Community Self-help Using Thought Field Therapy in a Traumatised Population in Uganda: a Randomised Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Community Self-help Using Thought Field Therapy in a Traumatised Population in Uganda: a Randomised Trial (NCT01681628), a randomized na study. The primary outcome is Change in Post-traumatic Stress Disorder Check List for Civilians (PLC-C) Score..

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial, with quadruple masking recorded in the registry. The record does not identify which participants, care providers, assessors, or other roles were masked, so the masking label alone does not show how masking was implemented. The trial enrolled 256 participants and is listed as completed. The recorded start was 2012-06 and primary completion was 2012-07.

The intervention arm received Thought Field Therapy from trained community leaders. The comparator was a wait-list group that did not receive therapy before the one-week assessment. After that assessment, the wait-list participants received the intervention and were assessed again after a further week. Thus, the principal randomized comparison concerns the initial period when one arm received therapy and the other did not. The later wait-list assessment describes outcomes after delayed treatment, rather than preserving a concurrent untreated randomized comparison.

The registry lists 114 participants for the Thought Field Therapy arm, 122 for the wait-list without therapy, and 107 for the wait-list after receiving therapy. The primary endpoint population excluded non-attenders at specified follow-up assessments. These analysis counts should therefore not be assumed to represent every participant enrolled or to be interchangeable across assessment times.

The record does not report the original sample-size justification, details of the randomization procedure, allocation concealment, a protocol, or a complete statistical analysis plan. It lists the study phase as not applicable. The trial’s registration title describes community self-help using Thought Field Therapy in a traumatised population in Uganda; the official title emphasizes delivery by newly instructed community workers. Both describe the same randomized trial.

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: Participants had been previously identified. As an excess number presented, only sufficient for the trial were assigned. One participant was excluded as not meeting age criteria (had to be 18 years old or over).

Time 1 (Entry) to Time 2 One Week Later.

Reasons for not completing this period:

Time 2 to Time 3 (One Week Later).

Reasons for not completing this period:

19 Months Follow-up

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants with symptoms suggestive of PTSD, excluding non-attenders at Time 2. Non-attenders at time 3 were excluded from the wait-list (Thought Field Therapy) arm.

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.

Population: Local population from Kasese district in Uganda who admitted to symptoms suggestive of PTSD and volunteered to participate. Non-attenders at Time 2 were not included in the analysis.

6. Primary Endpoint

The registry lists the primary outcome as Change in Post-traumatic Stress Disorder Check List for Civilians (PLC-C) Score., measured over Baseline (Time 1), One week later (Time 2) and Two weeks later (Time 3 for Wait-list: Thought field Therapy Arm).

The primary outcome was change in the Post-traumatic Stress Disorder Check List for Civilians score, described in the registry as a measure of PTSD symptom severity that can also be used to screen populations. The questionnaire is self-completed and has 17 questions, each scored from 1 to 5. The item scores are added to form a total, with a possible range from 17 to 85. The registry states that a score above 50 is accepted as diagnostic for PTSD.

A continuous score can represent differences in symptom severity across the scale rather than only distinguishing participants above or below a screening threshold. Comparing mean changes can therefore retain information from the full score, while the diagnostic cutoff answers a different, categorical question. The registry lists a secondary outcome concerning the percentage with scores diagnostic for PTSD, using PCL-C > 50 before and after treatment.

The primary outcome title uses the label PLC-C, while the description uses PCL-C. The registry does not explain this difference in labeling. It describes measurements at baseline, after one week, and, for the wait-list group after treatment, after a further week. The treatment arm was assessed immediately before treatment and one week later; the wait-list arm was assessed before and after its initial no-treatment week, then again after receiving therapy.

7. Results Posted in the Registry

• Wait List: no Therapy: mean 12.40 versus 32.40 on control; difference -20.00 (95% CI -23.69 to -16.31), standardized effect d = -1.38; z-test p = <0.001.
• Wait-list: Thought Field Therapy: mean 20.60 versus 32.40 on control; difference -11.80 (95% CI -15.52 to -8.08), standardized effect d = -0.83; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean difference describe?

