Two Stepped Care Models for Posttraumatic Stress Disorder (PTSD) Among Cambodian Refugees With PTSD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Two Stepped Care Models for Posttraumatic Stress Disorder (PTSD) Among Cambodian Refugees With PTSD (NCT01542372), a randomized phase 4 study. The primary outcome is Change in the PTSD Checklist (PCL) at 12 Weeks in Step II.

2. Clinical Question

3. Trial Design

The registry describes a Phase 4, randomized, single-masked, parallel-group trial sponsored by Massachusetts General Hospital. The study was terminated. Enrollment was 114, and the registry lists a start of 2012-04 and primary completion of 2014-03. The identity of the party or parties masked is not reported in the registry record, nor is the method used to generate or conceal the random allocation.

The study addressed two stepped-care models for PTSD among Cambodian refugees. Both began with medication, described in the registry summary as an SSRI/SNRI approach with paroxetine as the first-line agent. Patients whose PTSD remained were eligible for Step II, in which the models diverged: one offered medication augmentation and the other culturally sensitive CBT augmentation. The posted primary outcome is specifically a Step II outcome, rather than an outcome for all 114 enrolled participants.

The posted primary endpoint has 11 participants in each arm. The registry does not report how many participants were randomized at the start of the trial, how many reached Step II, or how participants moved through the first step. It also does not report the length of follow-up beyond the endpoint timeframe, the randomization ratio, the masking procedures, or reasons for termination. These omissions limit what can be reconstructed about the trial's conduct from the registry record alone. A detailed protocol and a full account of recruitment, treatment delivery, and participant flow are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: In Step 1 of the protocol, patients were given an SSRI/SSRN to determine eligibility for Step 2. If a patient still had PTSD (PCL \> 30) after Step I, the patient was assigned to an arm of Step 2, which was Medication or CBT augmentation. Of 114 patients started in Step I, 22 met criteria for entry to Step 2, the study proper, addressing study aims.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Patients with PTSD checklist scores of 30 or above were eligible for Step II, which was medication augmentation or CBT augmentation.

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: Patients who meet inclusion criteria for Step 1 such as PTSD and impairment.

6. Primary Endpoint

The registry lists the primary outcome as Change in the PTSD Checklist (PCL) at 12 Weeks in Step II, measured over Baseline and 12 weeks.

The primary endpoint is mean change in the PTSD Checklist (PCL) score from baseline to 12 weeks in Step II. The PCL is a measure of PTSD severity. Its total score ranges from 17 to 85, with higher scores indicating greater psychopathology. Change scores were calculated, so the posted endpoint expresses change over time rather than the score at a single assessment.

A change measure is used to describe how scores shifted between baseline and follow-up. Negative mean changes in the posted arms indicate that the average score decreased over that interval. The comparison between arms asks whether their average changes differed; it is not simply a test of whether scores changed within either arm. The endpoint is restricted to participants eligible for Step II, defined in the record as those with PCL scores of 30 or above. It therefore addresses the relative Step II approaches among that eligible group, not the full stepped-care pathway for every enrolled participant.

The record identifies the measure, timeframe, scale direction, and Step II eligibility criterion. It does not report a clinically important threshold for interpreting a change, a target difference between arms, or details of how missing endpoint measurements were handled. Those details should not be inferred from the posted means and standard deviations.

7. Results Posted in the Registry

• CBT Augmentation: mean -11.40 versus -5.00 on control; difference -6.40 (95% CI -13.60 to 0.80), standardized effect d = -0.74; z-test p = 0.081.

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.

Change in the HSCL Anxiety Scale at 12 Weeks in Step II

• 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:
  • CBT Augmentation: mean -0.60 versus -0.40 on control; difference -0.20 (95% CI -0.84 to 0.44), standardized effect d = -0.26; z-test p = 0.540.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CBT Augmentation and the control. The interval gives the range of effects that remain compatible with the data.

