Treatment of Post-Traumatic Brain Injury (TBI) Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Treatment of Post-Traumatic Brain Injury (TBI) Depression (NCT00233103), a randomized phase 2 study. The primary outcome is Depression at Baseline.

2. Clinical Question

The clinical topic is treatment of post-traumatic brain injury depression, but the statistical question answered by the primary result is narrower: how the recorded baseline HAM-D means compare between the randomized groups. The posted baseline mean was 27.5 for sertraline and 25.2 for placebo. The unadjusted mean difference was 2.3, with a 95% confidence interval from -2.363 to 6.963 and p = 0.334. Those estimates describe the baseline groups; they do not establish whether scores changed during the intervention. The post-intervention secondary measures offer additional descriptive information, but they do not replace a posted primary treatment outcome.

3. Trial Design

The registry record describes a completed phase 2 clinical trial of sertraline versus placebo for post-traumatic brain injury depression. It lists randomized allocation, double masking, and a parallel-group model. Enrollment was 52. The record gives a start date of 2003-06 and a primary completion date of 2008-09. The sponsor is Icahn School of Medicine at Mount Sinai.

The intervention description specifies daily oral sertraline, starting at 25mg and increasing to therapeutic levels up to 200mg. The placebo arm is described as receiving placebo for 10 weeks. The record does not report further details about treatment visits, dose adjustment, adherence, or the procedures used to preserve masking.

Outcome summaries are posted for 22 participants assigned to sertraline and 19 assigned to placebo. The registry record does not explain how these outcome counts relate to the enrollment total, or whether they represent all participants assigned, all participants treated, or participants with available measurements. It does not report analysis populations or the handling of missing outcomes.

The posted primary outcome is titled Depression at Baseline and is measured in units on a scale. The record describes the HAM-D as a clinician-rated depression-severity measure and includes self-report and DSM-IV diagnosis in the outcome description. The two posted secondary outcomes are BAI, measured immediately post-intervention, and Life-3, measured immediately post-intervention at 10 weeks. Although the trial is a treatment comparison, the primary outcome information posted is baseline data. The registry record does not post a primary post-treatment score, a change-from-baseline estimate, or a treatment contrast for a later HAM-D assessment.

The record lists four outcome measures, while two secondary outcomes are posted. It lists no reported analyses. The original statistical analysis plan and the study's own sample-size assumptions are not reported in the registry record. Consequently, design features beyond the recorded allocation, masking, model, and intervention descriptions should not be inferred.

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 Depression at Baseline, measured over baseline.

The primary endpoint is titled Depression at Baseline. It uses the 17-item Hamilton Rating Scale for Depression (HAM-D), a clinician-rated measure of depression severity. The registry record describes score bands: 0-7 is considered normal, 8-13 mild depression, 14-18 moderate depression, 19-22 severe depression, and 23 or higher very severe depression. This scale provides a structured way to quantify symptom severity rather than relying only on a diagnostic label.

The reported outcome is a mean and standard deviation for each group. A mean gives the group's average recorded score; the standard deviation describes how spread out the individual scores are around that average. At baseline, the posted mean was 27.5 with a standard deviation of 7.1 in the sertraline group, and 25.2 with a standard deviation of 8.0 in the placebo group.

Baseline measurement can describe the groups before the intervention and help readers understand the severity represented in the outcome sample. It is not, by itself, a measure of response to treatment. The registry record does not post a later HAM-D outcome or a change from baseline as the primary result. Thus, the primary outcome as recorded can be interpreted as a baseline group comparison, not as evidence about whether depression severity changed during the trial.

7. Results Posted in the Registry

• Sertraline: mean 27.50 versus 25.20 on control; difference 2.30 (95% CI -2.36 to 6.96), standardized effect d = 0.31; z-test p = 0.334.

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.

BAI

• 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:
  • Sertraline: mean 11.10 versus 13.10 on control; difference -2.00 (95% CI -9.21 to 5.21), standardized effect d = -0.17; z-test p = 0.586.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sertraline and the control. The interval gives the range of effects that remain compatible with the data.