A mean is the average value of a continuous outcome in a group. A mean difference compares the averages for two groups or time points. Its sign depends on which group or time point is subtracted from the other, so the direction of subtraction must be understood before interpreting a positive or negative estimate. For this trial, the posted re-calculation gives a mean difference of -20.0 for the initial comparison, based on listed arm means of 32.4 and 12.4. The registry reports its own final-values mean difference as 20.9. Because the registry does not fully specify the relationship between these outcome definitions, the estimates should be described separately.

What does a t-test assess?

The registry reports two-sided t-tests. In general, a two-sided test evaluates whether the data are compatible with a difference in either direction, rather than testing only for a difference favoring one specified direction. A t-test uses the observed group averages and their variation to assess the evidence against a null difference, under assumptions about the data and the analysis. The registry does not report diagnostic checks for those assumptions, details of any adjustment, or a full analysis plan. The p-values reported for the comparisons are < 0.001. These values describe evidence under the test assumptions; they are not measures of the size of the difference.

What does the re-calculation use?

The posted re-calculations are simple unadjusted comparisons based on the arm-level summaries. They report mean differences, confidence intervals, pooled standard deviations, standardized effect estimates, and p-values. The two initial arm counts are 114 and 122, and the arm-level standard deviations are 14.7 and 14.2. No covariate adjustment is represented in these simple re-calculations. They are not a substitute for a more detailed analysis based on participant-level data and a prespecified analysis plan.

The standardized effect estimates are -1.385 for the initial comparison and -0.835 for the comparison between the initial treatment arm and the later treated wait-list group. Standardization expresses a difference in units of pooled variability, which can help compare differences measured on the same scale across settings. It does not remove design limitations, make different analysis populations equivalent, or show that the outcome is clinically important. The direction again follows the comparison’s sign convention.

Why is the wait-list sequence important?

The initial wait-list period provides a randomized comparison between participants assigned to therapy and those assigned to no therapy before the one-week assessment. Once the wait-list group receives therapy, its later outcome is measured after a different treatment sequence. The later assessment can describe the group’s scores after delayed treatment, but it no longer provides a concurrent randomized estimate against an untreated control. Changes over time can reflect more than the intervention, and a before-and-after comparison within a group does not isolate those influences in the way a concurrent randomized comparison can.

The primary outcome description specifies assessments at baseline, after one week, and after a further week for the wait-list group. It also states that non-attenders at the one-week assessment were excluded, with non-attenders at the later assessment excluded from the wait-list treated arm. The registry does not report how missing outcome values were otherwise handled, whether any imputation was used, or whether the reported analyses followed an intention-to-treat principle. Such details matter because an analysis restricted to attendees may differ from an analysis retaining all randomized participants.

Why does variation matter?

Standard deviations describe how dispersed participant scores are around a group mean. Variation affects the precision of a mean estimate: greater dispersion generally makes it harder to estimate a group average or difference precisely when other features are held constant. The posted summaries give standard deviations for each arm and pooled standard deviations for the recalculated comparisons. In general, pooling combines information about variability across the groups under assumptions that make the pooled estimate appropriate. The record does not report a variance-assumption check for the sponsor’s t-test.

Over-dispersion is a general concern when a statistical model assumes less variability than the data contain. If variability is underestimated, standard errors can be too small and confidence intervals too narrow. This trial reports a continuous score and standard deviations; the registry does not report an over-dispersion assessment or a model designed to address it. That concept is therefore relevant as a general statistical issue, not as a documented problem in this study.

What do the analyses not establish?

The record lists 4 outcome measures and 8 analyses, while one secondary outcome is posted. It does not report whether multiple comparisons were anticipated, how the primary analysis was selected among possible analyses, or whether multiplicity adjustments were applied. When many outcomes or comparisons are tested, the chance of observing a small p-value by chance can increase; methods for managing that issue depend on the analysis plan and scientific aims. No particular multiplicity procedure can be inferred from the registry information.

