Spinal Cord Stimulation for the Treatment of Major Depressive Disorder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Spinal Cord Stimulation for the Treatment of Major Depressive Disorder (NCT03433339), a randomized phase 1, phase 2 study. The primary outcome is Montgomery Asberg Depression Rating Scale (MADRS) Score Change.

2. Clinical Question

This is a comparison of changes in a continuous symptom score, not a comparison of the proportion of participants who reached a particular response or remission threshold. The primary outcome description specifies a change from baseline, so the direction and size of the mean change matter, as does the uncertainty around the difference between groups. The registry record reports that 10 participants in the active group and 9 in the sham group contributed to the primary analysis. Its stated analysis population includes participants with at least one post-baseline MADRS assessment. The record does not report how many participants were randomized to each group, whether any participants discontinued, or how missing assessments were handled beyond using the last available observation where applicable.

3. Trial Design

This was a randomized, quadruple-masked, parallel-group pilot clinical trial registered as phase 1 and phase 2. The condition listed was major depressive disorder, and the sponsor was the University of Cincinnati. The study status is completed. The recorded start date is 2018-08-29, and the primary completion date is 2022-09-13. The registry lists an enrollment of 20.

The two groups were Active Treatment and Sham Treatment. Active treatment was described as thoracic anodal transcutaneous spinal direct current stimulation at 2.5 milliampere for 20 min, three times per week for 8 weeks. The sham group was described as receiving thoracic anodal transcutaneous spinal direct current sham stimulation in sessions lasting 20min, three times per week for 8 weeks. The registry identifies the groups and schedule but does not report further procedural detail about the sham condition or which roles were masked under the quadruple-masking designation.

The primary outcome was change in MADRS summed total scores from baseline to week 8, or the last available observation. The analysis population was participants with at least one post-baseline MADRS total-score assessment. The registry describes a least-squares mean endpoint parameter and standard error as the dispersion measure, and reports an ANOVA comparison. The record does not describe the ANOVA model in greater detail, including covariates or how repeated observations were treated.

Other listed outcomes include skin redness, Clinical Global Impression-Improvement, several symptom and questionnaire measures, and systolic blood pressure change. The registry lists 17 outcome measures and 16 posted secondary outcomes. Its record does not report a detailed statistical analysis plan, a sample-size justification, or interim analyses. Accordingly, the design features and methods are limited to what is recorded for the trial.

4. Participant Flow and Accounting

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

Before assignment: Pre-screened 671 participants. Forty two were assessed for eligibility on site. A total of 22 were excluded for not meeting eligibility criteria (n=19) or declined to participate (n=3). A total of 20 participants were randomized.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All participants with at least one post baseline Montgomery Asberg Depression Rating Scale total scores assessment were included in the analysis.

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: Although 10 participants were randomized on each arm, one participant on the Sham group withdrew after only completing the baseline visit and was excluded from the analysis. Baseline and longitudinal results analyze 10 participants on the active group and 9 on the sham group.

6. Primary Endpoint

The registry lists the primary outcome as Montgomery Asberg Depression Rating Scale (MADRS) Score Change, measured over 8 weeks (or last available observation)..

The primary endpoint is change in the Montgomery Asberg Depression Rating Scale (MADRS) summed total score. The scale ranges from 0 to 60, and higher scores indicate worse depressive symptom severity. The endpoint compares change from baseline to week 8, or the last available observation, between the active and sham groups. It is a continuous measure: each participant’s score can take different values across the scale, and the analysis summarizes group changes rather than classifying participants into response categories.

A change score describes how the measured outcome differs from baseline during the specified follow-up period. The registry reports the mean change for each group as a least-squares mean, with standard error as the dispersion measure. In principle, the central comparison is the difference between the two groups’ changes. The record does not give a numerical estimate or confidence interval for that between-group contrast, so its size and precision cannot be read directly from the posted result.

General explanation: a continuous symptom scale can retain information about variation in severity that would be lost if scores were converted into categories. That can make it useful for comparing average changes across randomized groups. Interpretation still depends on the measurement schedule, the analysis population, and the handling of missing observations. Here, the registry specifies week 8 or the last available observation and requires at least one post-baseline MADRS assessment for inclusion. It does not report additional details about the scale’s administration or missing-data procedures.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean change describe?

