[Study Evaluating the Efficacy and Safety of a Digital Therapeutic as an Adjunct to TAU in Postpartum Depression]: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for [Study Evaluating the Efficacy and Safety of a Digital Therapeutic as an Adjunct to TAU in Postpartum Depression] (NCT05551195), a randomized na study. The primary outcome is Hamilton Rating Scale for Depression (HAM-D).

2. Clinical Question

3. Trial Design

NCT05551195 is a multicenter, randomized, parallel-group controlled study sponsored by Woebot Health. The registry lists masking as triple. Its official title describes the study as double-blind, so the record contains both descriptions; it does not explain the difference or identify which roles were masked. The phase is listed as NA.

The study began on 2022-11-16, and the primary completion date is 2023-05-10. Its current status is terminated. The registry record does not report why the study was terminated, whether recruitment stopped early, or whether the termination affected the planned analyses. Those details should not be inferred from the status label alone.

The intended groups were WB001 with adjunctive treatment as usual and educational control, ED001, with adjunctive treatment as usual. In both groups, participants were asked to download and use a study application. The record identifies both digital therapeutics as devices. It does not give enough detail to characterize their therapeutic content, the participants' usual care, or the duration and intensity of application use beyond the primary outcome timeframe.

The registry labels OG000 as the control group, while the posted arm descriptions identify OG000 as WB001 and describe ED001 as the educational control. This designation is not reconciled in the record. The primary outcome summaries show 1 participant in the WB001 group and 5 in the educational-control group. These outcome counts do not match the enrollment total of 7, and the primary outcome record notes one participant missing at Week 8. The record does not explain the remaining accounting or describe an analysis population.

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

Analysis population for the primary outcome, as defined in the registry: One participant missing data at Week 8

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 Change from Baseline to Post-treatment at Week 8.

The primary endpoint is change in the Hamilton Rating Scale for Depression, or HAM-D, from baseline to post-treatment at Week 8. The registry describes HAM-D as a measure of depression based on a 6-item questionnaire. Total scores range from 0 to 22, and higher scores indicate greater depression severity. The endpoint is reported as a mean change with standard deviation.

In general, a repeated measure of symptom severity can describe how scores differ between the start and end of a treatment period. A comparison between randomized groups can help distinguish the observed course in one group from the course in another, subject to the study's design and the uncertainty in the estimates. The posted arm summaries report a mean change of 0.0 with standard deviation 0.0 for WB001, based on 1 participant, and a mean change of -5.4 with standard deviation 3.36 for ED001, based on 5 participants.

The registry does not specify the calculation convention for change, such as the order in which follow-up and baseline scores were subtracted. It also does not report a between-group estimate or its uncertainty. The arm-level values therefore describe the available summaries, but do not establish a comparative effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical information available for this study is limited. The primary endpoint is continuous: each participant has a HAM-D score, and the registry reports mean change and standard deviation by group. No inferential analysis is posted for the primary endpoint. The record indicates that a two-arm comparison is not estimable from the available values, so there is no posted treatment contrast to interpret.

What does an arm-level mean describe?

In general, a mean is the arithmetic average of the observed values in a group. It offers a compact description of the center of those observations. Here, the registry reports a mean change of 0.0 for WB001 and -5.4 for educational control. Those means describe the participants represented in each summary; they are not, by themselves, estimates of a randomized between-group effect. The group sizes are unequal, and the WB001 summary contains only 1 participant.

A standard deviation describes the spread of observed values around a mean. The ED001 standard deviation is 3.36. The WB001 standard deviation is 0.0, but a group containing 1 participant cannot reveal between-participant variation. The zero value in that summary should therefore not be interpreted as showing that the intervention produces identical outcomes for different people. More generally, a measure of spread is difficult to assess when very few observations contribute to it.

Why is a direct comparison needed?

Random assignment is intended to support a comparison between groups by assigning participants to the study arms through a random process. The statistical question is ordinarily expressed as a contrast between the groups, rather than as separate descriptions of their outcomes. A contrast is accompanied by an uncertainty estimate, often a confidence interval, and may also be assessed with a statistical test. In general, a confidence interval gives a range of values compatible with the data and the assumptions of the analysis; its width communicates precision. The registry does not report a contrast, confidence interval, or p-value for this primary endpoint.

Separate arm means cannot substitute for that missing analysis. In particular, no difference, ratio, or other contrast should be calculated from these summaries as though it were the sponsor's analysis. The record states that a two-arm comparison is not estimable for this continuous endpoint. There is no discrepancy between a posted sponsor comparison and a recalculated comparison to resolve, because no comparative analysis is posted.