Likewise, the listed masking status does not establish how masking was implemented or whether it was maintained. The registry does not report who was masked, participant-level baseline characteristics, details of treatment adherence, or a complete protocol. These omissions do not change the arithmetic of the posted summaries, but they limit how fully the estimates can be assessed for risk of bias and reproducibility.

10. Confidence Intervals and Precision

A confidence interval gives a range of values compatible with the observed estimate under the statistical method and its assumptions. It conveys statistical precision: a narrower interval indicates less uncertainty in the estimate than a wider interval, all else equal. It is not a guarantee that the true value lies in the interval, and it does not account automatically for bias, missing data, or an unsuitable analysis.

For the initial comparison, the recalculated mean difference is -20.0 units, with a 95% confidence interval from -23.69 to -16.31. The interval does not include zero, and the corresponding unadjusted p-value is < 0.001. For the comparison with the wait-list group after it received therapy, the estimate is -11.8, with a 95% confidence interval from -15.52 to -8.082 and p < 0.001. Both intervals are reported for the unadjusted recalculations and should be interpreted in light of their different comparison structures.

The registry’s own final-values analysis for the initial groups reports a mean difference of 20.9 and a 95% confidence interval from 16.1 to 24.0. The direction and magnitude differ from the recalculated change comparison. The registry does not report enough detail to reconcile the estimands fully, so precision for one estimate should not be transferred to the other.

Confidence intervals describe sampling uncertainty under the analysis method; they do not capture every source of uncertainty. They cannot correct for possible selection caused by excluding non-attenders, clarify the effect of unspecified outcome handling, or account for unreported design choices. The reported intervals are also tied to the score scale and the study’s analysis population. They should be read as estimates of the specified comparisons, not as a complete statement about all possible populations or treatment settings.

11. 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 display gives planning illustrations based on the observed effects, not the trial’s original design justification. The original sample-size assumptions are not reported in the registry record. For these illustrations, the stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation; each planning scenario targets a stated fraction of the observed effect. The figures for this continuous outcome are participants per arm, not total event counts.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding per-arm count for a design with a 90% chance. These are design-planning probabilities under a model, not probabilities that the treatment effect is true. This is not observed power.

The rows plan for 100%, 75%, and 50% of the observed effect. Planning for a smaller effect requires more participants per arm to detect it at the same stated power, because a smaller difference is harder to distinguish from sampling variation. The display lists 9 and 11 participants per arm for the 80% and 90% targets when planning for 100% of the observed effect in the initial comparison; for 75%, it lists 15 and 20, and for 50%, 33 and 44. For the later treated wait-list comparison, the corresponding figures are 23 and 31, then 41 and 54, then 91 and 121. These are illustrations, not observed trial outcomes or evidence that the original trial was designed with those assumptions.

12. 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 = -20.000; pooled standard deviation σ = 14.444.
2. n = 2 × (1.96 + 0.842)² × 14.444² / -20.000² = 8.2, rounded up to 9 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 <- -20                      # observed difference in means
sigma <- 14.4436142101965                     # 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(1681628)
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 = 9 (80% power), 11 (90% power)
 75% of observed effect: n per arm = 15 (80% power), 20 (90% power)
 50% of observed effect: n per arm = 33 (80% power), 44 (90% power)
power.t.test: 9.3 
Simulated power at n = 9 per arm: 0.787 (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.

13. Analyses Reported by the Sponsor

The registry’s reported analysis for the initial treatment and wait-list groups is a two-sided t-test of final values. It gives a mean difference of 20.9, a 95% confidence interval from 16.1 to 24.0, and p < 0.001. Separate reported analyses list final-value results for the individual groups as well.

The unadjusted re-calculation compares the posted arm-level means and reports a mean difference of -20.0, with a 95% confidence interval from -23.69 to -16.31 and p < 0.001. These estimates are not numerically identical and have opposite signs. The re-calculation is presented as a change comparison, whereas the registry labels its reported analysis as a final-values comparison. The registry record does not provide enough detail about coding, time points, or analysis populations to determine fully how the two definitions relate.

