with a range of possible directions for that final-value contrast. The interval’s endpoints also show that the estimate is not exact. A confidence interval does not account for every possible source of bias, such as missing outcome data or departures from the planned analysis.

How is non-inferiority assessed?

In general, a non-inferiority analysis asks whether an intervention can be ruled out as worse than a comparator by more than a prespecified clinically meaningful margin. The margin must be interpreted in the direction and on the scale defined by the protocol. The registry specifies a 5-point margin for CAPS-5 and says the lower limit of the 95% confidence interval for the difference in change rate must be greater than negative 5 points. This is a one-sided decision boundary applied to a two-sided 95% confidence interval, as described in the record.

That criterion depends on the estimated contrast matching the stated contrast. The record describes a difference in change from baseline to 6-month follow-up, but labels the posted comparison as a mean difference in final values. A change contrast accounts for the baseline-to-follow-up pattern; a final-value contrast compares only endpoint values as labeled. Without clarification that the reported interval corresponds to the specified change-rate analysis, the non-inferiority rule cannot be applied unambiguously. The record does not report a separate confidence interval for the change-rate contrast.

What role does the mixed-effects model have?

The registry says linear mixed-effects models were used for continuous outcomes, with time and the time-by-treatment interaction included as fixed effects. In general, a mixed-effects model can analyze measurements taken over time while accounting for the correlation among repeated observations from the same participant. Fixed effects represent modeled average relationships, such as change over time and whether that pattern differs by treatment. The time-by-treatment interaction is relevant to a question about whether change trajectories differ between groups.

The record does not provide the model equation, the full set of covariates, the covariance structure, the estimation method, or the way missing measurements were handled. It also does not state whether the posted final-value contrast came from the mixed model or another calculation. Those missing details limit the ability to reproduce or fully assess the analysis.

Why distinguish the estimate from the design question?

A study’s stated hypothesis, statistical model, and posted result need to align for a conclusion to follow. A label such as “non-inferiority” describes the analysis category, but does not by itself demonstrate that the estimate has the required scale, direction, follow-up definition, or comparison. Here, the stated margin is tied to a change-rate contrast, while the reported analysis is explicitly labeled a final-values difference. The numerical interval can be described as posted, but the registry record does not resolve this mismatch.

General statistical interpretation also distinguishes failure to demonstrate a difference from evidence of non-inferiority or equivalence. A confidence interval that includes zero does not establish equivalence. Non-inferiority requires the relevant confidence bound to exclude effects worse than the chosen margin. The record’s stated criterion gives that boundary, but the fit between the posted contrast and the criterion remains unclear.

No p-value is reported for the comparison. A p-value, if available, would address a specified null hypothesis under a statistical model; it would not replace the margin-based confidence-interval assessment required for non-inferiority. Nor would a statistically nonsignificant result alone show that the treatments are similar. The posted estimate and interval, together with clarity about the contrast and analysis population, are central to interpreting the registry’s claim.

10. Analyses Reported by the Sponsor

The registry reports an analysis classified as non-inferiority and labeled “Mean Difference (Final Values).” Its estimate is 2.09, with a 95% confidence interval from -2.59 to 6.78. The registry also describes a separate decision criterion based on the lower limit of an interval for the difference in change rate from baseline to 6-month follow-up, using a 5-point CAPS-5 margin. The record does not state that the reported final-values interval is the change-rate interval in that criterion.

A separate unadjusted re-calculation of the relevant two-arm continuous-outcome comparison is not reported in the registry record. There is therefore no independent recalculated estimate to compare with the registry analysis. The posted analysis is the available treatment-contrast result, but its label and the stated non-inferiority estimand should be kept distinct when interpreting it.

The arm means and their confidence intervals are also reported, but they are endpoint summaries and do not supply the baseline-to-follow-up contrast specified in the non-inferiority description. The registry does not provide enough information to resolve whether the reported analysis is merely labeled differently or represents a different contrast. No claim beyond the reported final-value comparison can be drawn from that ambiguity.

11. Multiplicity, Missing Data and Analysis Populations

The record lists two outcome measures and two analyses, but the posted results described in the registry record include only the CAPS-5 comparison. A result for the other outcome is not reported in the registry record. The trial summary mentions depressive symptoms, but no depression measure, estimate, confidence interval, or statistical test is included in the reported results. The record also does not report results for the proposed mediation hypothesis.