What do the planning assumptions mean?

The registry includes assumptions for planning illustrations: a two-sided significance level of 0.05, equal allocation, and a normal approximation. It describes the target effect for those illustrations in relation to a portion of the observed effect. These are planning assumptions, not evidence that the trial was designed using those values. The protocol's sample-size justification and original design assumptions are not reported in the registry record. Any planning illustration based on the observed effect is not observed power and should not be interpreted as proof that the trial had adequate power.

In general, a significance level is a decision threshold used in a specified testing procedure, and two-sided testing allows evidence in either direction. Equal allocation means planning for the same number of participants in each group. A normal approximation uses a mathematical distribution to approximate uncertainty or test behavior. These concepts describe the listed planning framework; they do not fill in an absent primary analysis or establish the suitability of the framework for this study.

Why can small samples limit precision?

In general, estimates from small samples tend to be less stable because each observation has substantial influence on the group summary. A single participant's score can materially affect a mean when only a few observations are available. The displayed arm sizes, 1 and 5, make that concern especially relevant. The registry does not report a confidence interval that would quantify uncertainty for either group or for a comparison between groups.

For continuous outcomes, an analysis plan may specify how baseline values are used, which participants are included, how missing observations are handled, and whether adjustments are made. Those choices can affect the estimate and its uncertainty. The registry does not report those details for the primary outcome. The fact that change from baseline is the endpoint does not establish whether additional adjustment for baseline was planned or conducted.

Because the primary comparison is not estimable in the record, neither the randomized design nor the descriptive endpoint summaries resolve the comparative question. The most defensible statistical reading is to distinguish observed summaries from inferential results: the former are posted, while the latter are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 secondary outcomes: EPDS, PHQ-9, GAD-7, MIBS, CGI-S, CGI-I, PGI, and CSQ-8. Together with HAM-D, the record reports 9 outcome measures. No analyses are posted. The registry record does not describe a multiplicity strategy, such as how the chance of false-positive findings across several outcomes would be addressed. Without posted analyses and a reported strategy, the secondary outcomes cannot be used to infer additional comparative findings.

One participant is identified as missing Week 8 primary outcome data. The record does not give the reason for missingness, identify the participant's group, state whether the missing value was imputed, or describe any other missing-data method. It also does not report whether all randomized participants were included in the primary analysis or define another analysis population. The posted primary outcome groups include 1 participant in WB001 and 5 in ED001, while enrollment is 7; the record does not explain the full accounting.

In general, missing-data methods rely on assumptions about why observations are absent and how missing outcomes relate to observed information. Different assumptions can lead to different estimates. A complete-case summary describes participants with available outcome values, but does not automatically answer what the results would have been for all randomized participants. The registry does not state whether the posted summaries are complete-case results or explain the assumptions used.

Interim analyses, stopping rules, and the statistical plan are not reported in the registry record. The terminated status does not by itself establish whether an interim analysis took place or why the study ended. The record also does not explain whether termination altered the analysis populations or outcome reporting.

11. 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 posted adverse-event figures list zero serious adverse events and zero other adverse events in both groups. The WB001 group had 1 participant at risk for these summaries, and the educational-control group had 6. Deaths are listed as 0 in each group. No serious adverse-event terms are listed.

These counts describe what the registry reports for the participants included in the adverse-event summaries. The exposed groups are very small, so zero reported events should not be interpreted as evidence that an event cannot occur or that the treatments have established safety. A small number of participants provides limited information about uncommon events and cannot characterize the full range of possible harms.

The registry does not provide additional details about event definitions, severity grading, collection periods, or whether adverse-event reporting was complete after termination. Nor does it provide a comparative safety analysis. The figures therefore support only a narrow description: no serious events, other events, or deaths are listed among the participants counted in these groups. They do not establish equivalent safety between the groups or settle the safety profile of either digital therapeutic.

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

The posted primary outcome values are descriptive summaries from a very small set of participants. They show a mean change of 0.0 in the WB001 group, based on 1 participant, and -5.4 in the educational-control group, based on 5 participants; the ED001 standard deviation is 3.36. One participant is noted as missing Week 8 data. These observations do not supply a comparative estimate for the randomized groups.

The registry states that values are missing or not estimable for a two-arm comparison of the continuous endpoint. No p-value, confidence interval, or primary comparative analysis is posted. Accordingly, the results do not show whether the groups differed, how large any difference might be, or how precisely it could be estimated. The arm-level summaries should not be treated as proof that one group performed better than the other.