The primary outcome is continuous, meaning that MADRS total scores take values along a numerical scale. A change score represents the difference between a participant’s follow-up score and baseline score. Group mean changes summarize those individual changes. In this trial, the active group’s reported mean change was -21.7 and the sham group’s was -14.6. Because higher scores indicate worse symptom severity, negative values indicate that the average follow-up score was lower than the average baseline score within each group.

General explanation: a group mean is a measure of the center of a distribution. It does not show how much participants differed from one another or whether every participant changed in the same direction. The standard error reported with a mean describes the estimated precision of that mean, not the spread of individual measurements. A standard deviation would instead describe variation among observations; the registry identifies standard error as the dispersion measure for the least-squares means.

What does the ANOVA test address?

The registry reports an analysis of variance (ANOVA) for the superiority comparison, with p =0.040. General explanation: ANOVA is a family of methods for assessing differences among means under a specified statistical model. In a two-group setting, the test evaluates evidence concerning a difference in the modeled group outcomes. The exact interpretation depends on the model and its assumptions. The registry record does not report model details such as covariate adjustment, residual assumptions, or the precise contrast tested beyond describing the primary comparison as a difference in change between active and sham groups.

The p-value is not an effect-size measure. A p-value also does not tell the reader the probability that a finding arose “by chance” in a simple sense. It reflects the compatibility of the observed data with a null hypothesis and statistical model, assuming the model’s conditions hold. Since no between-group estimate or confidence interval is reported, the p-value cannot supply the missing magnitude or precision information.

Why does the contrast matter more than the group means alone?

The primary question concerns whether the groups differed in change, rather than whether either group changed from its own baseline. Within-group means answer the latter descriptive question. A randomized comparison is intended to assess the difference between groups, which is the relevant contrast for the trial question. The registry reports the ANOVA test of that comparison but does not post a numerical estimate of the contrast or an interval around it. It would be inappropriate to reconstruct and present a new contrast from the posted group means, particularly because the registry’s comparison may depend on model details that are not reported.

General explanation: a confidence interval gives a range of effect estimates compatible with the data and analysis at a stated confidence level. Its width conveys precision, while its location conveys the range and direction of estimates under the model. An interval for the difference between groups would be more directly informative about the primary contrast than separate standard errors for each group. The registry record does not report such an interval.

What does the analysis population imply?

The stated primary analysis population includes participants who had at least one post-baseline MADRS total-score assessment. The outcome timeframe allows week 8 or the last available observation. General explanation: restricting an analysis to participants with post-baseline measurements can make the analyzed group differ from the randomized group if some participants lack measurements. The direction and size of any resulting bias depend on why measurements are missing and on the analysis method. The record does not report the number or reasons for missing assessments, whether the last available observation was carried forward, or other details of missing-data handling.

The registry reports 10 participants in the active group and 9 in the sham group for the primary outcome. These are analysis counts in the posted results, not a substitute for a full account of participant flow. The record does not report how many participants were randomized to each group or give a detailed accounting of exclusions and missing outcome data.

Why can multiple outcomes affect interpretation?

The registry lists 17 outcome measures and 16 posted secondary outcomes, in addition to the primary outcome. General explanation: when many outcomes are tested, the chance of at least one apparently notable result can increase if each test is interpreted without accounting for the family of comparisons. Methods for addressing this include adjustments to testing thresholds or a prespecified hierarchy of outcomes. The registry record does not report whether multiplicity adjustment was planned or used. The primary p-value should therefore be interpreted alongside the full outcome context, without assuming that an adjustment was or was not applied.

The record identifies one reported analysis for the primary endpoint and provides the ANOVA p-value. It does not give a detailed statistical analysis plan, so the prespecification of model choices, handling of deviations, or analysis hierarchy is not reported in the registry record. General statistical explanations describe how methods usually work; they do not establish which unreported decisions were used in this trial.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Active Treatment", "Sham Treatment")
n   <- c(10, 9)
m   <- c(-21.7, -14.6)
sp  <- c(2.3, 2.5)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Active Treatment  10  -21.700  7.273  -26.903  -16.497
CARM  Sham Treatment  9  -14.600  7.500  -20.365  -8.835

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as an ANOVA superiority analysis with p =0.040. It also gives group mean changes and standard errors: -21.7 with standard error 2.3 for active treatment, and -14.6 with standard error 2.5 for sham treatment. No numerical between-group estimate or confidence interval is reported. The registry record does not provide enough detail about the ANOVA model to reproduce the sponsor’s comparison from the group summaries alone.