Life-3

• 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:
  • Sertraline: mean 6.30 versus 4.90 on control; difference 1.40 (95% CI -2.29 to 5.09), standardized effect d = 0.22; z-test p = 0.458.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sertraline and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes are BAI and Life-3, both measured after intervention. Their unadjusted mean differences point in opposite directions, and the confidence intervals for both include zero. Taken together, the estimates do not present a consistent, precisely estimated separation between the groups across these secondary measures. They add post-intervention descriptive information, but their uncertainty means they should not be read as decisive evidence for a treatment effect.

The secondary results relate to the primary result in an important way: they address later measurements, while the posted primary outcome is depression at baseline. They therefore provide information about different outcomes and time points, rather than confirming or correcting a primary post-treatment HAM-D estimate. The registry record does not post such a primary treatment contrast. The secondary findings cannot be substituted for it.

These comparisons carry less inferential weight because they are unadjusted recalculations and the registry record lists no reported analyses. The record does not say whether these outcomes were prespecified, how repeated outcomes were handled, or whether multiplicity adjustments were planned or applied. When several outcomes are examined, the chance of finding an apparently notable result by chance can increase; adjustment or a clear hierarchy of outcomes can address that concern. That is a general statistical issue, not a description of a method used in this trial.

The record lists four outcome measures but posts two secondary outcomes. It does not identify the other measures in the posted outcome summaries or explain why they are absent. Readers should therefore treat the available secondary pattern as partial rather than as a complete inventory of all outcome findings. The registry record also does not report the analysis populations or missing-data procedures, which limits assessment of how the observed secondary summaries represent the randomized groups.

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

• BAI
• Life-3

9. Statistical Interpretation

10. Statistical Methods Explained

The posted continuous outcomes are expressed as group means and standard deviations. A mean describes the average observed score in a group, while the standard deviation describes how much individual scores vary around that group average. These summaries are useful for understanding scale values and group-level distributions, but they do not show every participant's score or the shape of the underlying distribution.

What does a mean difference represent?

A mean difference compares the average score in one group with the average in the other. The contrasts discussed for this trial are defined as sertraline minus placebo. A positive estimate means the sertraline group's average was higher on the scale; a negative estimate means it was lower. Interpretation of the direction depends on what higher scores mean for the particular measure. The posted primary comparison is 2.3 points on the baseline HAM-D scale. The BAI contrast is -2.0, and the Life-3 contrast is 1.4. These comparisons are unadjusted recalculations from the posted summary statistics rather than reported sponsor analyses.

Group means alone do not establish whether a difference is important or whether it is attributable to treatment. The outcome timing matters: a baseline contrast describes groups before the intervention, whereas a post-intervention contrast describes measured outcomes after it. A change-from-baseline analysis would address a different question again. The registry record posts the baseline HAM-D comparison and post-intervention secondary summaries, but it does not report a primary post-intervention HAM-D treatment contrast.

How should a confidence interval be understood?

A confidence interval expresses the range of values reasonably compatible with the observed data under the method used. It reflects sampling uncertainty in the estimated contrast. A narrower interval indicates greater precision, while a wider interval indicates less precision. The interval does not show the range of individual participant scores, and it does not guarantee that the true effect lies within its limits. Interpretation also depends on the design, measurement, and analysis used.

For the baseline mean difference, the 95% confidence interval ranges from -2.363 to 6.963. It includes zero, which represents no difference in group means, and includes estimates in either direction. The secondary confidence intervals also include zero. These intervals therefore do not rule out a zero contrast, and their breadth indicates that the estimates are not highly precise. A confidence interval should be considered alongside the point estimate and the clinical meaning of the scale, not reduced to a binary decision based only on whether it crosses zero.

What does a p-value add?

A p-value describes how incompatible the observed data are with a specified null model, assuming the analysis model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. The posted unadjusted p-values are 0.334 for the baseline HAM-D contrast, 0.586 for BAI, and 0.458 for Life-3. They are interpreted with the estimates and confidence intervals; none changes the fact that the primary posted measure is at baseline.

