Safety and Effectiveness of S-adenosyl-l-methionine (SAMe) for the Treatment of Major Depression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Effectiveness of S-adenosyl-l-methionine (SAMe) for the Treatment of Major Depression (NCT00101452), a randomized na study. The primary outcome is Hamilton Rating Scale for Depression (HAM-D).

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group study with triple masking. Its official title describes the study as double-blind; the registry’s masking classification is triple. The listed phase is NA. The trial is marked completed, with enrollment of 199, a start date of 2005-04, and primary completion in 2010-06.

The interventions listed are SAMe, escitalopram, and placebo. In the primary outcome results, the SAMe arm included 59 participants, the escitalopram arm included 55, and the placebo arm included 52. These are the reported group sizes for the outcome results, not a complete account of participant flow. How many participants were randomized, treated, or excluded from the primary analysis, and why those counts differ from the enrollment figure, is not reported in the registry record.

The primary outcome was change in total HAM-D score between baseline and endpoint, with the endpoint identified as 24 weeks. The outcome population was based on having at least one post-baseline visit. The adverse-event information separately lists Weeks 1-12 and Weeks 12-24 continuation groups, as well as a SAMe plus escitalopram crossover group during Weeks 12-24. These safety group descriptions do not establish the details of the primary-outcome treatment schedule. Doses, visit schedules, randomization ratios, allocation concealment, and the sequence or criteria for continuation and crossover are not reported in the registry record.

The registry lists one primary outcome measure and one reported analysis. Its stated enrollment is a study-level figure; the endpoint results provide the arm-specific numbers used in those results. The original sample-size rationale and the trial’s own planned assumptions 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: Only five participants completed the allowed wash-out period prior to randomization. Participants screened at Day 0-7 and were asked to return for a baseline visit at Day 0. Subjects not randomized were excluded for being ineligible at the baseline or for being lost to follow-up between screen and baseline.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Based on having at least one post-baseline visit.

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 Hamilton Rating Scale for Depression (HAM-D), measured over baseline and 24 weeks.

The primary endpoint was change in total score on the Hamilton Rating Scale for Depression (HAM-D) between baseline and endpoint, with endpoint specified as 24 weeks. The HAM-D is described as a clinician-rated inventory of depressive symptoms. Its items are scored from zero to four, and the summed total can range from 0 to 68. Higher scores indicate poorer outcomes.

Change from baseline expresses how the score moved over the study period. In the posted results, the arm means are negative, indicating reductions in average score from baseline. A more negative mean change represents a larger average reduction in the measured score; it does not by itself establish that a between-group difference is precise or statistically persuasive. The clinically meaningful size of a difference is not reported in the registry record.

A clinician-rated scale provides a defined way to quantify symptoms for an outcome comparison. The reported measure is continuous, so the results are presented as means and standard deviations rather than counts of participants meeting a threshold. The registry does not report a response or remission definition, and those outcomes should not be inferred from mean score changes.

7. Results Posted in the Registry

• 1. SAMe: mean -6.70 versus -5.70 on control; difference -1.00 (95% CI -3.66 to 1.66), standardized effect d = -0.14; z-test p = 0.462.
• 2. Escitalopram: mean -7.50 versus -5.70 on control; difference -1.80 (95% CI -4.53 to 0.93), standardized effect d = -0.25; z-test p = 0.196.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is continuous: each participant has a numerical change in HAM-D score, and the reported group summaries are means with standard deviations. A mean describes the group’s average change. A standard deviation describes the variation in individual changes around that average. Variation matters because the precision of a mean difference depends not only on the estimated difference but also on the spread of individual outcomes and the number of observations.

What does the reported ANOVA test?

In general, an analysis of variance (ANOVA) tests whether the group means can be treated as equal under its null hypothesis. The registry reports an ANOVA across SAMe, escitalopram, and placebo with p < 0.05. This is an overall test: it does not, on its own, identify which particular groups differ, give the size of a difference, or provide a confidence interval for each contrast. The registry does not report the ANOVA statistic, degrees of freedom, exact p-value, or follow-up testing details.

The recalculated comparisons are different in scope. They estimate the mean change in each active-treatment arm minus the mean change in placebo. For SAMe, the estimate is -1.0; for escitalopram, it is -1.8. Negative values indicate a larger average reduction in the active-treatment group than in placebo, because lower follow-up scores represent better measured outcomes and the comparison is based on change. These are unadjusted estimates from the posted summary statistics, not estimates reported as the sponsor’s pairwise analysis.

How should a confidence interval be read?