The reported values should therefore be read with their stated analysis descriptions rather than treated as direct replications of the same estimate. The record does not report additional analysis details that would allow a more complete reconciliation. The re-calculation is useful as a transparent comparison of the posted summaries, but it does not replace the registry’s own reported analysis or establish which analysis was prespecified.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 4 outcome measures and 8 analyses, while one secondary outcome is posted. The primary outcome concerns change in the civilian PTSD checklist score. The secondary outcome concerns the percentage with a score above the diagnostic threshold, PCL-C > 50, before and after treatment in the treatment and control groups. The registry does not report a complete protocol or analysis plan that would show how these outcomes and analyses were prioritized.

Multiplicity arises when several outcomes, time points, or group comparisons are examined. In general, conducting multiple tests can increase the chance that at least one result appears statistically significant even when there is no underlying difference. Whether to adjust for multiplicity depends on the study aims and prespecified analysis plan. The registry record does not report whether a multiplicity adjustment was planned or used, so none should be assumed.

Missing-data handling is also incompletely described. The primary endpoint population excludes participants who did not attend the one-week assessment, and the wait-list group’s later treated assessment excludes non-attenders at that time. The registry does not report the number of missing observations at each time point beyond the posted group counts, reasons for non-attendance, imputation procedures, or sensitivity analyses. It also does not state whether the analyses retained participants according to their randomized assignments regardless of attendance.

The registry gives no detailed analysis-population definition beyond the endpoint description. It does not report a full intention-to-treat analysis, a per-protocol analysis, or how deviations from assigned treatment were handled. These omissions limit assessment of whether the observed groups remained comparable after exclusions. They also make it difficult to determine how robust the estimates are to different reasonable approaches to missing outcomes and analysis populations.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death (0.4%).

The registry’s adverse-event record lists 250 participants at risk in the Thought Field Therapy group for serious adverse events and one participant affected. For other adverse events, it lists 250 at risk and two affected. The separately recorded number of deaths is not populated. The top serious-event term is listed as Death, with 0.4%.

These entries describe the registry’s event counts and should be read as recorded, without inferring more detail about event timing, severity beyond the category, relatedness to the intervention, or clinical course. The record does not report event narratives or explain the relationship between the serious-event count, the listed serious-event term, and the unpopulated deaths field. It also does not provide comparative adverse-event counts for the wait-list arm in this entry.

The figures are not a complete assessment of safety in all settings. They reflect the recorded trial information and the stated at-risk group, and do not establish whether events were caused by the intervention. Absence of further detail in the registry should not be interpreted as evidence that no other events occurred.

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

The posted results show that the analyzed groups had different mean checklist outcomes under the comparisons reported. The initial unadjusted re-calculation gives a mean difference of -20.0 units, with a 95% confidence interval from -23.69 to -16.31 and p < 0.001. The later comparison involving the wait-list group after it received therapy gives a mean difference of -11.8, with a 95% confidence interval from -15.52 to -8.082 and p < 0.001. The registry’s own analysis of final values reports a mean difference of 20.9, with a 95% confidence interval from 16.1 to 24.0 and p < 0.001.

These figures describe different stated analyses and should not be collapsed into a single estimate. The registry does not give enough detail to reconcile the final-values analysis with the re-calculated change comparison. The initial randomized wait-list comparison is the most direct comparison of therapy with no therapy before the one-week assessment. The later wait-list assessment followed treatment and does not provide a concurrent randomized untreated comparison.

A statistically small p-value does not establish that an intervention is effective, establish a causal explanation, or indicate the importance of an outcome to participants. The mean difference is measured on a symptom checklist scale; the record does not report a threshold for a clinically important change. The screening cutoff above 50 is a diagnostic-score definition, not a minimum meaningful change threshold.