Change in the HSCL Depression Scale at 12 Weeks in Step II

• 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:
  • CBT Augmentation: mean -0.50 versus -0.30 on control; difference -0.20 (95% CI -0.79 to 0.39), standardized effect d = -0.29; z-test p = 0.503.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CBT Augmentation and the control. The interval gives the range of effects that remain compatible with the data.

Change in the SCL Anger Severity at 12 Weeks in Step II

• 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:
  • CBT Augmentation: mean -0.90 versus -0.30 on control; difference -0.60 (95% CI -1.23 to 0.03), standardized effect d = -0.80; z-test p = 0.061.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CBT Augmentation and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes cover anxiety, depression, anger severity, a culturally sensitive complaint profile, and health status measured with the SF-12, alongside the primary PTSD severity outcome. The reported secondary comparisons are continuous changes from baseline to 12 weeks in Step II. Across the secondary endpoints for which arm comparisons are posted, the estimated differences generally favor a greater mean decrease in the CBT augmentation arm, while the precision varies and the intervals include values compatible with little or no difference. This broad pattern is directionally consistent with the primary estimate, but it is not an independent confirmation of it.

Secondary outcomes can help show whether an apparent pattern in the primary measure is accompanied by changes in related domains. They also raise interpretive challenges: multiple outcomes create multiple opportunities for chance findings, and measures may be correlated rather than representing entirely separate evidence. The registry record does not report a multiplicity adjustment or a hierarchy specifying which secondary outcomes were confirmatory. Accordingly, the unadjusted secondary results carry less weight than a single prespecified primary comparison and should be treated as exploratory evidence.

The record lists five secondary outcome measures, but reported analyses are empty for each. The numeric comparisons are therefore recalculations from posted summary statistics, not sponsor-reported inferential analyses. No clinical-importance thresholds or interpretation guidelines for the secondary measures are reported in the registry record. Nor does the record provide a joint analysis across endpoints. The secondary pattern can be considered alongside the primary result, but it cannot resolve the uncertainty in that result or establish broad benefit across all measured domains.

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

• Change in the HSCL Anxiety Scale at 12 Weeks in Step II
• Change in the HSCL Depression Scale at 12 Weeks in Step II
• Change in the SCL Anger Severity at 12 Weeks in Step II
• Change in the Cambodian Culturally Sensitive Complaint Profile at 12 Weeks in Step II
• Change in the SF-12 at 12 Weeks in Step II

9. Statistical Interpretation

10. Statistical Methods Explained

The registry identifies a continuous primary outcome summarized as a mean and standard deviation. A mean gives the average change in PCL score within an arm. The standard deviation describes how spread out individual change scores are around that arm's mean; it is not the standard error of the mean and does not describe uncertainty in exactly the same way. The posted primary means are -5.0 and -11.4, with standard deviations of 9.6 and 7.5, respectively. Each Step II arm has 11 participants.

What does the arm comparison estimate?

The main contrast is the difference between the two arm means, with CBT augmentation compared with medication augmentation. The recalculated estimate is -6.4 PCL units. A negative contrast means the average change in the CBT arm is more negative than the average change in the medication arm under this subtraction order. It is a contrast between group averages, not a statement that every participant had that change. It also does not describe how much either arm changed from baseline in a way that isolates treatment from all other influences.

The registry comparison is an unadjusted recalculation from posted summary statistics. It does not include adjustment for baseline values, additional covariates, clustering, or other model features. Whether such approaches were specified or used by the sponsor is not reported in the registry record. The record lists zero reported analyses, so the displayed recalculation should not be attributed to a sponsor's prespecified analysis.

How should the standard deviation be understood?

Standard deviations quantify the dispersion of participant-level scores within each arm. Greater dispersion means individual changes varied more around the arm average. The standard error, by contrast, describes the estimated sampling uncertainty of a mean. The arm standard errors associated with the posted primary comparison are 2.895 and 2.261. In general, uncertainty in a mean depends both on how variable observations are and how many observations contribute; small groups can leave estimates imprecise even when a mean is available.