In general, a confidence interval gives a range of values compatible with the data and the statistical procedure used, under its assumptions. It communicates uncertainty about an estimate rather than a range containing a known fixed percentage of individual outcomes. The SAMe interval is -3.663 to 1.663, and the escitalopram interval is -4.528 to 0.928. Both include zero, the value corresponding to no mean difference. Each also includes estimates in the opposite direction from the negative point estimate. This indicates limited precision for these contrast-specific estimates.

Why can the overall and pairwise results differ?

In general, an overall test can detect evidence that not all group means are the same without specifying which pair is responsible. Pairwise contrasts test particular comparisons and have their own estimates and uncertainty. A reported overall p-value and unadjusted recalculated contrast p-values can therefore differ without being direct contradictions. Interpretation also depends on which comparisons were prespecified and how multiplicity was handled. The registry does not report the planned contrast structure or adjustment procedures.

Why does outcome variation matter?

In general, greater variation among individual outcomes makes a group mean less precise when other factors are held fixed. The reported standard deviations are 7.3 for SAMe, 7.4 for escitalopram, and 7.0 for placebo. Those values describe variability within each arm, not uncertainty in the mean difference itself. The confidence intervals incorporate uncertainty in the comparison. If outcomes are over-dispersed relative to a model’s assumptions, standard errors can be too small and confidence intervals too narrow; this is a general statistical concern, not a reported problem in this trial. The registry provides no assessment of over-dispersion or model diagnostics.

The primary outcome is a change score. In general, change scores compare the difference between an individual’s baseline and endpoint values and can be summarized across participants. The registry identifies the measure and timeframe but does not give baseline arm means, endpoint arm means, or a detailed model specification. It therefore does not establish whether the sponsor analyzed change scores directly, adjusted for baseline in a model, or used another approach for the reported ANOVA.

The standard deviations and group sizes also permit a simple unadjusted recalculation of the active-treatment contrasts. Such a calculation uses the posted summary statistics and does not reproduce an analysis that may account for covariates, repeated measures, imputation, or other design features. The record does not report whether such features were included in the sponsor’s analysis. The reported ANOVA remains the analysis attributed to the registry; the recalculated estimates offer a separate, transparent contrast-specific view of the posted means.

10. Confidence Intervals and Precision

Precision is best judged from the confidence intervals, not from the point estimates alone. The recalculated SAMe-minus-placebo mean difference is -1.0, with a 95% confidence interval from -3.663 to 1.663. The escitalopram-minus-placebo mean difference is -1.8, with a 95% confidence interval from -4.528 to 0.928. Both intervals include zero, so the posted summary statistics do not rule out no mean difference for either contrast.

The intervals also extend to positive differences, while their point estimates are negative. The data are consequently compatible with a range of possible contrast sizes and directions. The confidence intervals do not mean that every value in the range is equally plausible, nor do they describe the range of individual participants’ changes. They express uncertainty about the estimated difference under the recalculation’s assumptions.

The standard deviations reported for the arms are similar in magnitude: 7.3 for SAMe, 7.4 for escitalopram, and 7.0 for placebo. These values indicate substantial individual variation relative to the estimated mean differences, while the confidence intervals directly convey uncertainty in those group contrasts. A clinical threshold for interpreting the size of a HAM-D difference is not reported in the registry record, so statistical uncertainty cannot be translated into a clinical-importance judgment from this record alone.

The registry’s overall ANOVA result, p < 0.05, is not accompanied by a confidence interval or effect estimate. It therefore does not provide the same information about the magnitude or precision of a specific active-treatment-versus-placebo contrast. The record also does not report the exact ANOVA p-value or the method used to obtain it.

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 planning figures are participants per arm, not total events. They illustrate sample sizes under the stated assumptions, using the observed contrast as a starting point and considering planning effects of 100%, 75%, and 50% of that observed effect. For the SAMe contrast, planning at 100% of the observed effect gives 806 participants per arm for 80% power and 1,078 participants per arm for 90% power. At 75%, the figures are 1,432 and 1,916 participants per arm; at 50%, they are 3,221 and 4,311 participants per arm.

For the escitalopram contrast, planning at 100% of the observed effect gives 252 participants per arm for 80% power and 338 participants per arm for 90% power. At 75%, the figures are 448 and 600 participants per arm; at 50%, they are 1,008 and 1,349 participants per arm. The 80% column describes a planning scenario with an 80% chance of detecting the specified effect under the assumptions; the 90% column uses a higher target chance. These are planning illustrations, not observed power.