The registry record lists no reported analyses. It therefore does not identify a sponsor-reported inferential procedure for these comparisons, nor does it report whether the analyses were prespecified. The posted comparison statistics are unadjusted recalculations from the means, standard deviations, and group counts. They should not be described as adjusted estimates or as a substitute for a complete analysis plan.

Why can variability matter?

Standard deviations describe observed spread, and greater variability generally makes it harder to estimate a group contrast precisely when other design features are unchanged. General statistical methods may need to account for over-dispersion when observed variability exceeds what a model assumes; otherwise, uncertainty can be understated. That is a general methodological point, not a claim that over-dispersion occurred in this trial. The registry record does not report distributional diagnostics or a model for these outcomes.

Randomization is intended to support a fair comparison of groups by allocating participants without choosing their treatment on the basis of their characteristics. Masking can reduce some sources of bias in treatment delivery, assessment, or reporting. Those design labels do not remove uncertainty from a small outcome sample, guarantee complete follow-up, or establish that the posted baseline groups were identical. The record does not report the randomization sequence, concealment procedures, masking assessment, or missing-data methods.

The sample-size illustrations use a normal approximation and equal allocation, as described in their assumptions. A normal approximation uses a reference distribution to estimate the number of participants needed under stated planning conditions. It is a planning calculation, not an analysis of observed outcomes. The registry record does not report whether the original trial used these assumptions or what its original design calculation was.

11. Confidence Intervals and Precision

Precision is about how narrowly the data estimate a group contrast. The posted baseline HAM-D mean difference is 2.3, with a 95% confidence interval from -2.363 to 6.963. This interval includes zero and extends to values on either side, so the observed baseline contrast is compatible with a range of group differences. Its width also indicates that the estimate is not exact. The p-value of 0.334 does not replace this interval; the interval makes the uncertainty and direction of plausible contrasts more visible.

For BAI, the 95% confidence interval extends from -9.206 to 5.206 around the mean difference of -2.0. For Life-3, it extends from -2.295 to 5.095 around the mean difference of 1.4. Each interval includes zero. The size and direction of the estimates differ across measures, and the intervals leave considerable uncertainty about the underlying group contrasts.

These are group-level estimates based on the outcome counts posted in the registry record, not a description of every participant. The record does not explain why the outcome counts differ from the enrollment total or report whether missingness could affect the estimates. That information would be relevant when judging how representative the analyzed observations are.

Precision does not answer whether a difference is clinically meaningful. That judgment requires context about the scale and what magnitude of change matters, which is not reported in the registry record. For the primary outcome, the central limitation is also one of timing: the reported HAM-D result is at baseline. Even a more precise baseline contrast would not by itself estimate change during treatment.

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 expressed as participants per arm for these continuous outcomes. They are not total event counts. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The original trial's design assumptions and sample-size justification are not reported in the registry record.

The 80% power column represents the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions. The 90% power column uses a higher target chance of detection and therefore generally requires more participants. These percentages describe the planned operating characteristics of a hypothetical study under the stated assumptions; they do not describe the chance that the observed trial result is correct. The figures are not observed power.

Rows that plan for a smaller fraction of the observed effect require more participants per arm. A smaller effect is harder to distinguish from random variation, so a study generally needs more information to detect it at the same significance and power targets. The displayed estimates are tied to the observed effect used as the planning reference; they are not independent evidence that this effect will recur in another study.

The calculations should be read as illustrations rather than as a reconstruction of the trial's original design. The registry record does not report the original target effect, its assumptions, or its sample-size calculation. The figures also rely on the stated equal-allocation and normal-approximation assumptions, so they do not account for design features or analysis choices that are not described in the registry record.