Multiplicity arises when several outcomes, time points, contrasts, or analyses are examined. In general, making many statistical comparisons can increase the chance that at least one appears compelling by chance. Common approaches include identifying a primary outcome and analysis in advance or adjusting inference for multiple comparisons. The registry identifies a primary CAPS-5 endpoint, but does not report a multiplicity adjustment or explain how the analyses across outcomes were prioritized. The statistical plan is not reported in sufficient detail to determine whether multiplicity was addressed elsewhere.

Missing-data handling is also not described. The registry does not report the number of participants with observed CAPS-5 outcomes at follow-up, the reasons for missing measurements, or the assumptions used to include participants with incomplete repeated measures. In general, mixed-effects models can use available repeated observations under assumptions about the missingness process, but that does not remove the need to state the assumptions and assess their plausibility. The specific missing-data method for this trial is not reported in the registry record.

The analysis population is not identified. The record does not say whether the posted results are based on all randomized participants, participants who received treatment, or another defined group. In non-inferiority studies, analysis-population choices can be especially consequential; general practice is to examine how the chosen population and missing-data assumptions affect the conclusion. No such sensitivity analyses are reported in the registry record.

The record also does not describe an interim-analysis plan, stopping rules, protocol deviations, or adjustments to the analysis plan. Those items should not be presumed. The absence of these details does not prove that they were absent from the trial’s full documentation, but they are not reported in the registry record and therefore cannot be evaluated from the posted information.

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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidality with intent or plan requiring staff intervention (4.4%), inpatient psychiatric admission (1.1%), suicide attempt (1.1%), risk of harm to others (1.1%), seizure (1.1%).

The registry lists serious adverse events affecting 5 in the LKM group and 3 in the CPT-C group. It lists other adverse events affecting 11 in the LKM group and 18 in the CPT-C group. Deaths are listed as 2 for LKM and 0 for CPT-C. These are counts recorded by the registry; the record does not describe the event-level follow-up period or provide enough detail to determine whether the entries represent unique participants, recurrent events, or another counting convention.

The most frequently listed serious-event terms are “Suicidality with Intent or Plan requiring staff intervention” at 4.4%, “Inpatient psychiatric admission” at 1.1%, “Suicide attempt” at 1.1%, “Risk of Harm to Others” at 1.1%, and “Seizure” at 1.1%. The record does not clearly assign each of these percentages to a particular arm or describe the denominator used. They should therefore be read as registry-listed terms and percentages, not as arm-specific comparative risks.

Adverse-event counts are descriptive. They do not by themselves establish that an intervention caused an event or that the difference between arms is statistically meaningful. The registry does not report comparative safety estimates, uncertainty intervals, event timing, seriousness definitions, or causal attribution details. The counts should be interpreted within those limits and alongside the incomplete description of how events were collected and classified.

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

The posted CAPS-5 results describe mean endpoint scores at 6 months post-intervention and a reported mean difference in final values. They provide an estimate and a 95% confidence interval, allowing readers to see both the direction represented by the reported contrast and the uncertainty around it. Because the interval extends across zero, the final-value comparison does not precisely establish a single direction of difference.

The record’s stated non-inferiority question is more specific than a general comparison of endpoint means. It uses a 5-point margin and defines the criterion in terms of the lower confidence limit for a difference in change rate from baseline to follow-up. The posted analysis is labeled as a final-values difference. Since the registry does not clarify that the reported interval is the required change-rate interval, it does not support an unambiguous determination that the stated non-inferiority criterion was met.

This is not evidence that the interventions are equivalent. Equivalence requires a defined range of differences considered acceptably similar and appropriate statistical evidence that the comparison lies within that range. Nor does an interval that includes zero prove that there is no difference. It indicates uncertainty compatible with values on both sides of zero for the reported contrast.

The result also does not establish whether either treatment reduces symptoms relative to no treatment, because both arms received active group interventions and there is no untreated comparison group. It does not establish a result for depression, the proposed mediation pathway, or outcomes beyond the stated follow-up. Those results are not reported in the registry record.