The stated calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is the stated fraction of the observed effect. In general, smaller effects are harder to distinguish from random variation and therefore require more participants for the same planned power. The figures are not the trial’s original design justification: the original sample-size assumptions and rationale are not reported in the registry record.

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 = -1.000; pooled standard deviation σ = 7.161.
2. n = 2 × (1.96 + 0.842)² × 7.161² / -1.000² = 805.0, rounded up to 806 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 <- -1                      # observed difference in means
sigma <- 7.16119768569322                     # 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(101452)
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 = 806 (80% power), 1,078 (90% power)
 75% of observed effect: n per arm = 1,432 (80% power), 1,916 (90% power)
 50% of observed effect: n per arm = 3,221 (80% power), 4,311 (90% power)
power.t.test: 806 
Simulated power at n = 806 per arm: 0.799 (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 attributes a three-group ANOVA to the reported analysis and gives p < 0.05. It does not provide an ANOVA effect estimate, confidence interval, exact p-value, or enough detail to determine which group differences contribute to that overall result. The registry also does not state whether the ANOVA was the only sponsor analysis or how it was specified.

The recalculated estimates use the posted arm-level means, standard deviations, and group sizes to compare each active-treatment arm separately with placebo. They are unadjusted mean differences: -1.0 for SAMe and -1.8 for escitalopram. Their unadjusted p-values are 0.462 and 0.196, respectively, and both corresponding confidence intervals include zero. These calculations are not the sponsor’s reported ANOVA and should not be described as a reproduction of it.

An overall test and individual pairwise contrasts address different questions. An overall ANOVA can indicate that the group means are not all alike without identifying a particular contrast. The pairwise recalculations focus on two active-treatment-versus-placebo comparisons. Since the registry does not report follow-up testing or multiplicity procedures, the relationship between the overall result and any specific pairwise conclusion cannot be resolved from the record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and one reported analysis. The reported analysis includes all three groups in an ANOVA, while the recalculations present two active-treatment-versus-placebo contrasts. The record does not say whether those contrasts, or any additional analyses, were prespecified. It also does not report a multiplicity adjustment or explain how the overall ANOVA result was followed up.

Multiplicity is a general statistical issue that arises when several hypotheses or comparisons are considered. As the number of tests increases, the chance of obtaining at least one apparently unusual result by chance can increase if no adjustment or other control is used. This general point does not establish that the analysis in this trial was improperly conducted; the registry does not give enough detail to assess the planned testing strategy. In particular, a single reported outcome does not mean that only one statistical comparison was considered.

The primary-outcome population is described as participants with at least one post-baseline visit. The registry does not report how many participants lacked a post-baseline score, how missing endpoint measurements were handled, or whether any imputation or sensitivity analyses were used. It also does not identify the outcome analysis as intention-to-treat, per-protocol, or another named population. These details affect how completely the reported results represent the randomized groups and how robust the estimates are to incomplete follow-up.

The record does not report the statistical model specification, covariates, assumptions checks, or whether baseline values were incorporated beyond defining the outcome as change from baseline. It does not report an interim-analysis plan, stopping rules, or a detailed statistical analysis plan. These items should be treated as not reported in the registry record, rather than inferred from the use of ANOVA.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

For Weeks 1-12, the registry lists 59 participants at risk for serious adverse events in the SAMe group, 55 in the escitalopram group, and 52 in the placebo group. Serious adverse events were recorded for 0 participants in each of those groups, and deaths were 0 in each group.

For Weeks 12-24 continuation, the listed at-risk counts for serious adverse events were 17 for SAMe, 11 for escitalopram, and 15 for placebo. Serious adverse events were recorded for 0 participants in these groups, and deaths were 0. The record also lists a SAMe plus escitalopram crossover group during Weeks 12-24, with 43 at risk, 0 serious adverse events, and 0 deaths.

These are the registry’s reported serious-event counts for the named periods and groups. The record does not list serious adverse-event terms. Other, non-serious adverse-event counts are a separate measure and should not be confused with serious-event counts. The different group sizes and periods, including continuation and crossover descriptions, mean these figures should not be treated as a single like-for-like comparison of safety risks. Further details about event definitions, ascertainment, and how events were attributed are not reported in the registry record.

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

The record reports average reductions in HAM-D score in all three primary outcome groups: -6.7 for SAMe, -7.5 for escitalopram, and -5.7 for placebo. Those are within-group changes. They do not by themselves isolate a treatment effect, because a treatment comparison depends on how the changes differ between groups. The recalculated contrasts with placebo are -1.0 for SAMe and -1.8 for escitalopram.