No recalculated two-group estimate is reported because the registry record does not provide the information needed for that comparison. The group means and standard errors are not a substitute for a reported model-based contrast, especially when model details are absent. There is no stated discrepancy between the sponsor’s result and a recalculated result; the registry record’s sponsor analysis is therefore the available inferential result to describe. It should not be expanded into claims about the magnitude or precision of the contrast that the record does not report.

General explanation: a sponsor-reported analysis is based on the study’s statistical model and analysis choices. A simple recalculation from summary statistics, when available, may differ because it is unadjusted and may not reproduce the original model. In this case, readers should distinguish the reported p-value from an effect estimate: the p-value is present, while the between-group estimate and its confidence interval are not reported in the registry record.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists 17 outcome measures and 16 posted secondary outcomes alongside the primary MADRS change endpoint. It reports one analysis for the primary endpoint. The existence of multiple outcomes is relevant to interpretation because testing many outcomes can create multiple opportunities to observe small p-values. General explanation: multiplicity procedures can control or describe this issue, for example through an adjustment or a prespecified testing sequence. Whether such a procedure was planned or used is not reported in the registry record.

The primary population is described as all participants with at least one post-baseline MADRS total-score assessment. The outcome timeframe is 8 weeks or the last available observation. These statements define the analysis population and measurement window at a high level, but they do not explain what happened when a participant lacked a week 8 assessment, how the last available observation was selected, or whether any imputation was used. The record does not report the number or reasons for missing assessments, participant-level completion, or a detailed participant flow by group.

General explanation: missing outcome data can affect estimates if the likelihood of missingness is related to a participant’s outcome or other characteristics. Different missing-data assumptions and methods can lead to different results. The impact cannot be assessed from group means and standard errors alone. The registry record does not describe sensitivity analyses for missing data or specify whether the primary analysis followed an intention-to-treat principle.

The trial lists an enrollment of 20, while the posted primary outcome groups contain 10 active and 9 sham participants. These counts describe different aspects of the record and should not be treated as a full explanation of participant disposition. Reasons for any difference in the enrolled and analyzed populations are not reported in the registry record. Details of allocation concealment, protocol deviations, and the prespecified statistical analysis plan are also not reported. Without those details, the degree to which outcome selection, missingness, and analysis choices were addressed cannot be determined from the registry entry.

13. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events among 10 participants at risk in the active group and zero among 9 participants at risk in the sham group. It also reports zero deaths in each group. No serious adverse event terms are listed. For other adverse events, the record reports 10 affected participants in the active group and 9 in the sham group.

These figures describe what was posted for the trial; they do not establish that the intervention has no risk. General explanation: the absence of a recorded serious event in a study does not rule out uncommon events or events that occur under different conditions. The meaning of an adverse-event count also depends on the registry’s definitions, collection procedures, and observation period. The record does not report detailed event descriptions, severity or relatedness assessments for the other adverse events, or further information about how events were collected.

The listed secondary outcomes include number of participants with skin redness, but the registry information does not give a result for that outcome. It should not be inferred from the adverse-event totals. The serious-event figures are best read as a limited safety description within this completed pilot trial, not as a comprehensive estimate of risk.

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

The posted primary results describe mean MADRS score changes of -21.7 in the active group and -14.6 in the sham group, with standard errors of 2.3 and 2.5, respectively. The registry reports an ANOVA p-value of p =0.040 for the superiority comparison. Together, these entries show the group summaries and the sponsor-reported statistical test. Because higher MADRS scores indicate worse depressive symptom severity, negative mean changes indicate lower average scores relative to baseline in both groups.

The record does not report a numerical between-group estimate or confidence interval. Consequently, readers cannot directly assess from the registry entry how large the modeled difference in change was or how precisely it was estimated. The reported p-value does not answer those questions, and it does not establish clinical importance. The registry also does not report a threshold for interpreting a change as clinically important.

This trial concerns a specific active stimulation procedure compared with a sham procedure, in the analysis population with at least one post-baseline MADRS assessment, over a timeframe of 8 weeks or the last available observation. Its findings should not be generalized beyond those conditions without evidence from other settings. The record’s pilot designation is part of the trial description; a pilot result does not by itself establish how the intervention would perform in a larger or different study.