In general, over-dispersion matters when outcomes vary more than a chosen statistical model expects. A model that understates variability can produce uncertainty intervals that are too narrow and tests that are too optimistic. The registry does not report a model-based analysis, a variance check, or a discussion of over-dispersion for this continuous endpoint, so it is not possible to say whether those issues were assessed in the trial's own analysis.

What does a confidence interval communicate?

A confidence interval gives a range of values that are reasonably compatible with the data under the method used to calculate it. It is not a range containing a fixed proportion of individual participant changes, and it is not a guarantee that the true effect lies inside the reported limits. The recalculated interval for the primary difference runs from -13.6 to 0.799. Because it includes zero, the interval is compatible with no difference as well as with differences in either direction. The confidence level and interval method are not reported in the registry record, so they should not be inferred from the interval alone.

What does the p-value say?

The recalculated unadjusted p-value for the primary comparison is 0.081. In general, a p-value describes how surprising results at least as inconsistent with a null model would be if that model and the assumptions of the test were correct. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. A p-value also does not communicate the range of effect sizes compatible with the data; that is why the estimate and its confidence interval matter as well. The specific test procedure used for the recalculation is not reported in the registry record.

The registry's comparison estimate includes a standardized effect-size value of -0.743. Such a standardized measure expresses a mean contrast relative to a measure of variability, allowing outcomes on different scales to be discussed in a common unit. It does not replace the PCL-unit difference, and its interpretation depends on the variability used in standardization. No sponsor analysis or prespecified threshold for interpreting this standardized value is reported in the registry record.

What cannot be reconstructed?

The record does not report an analysis population definition, an intention-to-treat or per-protocol designation, a missing-data procedure, a covariate-adjustment plan, or handling of protocol deviations. It does not report a prespecified statistical analysis plan, interim analyses, a sample-size justification for the original trial, or the original design assumptions. These are distinct from the planning illustrations associated with the recalculated effects. Without these details, the posted summary statistics allow a basic unadjusted comparison but not a full audit of the trial's statistical methods.

11. Confidence Intervals and Precision

The primary difference in mean PCL change is estimated at -6.4 units for CBT augmentation compared with medication augmentation. Its confidence interval extends from -13.6 to 0.799. The interval crosses zero, meaning that the recalculated results are compatible with no mean difference as well as with a difference in either direction. Its upper endpoint is near zero, while its lower endpoint allows a substantially more negative difference. The estimate therefore has notable uncertainty.

Precision concerns how narrowly the data locate an effect, not whether the effect is important to patients. Here, only 11 participants per arm contribute to the posted Step II primary results. The standard deviations, 9.6 and 7.5, also show that participant changes varied around their respective arm means. With a limited number of observations and variability in scores, a point estimate can be quite different from the range of effects compatible with the data.

The unadjusted p-value of 0.081 is consistent with the interval's inclusion of zero, but the p-value and interval answer different questions. The interval communicates the range of effect estimates compatible with the calculation; the p-value summarizes how the data compare with a null model under test assumptions. Neither should be read in isolation. The confidence level and calculation method for the interval are not reported in the registry record.

These uncertainty statements are based on a simple recalculation from the posted summaries, not a sponsor-reported model. They do not account for unknown features such as missing observations, covariate adjustment, or the analysis population. The record does not give a clinically meaningful difference threshold, so the interval cannot be classified against such a threshold from the available information.

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 figures are planning illustrations based on the observed effect, not the trial's original sample-size justification. For these continuous outcomes, the entries are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and detection of the stated fraction of the observed effect. The original trial's design assumptions are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a design that would have an 80% chance of detecting the specified effect under the stated assumptions, if that effect and assumptions were correct. The 90% power column gives the corresponding planning target for a higher chance of detection under those same assumptions. For example, planning for 100% of the observed primary effect gives 29 participants per arm for 80% power and 39 participants per arm for 90% power. Planning for 75% of the observed effect gives 51 participants per arm for 80% power and 68 participants per arm for 90% power. The 50% planning row requires more participants per arm than either of those rows.

Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants. These figures depend on using the observed effect as a starting point; an effect estimated from a small group may be unstable. They are illustrations for future planning, not observed power for this trial. They do not show the probability that the trial's findings are true or that the observed result will recur.

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 = -6.400; pooled standard deviation σ = 8.614.
2. n = 2 × (1.96 + 0.842)² × 8.614² / -6.400² = 28.4, rounded up to 29 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 <- -6.4                      # observed difference in means
sigma <- 8.61423240921674                     # 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(1542372)
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 = 29 (80% power), 39 (90% power)
 75% of observed effect: n per arm = 51 (80% power), 68 (90% power)
 50% of observed effect: n per arm = 114 (80% power), 153 (90% power)
power.t.test: 29.4 
Simulated power at n = 29 per arm: 0.795 (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 many outcomes, comparisons, or analysis choices are examined. Even if each test is conducted at a conventional threshold, considering many tests increases the chance that at least one result will appear noteworthy by chance. The registry lists one primary endpoint and five secondary outcome measures. It does not report whether the secondary outcomes were prespecified as confirmatory, whether a hierarchy was used, or whether any adjustment for multiple testing was planned or performed. The secondary recalculations are unadjusted, so isolated p-values should not be interpreted as if they were the sole planned test.

Missing-data handling is also important because the posted endpoints are measured over time. The registry record provides arm-level sample sizes for the Step II outcomes, but does not explain whether all randomized participants contributed measurements, why observations might be absent, or how missing outcomes were addressed. It does not report imputation, a likelihood-based model, sensitivity analyses, or a complete-case rule. No particular missing-data method should be assumed.

The analysis population is not described. The registry does not state whether results use all randomized participants, only those who entered Step II, participants who completed treatment, or another defined group. The Step II eligibility criterion is reported, but an eligibility description does not establish the analysis population or how participants who did not progress from Step I were handled. The number of participants randomized, reasons for discontinuation, adherence, and participant flow are not reported in the registry record.

These gaps matter because analysis choices can affect both the estimate and its uncertainty. The registry does not report a statistical analysis plan, covariate adjustments, interim analyses, or rules for deviations from the intended treatment. A reader can interpret the posted unadjusted summaries, but cannot determine whether they match a prespecified analysis or how robust the conclusions are to alternative approaches.

15. 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 adverse-event entry covers the group labeled Step I and Step 2, with 114 participants at risk. The record reports zero participants affected by serious adverse events, zero affected by other adverse events, and zero deaths. No serious adverse-event terms are listed.

These figures describe what is entered in the registry record for the specified group. They do not give event-level detail, timing, severity categories beyond the serious/other distinction, or exposure duration. The record does not explain the method or period of adverse-event collection, whether events were actively solicited, or how events were attributed to treatment. A zero reported count is therefore not evidence that no unrecorded events occurred, and these figures should not be extended beyond the registry's stated group and reporting.

The safety entry also covers the full Step I and Step 2 group rather than giving separate counts for the Step II arms. It cannot be used to compare safety between medication augmentation and CBT augmentation. The registry provides no further adverse-event breakdown that would support such a comparison.

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

The posted results describe average PCL score changes among Step II participants and a recalculated contrast between medication augmentation and CBT augmentation. The CBT arm had a more negative average change, and the estimated difference was -6.4 PCL units. The confidence interval includes zero and spans values in both directions, while the unadjusted p-value is 0.081. Together, these results indicate that the comparative estimate is uncertain; the observed point estimate alone does not settle the size or direction of the difference.

The endpoint is a scale score, not a direct report of each participant's experience or a measure of every aspect of PTSD. A group mean does not mean that all participants changed in the same way. The record does not state a threshold for a clinically important PCL change, so the numeric contrast cannot be classified as clinically important or unimportant using the registry information alone.

The Step II comparison does not answer whether the two complete stepped-care models differ across the full course of care. It does not give the first-step outcome, the proportion of participants who reached Step II, or results for those who did not meet the Step II eligibility criterion. The trial's termination status and missing details about participant flow further limit interpretation of the posted results.