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 = 2.300; pooled standard deviation σ = 7.529.
2. n = 2 × (1.96 + 0.842)² × 7.529² / 2.300² = 168.2, rounded up to 169 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 <- 2.3                      # observed difference in means
sigma <- 7.52876534979725                     # 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(233103)
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 = 169 (80% power), 226 (90% power)
 75% of observed effect: n per arm = 300 (80% power), 401 (90% power)
 50% of observed effect: n per arm = 673 (80% power), 901 (90% power)
power.t.test: 169.2 
Simulated power at n = 169 per arm: 0.797 (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

The registry record lists four outcome measures and posts two secondary outcomes, BAI and Life-3. It lists no reported analyses. It does not report a prespecified testing hierarchy, multiplicity adjustment, or a plan for choosing which outcomes to emphasize. Without those details, the secondary p-values should be treated as unadjusted and exploratory in interpretation. When several outcomes or comparisons are considered, the chance of at least one result appearing notable by chance can increase. Multiplicity procedures can manage that issue, but the registry record does not say whether any were used.

Details about missing data are also absent. The overall enrollment is 52, while the posted outcome counts are 22 for sertraline and 19 for placebo. The registry record does not explain whether those counts reflect participants with available measurements, an analysis subset, or another reporting convention. It does not describe withdrawals, reasons for missing observations, imputation, or sensitivity analyses. Without this information, it is not possible to assess from the posted record whether missingness could have influenced the outcome summaries.

The analysis population is not reported in the registry record. The record does not state whether participants were analyzed according to randomized assignment, treatment received, or another definition. Randomization supports comparisons when the analysis appropriately preserves the assigned groups, but the analysis population and handling of deviations matter to interpretation. No particular approach should be assumed from the design labels alone.

The original statistical analysis plan, model specifications, covariates, and approach to repeated measurements are not reported in the registry record. Nor does the record state how the primary outcome was intended to be analyzed. These omissions do not establish that the trial lacked a plan; they mean that the plan and its implementation cannot be assessed from the registry record. The posted statistics can be described, but a fuller evaluation of analysis decisions requires details that are not reported.

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 non-serious adverse event terms

The registry record lists 0 serious adverse events in the sertraline group and 0 in the placebo group. It lists 22 participants at risk in the sertraline group and 19 at risk in the placebo group for these figures. The record also lists 1 other adverse event in the sertraline group and 2 in the placebo group, with the same respective group counts at risk. The registry record does not provide serious adverse event terms.

These counts should be read as the adverse-event figures recorded for the groups, not as a complete description of safety. The registry record does not report the event definitions, severity criteria, timing, duration, or whether the other adverse events were considered related to treatment. It also does not report deaths. The counts alone do not support conclusions about comparative risk, especially without details on follow-up and event ascertainment. They describe what was posted, not the full clinical context needed to interpret adverse events.

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

The posted results describe a randomized, double-masked comparison of sertraline and placebo, but the primary outcome result is depression at baseline. That result reports baseline group means and an unadjusted contrast; it does not estimate a change during the intervention or a post-treatment difference on HAM-D. This distinction is central to what the registry record can support. The trial's treatment question is not answered by a baseline comparison alone.

The post-intervention BAI and Life-3 summaries are relevant as secondary observations. Their estimates have confidence intervals that include zero, and the record does not list reported analyses or describe any multiplicity approach. They should be interpreted as uncertain unadjusted comparisons, not as definitive evidence about treatment effect. The record also does not report the analysis population or missing-data handling, leaving additional uncertainty about how the summaries relate to all enrolled participants.

A confidence interval crossing zero does not prove that groups are identical, just as an estimate in one direction does not by itself demonstrate a meaningful effect. The intervals show uncertainty around the contrasts. Whether any difference would matter clinically depends on scale interpretation and a meaningful-change threshold, neither of which is reported for these comparisons. The baseline HAM-D score bands describe severity categories, but they do not supply a threshold for meaningful treatment-related change.

The registry record can establish what outcomes and group summaries were posted, and it can show the uncertainty associated with those unadjusted contrasts. It cannot, from the posted primary result, establish the effect of sertraline relative to placebo on post-treatment depression. It also cannot establish the study's original statistical plan, the reasons for the outcome counts, or a complete safety profile. Those are limitations of what is reported, not evidence that the unreported procedures or results did not exist.