The endpoint concerns change in HAM-D, a depression measure with higher scores indicating greater severity. The record does not specify the direction used to calculate change, so the sign of a change value should be interpreted with that limitation in mind. It also does not report baseline and follow-up scores, individual participant data, or an explanation for all differences between enrollment and the outcome summaries.

The study compared two digital applications, each alongside treatment as usual. It did not, in the described arms, compare an application with no application or with treatment as usual alone. The registry does not report the content of treatment as usual, the applications' therapeutic components, or the reason for termination. Those omissions limit what can be concluded about the intervention contrast and the study's conduct.

The safest interpretation is bounded: the record contains arm-level outcome and adverse-event summaries, but not an inferential result addressing the primary comparison. It neither establishes a treatment benefit nor demonstrates an absence of benefit. It also does not establish safety or lack of safety. The small outcome counts, missing Week 8 data, and unreported analysis details mean the posted record offers limited precision and limited basis for broader conclusions.

13. Why This Trial Matters Statistically

This trial illustrates why a randomized design and a reported comparative analysis are both important. Randomization defines the intended comparison, but the posted means alone do not quantify that comparison. Without an estimated between-group effect and its uncertainty, readers cannot assess how compatible the observed data are with a range of possible group differences. In this registry record, the primary two-arm comparison is not estimable from the posted values.

The scale of the outcome data also matters. The primary summaries include 1 participant in WB001 and 5 in educational control, against a total enrollment of 7. The registry notes one participant missing Week 8 data but does not explain all participant accounting. With so few observations, individual outcomes can strongly influence summaries, and the data cannot provide a precise picture of variation across a larger population.

The record also highlights the distinction between outcomes measured and outcomes analyzed. It lists 8 secondary measures in addition to HAM-D, but no analyses are posted. Knowing which measures were collected does not reveal their results, how they were analyzed, or how multiple outcomes were handled. Those questions require a reported analysis plan and results.

Finally, status and safety counts need context. Termination is recorded, but the reason is not. Zero listed adverse events among small groups is a factual description, not a comprehensive safety conclusion. For statistical interpretation, the important point is to keep claims proportionate to the information available: the registry offers limited descriptive data, while key inferential and conduct details are not reported.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several trial-specific limitations restrict interpretation. The overall enrollment is 7, while the primary outcome summaries include 1 participant in WB001 and 5 in educational control. The registry notes one participant missing Week 8 data but does not account for all participants or state why the observation is missing. The group sizes are therefore small, and the posted primary outcome is not suitable for a precise comparative reading.

No between-group estimate, confidence interval, p-value, or analysis is posted for the primary endpoint. The record says a two-arm comparison is not estimable from the available values. It does not report the analysis population, missing-data method, baseline score summaries, calculation convention for change, or any adjustment strategy. These omissions prevent assessment of how the primary result was constructed and how uncertain a comparison might be.

The record lists multiple secondary outcomes but no analyses and no multiplicity approach. It does not provide the protocol's sample-size justification or clarify whether the planning assumptions listed for illustrations correspond to the original study design. Interim analyses, stopping rules, and the reason for termination are also not reported.

The intervention descriptions are brief: participants in both groups were asked to download and use an application, and both groups received adjunctive treatment as usual. The registry does not detail application content, adherence, the nature of usual care, or how the applications differed. It also labels OG000 as the control while describing OG000 as WB001 and ED001 as the educational control; that designation is not explained.

These limitations concern the information in the registry record. They do not establish that additional information or analyses do not exist elsewhere; they indicate that such details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: [Study Evaluating the Efficacy and Safety of a Digital Therapeutic as an Adjunct to TAU in Postpartum Depression] — NCT05551195.
• Registry record last updated 2025-02-24; status last verified 2025-01.
• Results first posted 2025-02-24.
• 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.

18. Record Summary

NCT05551195 is a terminated, randomized, parallel-group multicenter study sponsored by Woebot Health. It evaluated WB001 and ED001, each as an adjunct to treatment as usual, among women with postpartum depression. The official title describes mild to moderate postpartum depression. The registry lists triple masking, while the official title says double-blind.

The primary endpoint was change in HAM-D score from baseline to Week 8. The posted summaries include 1 participant for WB001 and 5 for educational control; one participant is noted as missing Week 8 data. No two-arm comparison is estimable from the posted values, and no analyses are posted. The record lists 8 secondary outcomes and reports zero serious adverse events, zero other adverse events, and zero deaths in both groups. The reason for termination and key statistical analysis details are not reported in the registry record.