These findings do not establish that either intervention works or does not work, and they should not be generalized automatically to other populations or care settings. The registry record reports a study focused on Cambodian refugees with PTSD in primary care, and the endpoint applies to participants eligible for Step II. The posted estimates are useful as a description of the available summary data, but uncertainty, small Step II arm sizes, and incomplete information about the analysis plan constrain stronger conclusions.

17. Why This Trial Matters Statistically

This trial is statistically informative because it compares two distinct Step II strategies within a stepped-care framework, while also illustrating how a broad enrollment figure can differ from the group contributing to a particular endpoint. Enrollment was 114, but the primary Step II results list 11 participants in each arm. Interpreting those figures correctly requires keeping the overall trial population separate from the participants represented in the posted endpoint.

The result also shows why an effect estimate should be read together with its interval and analysis context. A more negative mean change in one arm may look compelling in isolation, but the interval for the recalculated difference includes zero and the estimate comes from a small group. The uncertainty is part of the finding, not a technical detail to set aside. The same principle applies to the secondary outcomes, where multiple unadjusted comparisons deserve caution.

The registry's limited reporting draws attention to the importance of transparent statistical plans, participant flow, missing-data methods, analysis populations, and multiplicity decisions. Those details help readers distinguish prespecified confirmatory analyses from exploratory recalculations and evaluate the credibility of estimates. They are not reported in the registry record for this trial.

Finally, the planning illustrations demonstrate how sample-size needs change when the target effect is made smaller. They should not be mistaken for the original design rationale or for evidence about the trial's observed power. Used with that distinction in mind, the record offers a practical example of the different roles played by a point estimate, uncertainty interval, p-value, and prospective planning calculation.

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 most direct limitation is the small number of participants in each posted Step II primary arm: 11 per arm. The primary estimate is consequently imprecise, as reflected by a confidence interval that includes zero. The registry does not report how many participants were initially randomized, how many proceeded through Step I, or why only the posted number contributed to the Step II results. The relation between overall enrollment and the endpoint population cannot be reconstructed from the record.

The registry lists the trial as terminated, but does not report the reason for termination. It also does not report whether the study stopped early relative to its plan or whether termination affected recruitment, follow-up, or analysis. No explanation should be inferred from status alone. The original sample-size target and design justification are likewise not reported.

The posted inferential comparisons are recalculations from summary statistics; the registry lists no reported analyses. A prespecified statistical analysis plan, covariate adjustment, missing-data approach, analysis population, interim monitoring, and multiplicity procedure are not reported. The confidence-interval method and confidence level are also not given. These omissions make it difficult to assess how the posted estimates relate to the sponsor's intended analysis.

Additional limits include the lack of reported participant-flow detail, treatment adherence, reasons for missing measurements, and clinically important thresholds for the outcome scales. The primary endpoint applies to Step II eligibility rather than the full enrolled cohort. The adverse-event counts are not separated by the two Step II arms and have no reported collection details. These limitations restrict both comparative interpretation and generalization beyond the population and endpoint described in the registry.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Two Stepped Care Models for Posttraumatic Stress Disorder (PTSD) Among Cambodian Refugees With PTSD — NCT01542372.
• Registry record last updated 2017-05-17; status last verified 2017-04.
• Results first posted 2017-05-17.
• 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 describes a randomized, single-masked, parallel-group Phase 4 trial of two stepped-care models for PTSD among Cambodian refugees. Both models began with medication; the Step II comparison was medication augmentation versus CBT augmentation for participants eligible by PCL score. The primary endpoint was mean PCL change from baseline to 12 weeks in Step II.

The posted Step II primary results include 11 participants per arm. The recalculated, unadjusted mean difference for CBT augmentation compared with medication augmentation is -6.4 units, with a confidence interval that includes zero and an unadjusted p-value of 0.081. No sponsor-reported analyses are listed. The record lists five secondary outcomes, with no reported analyses, and zero serious adverse events, other adverse events, and deaths in the combined group at risk. The study status is terminated; the reason and key statistical-analysis details are not reported in the registry record.