Accordingly, the numerical results should be understood as descriptions of the posted baseline and secondary summaries, with their associated uncertainty. They do not justify a treatment recommendation or a conclusion about whether sertraline works. Resolving the treatment question would require a reported primary post-intervention outcome and enough information to evaluate its analysis and completeness.

17. Why This Trial Matters Statistically

This trial is a useful example of why outcome timing and outcome role matter as much as the numerical estimate. A record can describe a treatment comparison while its posted primary outcome is a baseline measurement. Recognizing that distinction prevents a baseline group contrast from being misread as evidence about change during treatment.

It also illustrates the value of reading estimates together with uncertainty. The baseline mean difference is accompanied by a confidence interval that spans zero, and the secondary contrasts have the same broad interpretive limitation. A point estimate alone can look more conclusive than the underlying data warrant. Confidence intervals make the range of compatible values explicit and help keep interpretation focused on precision as well as direction.

The record highlights the importance of complete reporting. Information about the original statistical plan, outcome populations, missing data, and multiplicity would help readers understand how the posted summaries were selected and analyzed. The recorded enrollment and outcome counts are not identical in scope, but the reason is not reported. Without that context, readers should avoid assuming how observations were handled.

Finally, the sample-size illustrations show how planning conclusions depend on assumed effects and design conditions. They are distinct from the trial's own original sample-size justification, which is not reported in the registry record. Keeping planning illustrations separate from observed results avoids treating a hypothetical design calculation as evidence about the trial's success or power. These statistical distinctions are broadly useful when interpreting registry results from clinical trials.

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.

A central trial-specific limitation in the registry record is that the posted primary outcome is at baseline. It provides no primary post-intervention HAM-D estimate or change-from-baseline comparison. As a result, the primary result cannot answer the treatment-effect question implied by the trial topic. The posted secondary outcomes occur after intervention, but they are not a substitute for a reported primary treatment endpoint.

The posted outcome counts are 22 in the sertraline group and 19 in the placebo group, compared with overall enrollment of 52. The registry record does not explain the difference, identify the analysis population, or describe missing observations. This makes it difficult to judge how representative the posted summaries are of all randomized participants.

The record posts two secondary outcomes and lists four outcome measures. It does not identify the remaining measures in the available outcome summaries, nor does it explain the analysis plan, multiplicity approach, or whether the posted comparisons were prespecified. These gaps constrain interpretation of the secondary results and prevent assessment of a complete outcome-reporting pattern.

Other details relevant to interpretation are not reported in the registry record, including randomization procedures, masking checks, adherence, treatment discontinuation, and analysis methods. The safety record gives counts but not event descriptions or assessment details, and deaths are not reported. These reporting limitations do not show that such information was never collected; they limit what can be concluded from the registry record itself.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Treatment of Post-Traumatic Brain Injury (TBI) Depression — NCT00233103.
• Registry record last updated 2015-11-13; status last verified 2015-10.
• Results first posted 2015-11-13.
• 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 record describes a completed phase 2, randomized, double-masked, parallel-group trial of sertraline versus placebo for post-traumatic brain injury depression. Enrollment was 52, and the recorded start and primary completion dates are 2003-06 and 2008-09. Sertraline was described as daily oral dosing beginning at 25mg and increasing to therapeutic levels up to 200mg; placebo was described for 10 weeks.

The posted primary outcome is Depression at Baseline, measured with HAM-D. The baseline group summaries and unadjusted comparison do not estimate a treatment effect after intervention. BAI and Life-3 are posted as secondary post-intervention outcomes, with unadjusted comparisons whose confidence intervals include zero. The registry record lists no reported analyses and does not report the original sample-size justification, analysis populations, missing-data methods, or multiplicity plan.

The record therefore supports description of the posted baseline and secondary summaries, but it does not provide a primary post-treatment depression comparison. Safety counts are posted, while event details and deaths are not reported in the registry record.