Both contrast-specific confidence intervals include zero, and their unadjusted p-values are 0.462 and 0.196. Thus the recalculated estimates are uncertain and do not establish a clear difference from placebo on this outcome. At the same time, the confidence intervals allow differences in either direction, so they should not be read as evidence that the groups were equivalent. The registry does not report an equivalence margin or a threshold for clinical importance.

The registry’s three-group ANOVA is reported with p < 0.05. That finding applies to the overall group analysis as reported; the record does not provide the exact test details or identify which specific comparisons account for it. It is not possible to reconcile that global finding with particular treatment contrasts from the registry information alone.

The endpoint is a clinician-rated symptom score, measured as change from baseline to 24 weeks. The record does not report longer-term outcomes, response or remission results, or the clinical importance of a score difference. Nor does it provide enough information about missing outcomes and analysis populations to determine how fully the results represent all randomized participants. These limitations constrain interpretation of the posted estimates, without changing what the registry reports.

Accordingly, the results describe the recorded score changes, an overall ANOVA result, and imprecise recalculated contrasts. They do not by themselves establish a clinical conclusion about SAMe or escitalopram. Any such conclusion would require broader evidence and a clinical interpretation that is not contained in this registry record.

17. Why This Trial Matters Statistically

This trial illustrates why an overall statistical test and treatment-specific effect estimates need to be read together. The registry reports an overall three-group ANOVA result of p < 0.05, but gives no group-specific estimate or interval from that analysis. The recalculated active-treatment contrasts give a different kind of information: each has an estimated direction and a confidence interval that shows substantial uncertainty.

The arm means also show why within-group change is not the same as a comparative effect. Each group’s average HAM-D score fell from baseline, but the relevant comparison for treatment interpretation is the difference in changes between groups. Without that distinction, the reductions within the active-treatment arms could be misread as isolating an effect relative to placebo.

The registry highlights another statistical distinction: a nonsignificant unadjusted pairwise result is not proof that the treatments and placebo are equivalent. Both contrast intervals include zero and extend to effects in either direction, and no equivalence margin is reported. Precision and the range of compatible effects are essential context for interpreting a p-value.

Finally, the record’s outcome population is based on having at least one post-baseline visit, but the handling of missing data and analysis population details are not reported. That gap prevents a full assessment of how the analyzed participants relate to the randomized groups. Together, these features make the trial a useful example of separating an overall test, specific contrasts, uncertainty, and the limits of registry reporting.

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 registry’s endpoint results include group sizes of 59 for SAMe, 55 for escitalopram, and 52 for placebo, while the overall enrollment is 199. Participant flow explaining the relationship between these figures is not reported in the registry record. The primary outcome population is described only as participants with at least one post-baseline visit, and the number or characteristics of participants without such a visit are not reported.

The record does not describe dosing, adherence, eligibility criteria in detail, allocation concealment, or the visit schedule. It also does not provide the full statistical analysis plan, the ANOVA statistic, its exact p-value, model assumptions, or the sponsor’s pairwise estimates. The registry’s double-blind wording in the official title differs from its triple-masking classification; further details about masking procedures are not reported.

The record gives a primary endpoint at 24 weeks and adverse-event information for Weeks 1-12 and Weeks 12-24, including continuation and crossover groups. It does not provide enough detail to connect all safety group definitions to the primary-outcome comparison or to assess event ascertainment and attribution. Serious-event terms are not listed.

The recalculated pairwise estimates rely on posted summary statistics and are unadjusted. They do not account for any analysis features not represented by those summaries. The registry does not report a clinically important difference threshold, an equivalence margin, or longer-term outcome results. These omissions limit conclusions about clinical importance, missing-data sensitivity, and the meaning of the overall reported result.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Safety and Effectiveness of S-adenosyl-l-methionine (SAMe) for the Treatment of Major Depression — NCT00101452.
• Registry record last updated 2017-04-04; status last verified 2017-03.
• Results first posted 2012-12-31.
• 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

NCT00101452 is a completed randomized parallel-group study of SAMe, escitalopram, and placebo for major depression. The registry classifies masking as triple and lists phase as NA. Enrollment was 199; the study start was 2005-04 and primary completion was 2010-06.

The primary outcome was change in HAM-D score from baseline to 24 weeks, among participants with at least one post-baseline visit. The registry reports an overall three-group ANOVA result of p < 0.05. Unadjusted recalculated mean differences versus placebo were -1.0 for SAMe and -1.8 for escitalopram; both confidence intervals include zero. Serious adverse events and deaths were recorded as 0 in each listed safety group. Further details on the analysis plan, missing data, and participant flow are not reported in the registry record.