The results also do not by themselves show whether findings generalize beyond the enrolled population or to other delivery settings. The record lacks details about participant characteristics, training, adherence, missing-data methods, and a complete analysis plan. The appropriate interpretation is therefore limited to the posted outcome summaries and their stated comparisons, with uncertainty about analysis definitions and potential effects of exclusions.

17. Why This Trial Matters Statistically

Statistically, this trial is notable for evaluating a community-delivered intervention in a randomized parallel-group design. The intervention was delivered by trained community leaders, while the initial comparator group waited without therapy until its first follow-up assessment. That design creates a direct randomized comparison during the initial period and a separate descriptive assessment after the wait-list group later received therapy.

The outcome is a continuous symptom score, so mean differences and their confidence intervals communicate both the direction and precision of group comparisons. The posted recalculations show sizeable differences relative to the reported score variability, while the confidence intervals convey the range of values compatible with those analyses. The separate registry analysis, however, is described as a final-values comparison and does not match the recalculated change estimate exactly. This highlights why clarity about the estimand—what quantity an analysis is intended to estimate—is essential when interpreting trial results.

The trial also illustrates how analysis populations affect interpretation. The endpoint description excludes non-attenders at follow-up, but the registry does not document missing-data methods or provide a complete account of participants included at each analysis stage. Such details are central to evaluating whether the randomized comparison was preserved.

Finally, planning illustrations based on observed effects can help readers understand how sample-size requirements change with the effect targeted, but they are not evidence of the original design assumptions or of achieved power. The registry does not report the original sample-size justification. Read together, these features make transparent distinctions between randomized comparisons, post-treatment follow-up, statistical precision, and unreported analytic choices particularly important.

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.

Several trial-specific limitations affect interpretation. First, the registry does not fully reconcile its reported final-values analysis with the unadjusted re-calculation based on listed arm summaries. The analysis descriptions differ, and the record does not report enough information about score coding, time points, or analysis populations to establish whether they target the same quantity.

Second, the primary endpoint population excludes non-attenders at follow-up, including non-attenders at the later assessment in the wait-list treated arm. The registry does not report reasons for non-attendance, detailed missing-data handling, or sensitivity analyses. The posted group counts therefore cannot alone establish whether exclusions altered comparability or the estimated outcomes.

Third, the wait-list group received therapy after its initial no-treatment assessment. Its later treated result is not a randomized comparison with a concurrent untreated group, and time-related changes cannot be separated from treatment effects using that later result alone.

The registry records quadruple masking but does not identify who was masked or how masking was maintained. It does not report allocation concealment, detailed participant characteristics, treatment adherence, community-worker training procedures, or a complete protocol and statistical analysis plan. The number of listed outcome measures and analyses also raises questions about multiplicity that cannot be resolved because adjustment procedures are not reported.

The outcome label varies between PLC-C and PCL-C in the registry, and the record does not explain the difference. Finally, the adverse-event entry contains limited detail and does not provide a comparative wait-list event count in the posted safety information. These limitations do not invalidate the reported summaries, but they constrain conclusions about bias, reproducibility, safety, and generalizability.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Community Self-help Using Thought Field Therapy in a Traumatised Population in Uganda: a Randomised Trial — NCT01681628.
• Registry record last updated 2017-05-01; status last verified 2017-03.
• Results first posted 2017-05-01.
• 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

NCT01681628 is a completed randomized, parallel-group trial of Thought Field Therapy delivered by trained community leaders in Uganda. It enrolled 256 participants and recorded quadruple masking. The primary outcome was change in the civilian PTSD checklist score, measured at baseline and follow-up; the wait-list group received therapy after its initial no-treatment assessment.

The posted unadjusted re-calculation reports a mean difference of -20.0 units for the initial Thought Field Therapy versus wait-list comparison, with a 95% confidence interval from -23.69 to -16.31 and p < 0.001. The registry’s own final-values analysis reports a mean difference of 20.9, with a 95% confidence interval from 16.1 to 24.0 and p < 0.001. The record does not provide enough detail to reconcile these estimates fully. Missing-data methods, original sample-size assumptions, and a complete statistical analysis plan are not reported in the registry record.