The posted data do not show whether results would be similar under other missing-data assumptions or after accounting for the other outcomes. The registry does not report a detailed statistical plan, multiplicity procedure, participant flow, or sensitivity analysis. Those omissions limit how fully the primary result can be evaluated. No medical recommendation follows from these statistical descriptions. They characterize the registry’s reported outcome summaries and test, not a decision about treatment for any individual.

15. Why This Trial Matters Statistically

This trial illustrates why a reported p-value should be read together with an effect estimate and its uncertainty. The registry gives a primary ANOVA result, but does not report a numerical between-group estimate or confidence interval. The group means and their standard errors provide useful descriptive information, yet they do not directly replace the modeled contrast that addresses the randomized comparison. Without that contrast and interval, the magnitude and precision of the primary result remain difficult to assess from the registry record.

The analysis population also matters. The primary endpoint includes participants with at least one post-baseline MADRS assessment, while enrollment is reported separately. The record does not provide detailed participant flow or the reasons for missing outcomes. These are important distinctions for interpreting a continuous change endpoint, because the observed analysis group may not fully reflect the randomized population when follow-up data are missing.

The trial also lists numerous secondary outcomes. General statistical reasoning highlights that a larger set of tested outcomes can complicate the interpretation of individual p-values, particularly when no multiplicity approach is described. The registry does not report such an approach, so it is not possible to determine whether or how multiplicity was handled.

These features make the record useful for learning how to distinguish descriptive group summaries, the sponsor’s inferential test, analysis-population definitions, and uncertainty. A statistically notable p-value is not a complete account of a treatment comparison. Clear reporting of effect estimates, confidence intervals, analysis details, missing-data handling, and outcome-testing strategy helps readers judge what the results support and what remains uncertain.

16. 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 record leaves several trial-specific details unreported. It does not give a numerical between-group estimate or confidence interval for the primary MADRS comparison, despite reporting group mean changes, standard errors, and an ANOVA p-value. It also does not describe the ANOVA model beyond identifying the method, so covariate adjustment and other model choices cannot be assessed.

The primary analysis population is defined as participants with at least one post-baseline MADRS assessment, and the endpoint may use week 8 or the last available observation. The registry does not report how many participants had missing week 8 measurements, how last available observations were handled, or whether imputation or sensitivity analyses were used. It does not provide a detailed account of participant flow or explain the relationship between the enrollment figure and the primary analysis counts.

The registry record does not report a sample-size justification, an interim-analysis plan, or a detailed statistical analysis plan. It also does not state whether the 16 posted secondary outcomes were adjusted for multiplicity or how they were prioritized. The reported results therefore cannot be assessed against an unreported testing hierarchy or plan.

Additional details about the masking procedures, sham delivery, participant characteristics, adherence, protocol deviations, and the clinical interpretation threshold for MADRS change are not reported in the registry record. These specific gaps constrain assessment of the comparability and conduct of the groups, the precision of the estimated contrast, and the meaning of score changes. They do not imply that a particular problem occurred; they identify what cannot be determined from the recorded information.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Spinal Cord Stimulation for the Treatment of Major Depressive Disorder — NCT03433339.
• Registry record last updated 2024-12-27; status last verified 2024-12.
• Results first posted 2024-04-23.
• 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.

20. Record Summary

The completed phase 1 and phase 2 randomized, quadruple-masked, parallel-group trial compared active with sham transcutaneous spinal direct current stimulation in major depressive disorder. The registry lists an enrollment of 20 and identifies the University of Cincinnati as sponsor. Active treatment was described as thoracic anodal stimulation at 2.5 milliampere for 20 min, three times per week for 8 weeks; sham sessions were described on the same frequency and duration.

The primary endpoint was change in MADRS summed total scores at 8 weeks or the last available observation. The active group mean change was -21.7 (standard error 2.3), and the sham group mean change was -14.6 (standard error 2.5). The sponsor-reported ANOVA result was p =0.040. A numerical between-group estimate and confidence interval are not reported in the registry record. The primary analysis included participants with at least one post-baseline MADRS assessment. The registry lists 16 posted secondary outcomes and reports no serious adverse events or deaths in either group.