The posted summaries do not describe missing data, analysis populations, or enough model detail to assess the robustness of the estimate. The enrollment entry and the planned number in the brief summary are not the same, and the record does not explain the relationship between them. These details limit conclusions about precision and generalizability. The findings should be read as the registry’s reported endpoint information, not as medical advice or a complete account of every trial analysis.

14. Why This Trial Matters Statistically

This trial raises a core statistical issue in active-comparator research: what evidence is needed to support a claim that one intervention is not meaningfully worse than another? A conventional test for a difference does not answer that question by itself. Non-inferiority requires a prespecified margin, a clearly defined contrast, and an interval whose bound can be evaluated against that margin. The registry states a 5-point CAPS-5 margin, making the direction and definition of the reported contrast central to interpreting the result.

The registry also illustrates why an estimate’s label matters. The stated criterion concerns change from baseline to follow-up, while the posted result is labeled a difference in final values. Even when both are expressed in the same score units, they answer distinct questions. A confidence interval cannot be transferred from one estimand to another without justification. Transparent reporting of the contrast, model, analysis population, and missing-data handling is essential for readers to assess whether the statistical result addresses the clinical question.

Another important feature is the role of uncertainty. The reported interval spans negative and positive values, so the point estimate alone should not carry the interpretation. In non-inferiority work, uncertainty must be judged in relation to the margin, not merely by whether an ordinary test detects a difference. This trial’s record gives the margin but leaves ambiguity about whether the posted interval corresponds to the quantity defined by the criterion.

Finally, registry completeness affects what can be learned from posted results. Depression and mediation are part of the stated aims, but corresponding results are not reported in the registry record. Missing-data methods and analysis populations are also absent. These gaps do not determine what the full trial documentation may contain, but they prevent a thorough assessment from the registry information alone.

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 most important trial-specific limitation in interpreting the posted result is the apparent difference between the estimand described for non-inferiority and the contrast reported. The registry defines the decision criterion using change from baseline through 6 months, while the reported comparison is labeled as a difference in final values. The record does not explain whether this is a labeling issue or a different analysis, and does not report a separate change-rate confidence interval.

The registry lists enrollment as 184, while its brief summary says 170 Veterans would be randomized. It does not explain how these entries relate to each other or how many participants contributed data to the CAPS-5 analysis. The arm summaries list 91 for LKM and 93 for CPT-C, but the record does not explain whether these are assigned, analyzed, or observed-outcome counts. The analysis population and follow-up completeness are not reported.

Details needed to evaluate the model are absent, including the covariance structure, covariates beyond time and the time-by-treatment interaction, and the specific procedure for handling missing observations. Baseline CAPS-5 means are not reported in the outcome information, limiting assessment of the baseline-to-follow-up contrast described in the trial rationale. The record also does not report a p-value for the posted comparison.

Although the stated aims include depressive symptoms and a mediation hypothesis, the corresponding results are not reported in the registry record. The registry does not describe multiplicity handling, interim analyses, allocation concealment, treatment adherence, or sensitivity analyses. These gaps constrain interpretation of the primary result and prevent assessment of whether findings were robust across outcomes or analysis choices.

Safety information is also limited to registry-listed counts and terms. Comparative uncertainty, event timing, causal attribution, and the denominators behind the listed serious-event percentages are not reported. These limitations are specific to the information available in this trial record and should not be filled in by assumptions about standard practice.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Loving-Kindness Meditation for PTSD — NCT01962714.
• Registry record last updated 2020-03-17; status last verified 2020-03.
• Results first posted 2020-02-27.
• 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

NCT01962714 is a completed randomized, single-masked, parallel-group trial sponsored by the VA Office of Research and Development. It compared a 12-week group Loving-Kindness Meditation course with a 12-week group Cognitive Processing Therapy course, cognitive-only version, among Veterans with current PTSD as described in the registry summary.

The primary endpoint was CAPS-5 score at 6 months post-intervention. The registry reports arm means and a mean difference in final values, with a 95% confidence interval. The trial description specifies a non-inferiority margin of 5 points for a change-rate comparison, but the record does not clarify that the posted final-values contrast is the same estimand.

The record lists enrollment as 184 and states in its brief summary that 170 participants would be randomized; it does not explain the difference. Results for depressive symptoms and the proposed mediation hypothesis, as well as details on missing data and analysis populations, are not reported in the registry record.