Trial to Test Effectiveness of Depression Intervention for Mistreated Older Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial to Test Effectiveness of Depression Intervention for Mistreated Older Adults (NCT04258579), a na na study. The primary outcome is Change in Clinically Significant Depressive Symptoms as Measured by the MADRS.

2. Clinical Question

The central statistical question is therefore about the course of symptoms within a single group, not a randomized contrast between treatments. A single-group study can describe the observed outcome pattern, but it cannot by itself separate intervention-associated change from other influences over time. Those influences might include changes in circumstances, other services, or the natural course of symptoms; the record does not report their contribution.

The stated clinical context is elder mistreatment, with PROTECT designed to work alongside mistreatment-resolution services that offer safety planning, support services, and links to legal services. The brief summary also describes community mental-health linkage as part of the comparison originally envisaged. However, the registered intervention and arm information identify Video PROTECT as the study arm, and no comparator outcome data are given. Accordingly, the question that can be addressed from the reported design is descriptive: what MADRS measurements were planned, how many participants were analyzed at each timepoint, and how precisely can a treatment effect be estimated from the posted record? The record does not provide the numerical outcome estimates needed to answer the last part.

3. Trial Design

NCT04258579 is titled Trial to Test Effectiveness of Depression Intervention for Mistreated Older Adults. Its official title is EM/PROTECT-Hybrid: Improving Depression in Elder Mistreatment Victims. The sponsor is Weill Medical College of Cornell University. The study status is COMPLETED, with a start date of 2020-07-15 and a primary completion date of 2022-08-11. The phase is listed as NA.

The structured design identifies a SINGLE GROUP model, allocation as NA, and masking as NONE. Enrollment is 40. The intervention is PROTECT with Technology Augmentation, classified as behavioral; the arm description is Video PROTECT. Participants were to receive therapy once a week for 9 weeks. These details describe a within-group follow-up design, rather than a randomized, masked comparison.

There is an important difference between the brief summary and the structured design details. The summary says investigators intended to compare EM/PROTECT with mistreatment services enhanced by staff training in links to community mental-health services, and that they intended to enroll 50 subjects. The registered enrollment is 40, while the structured design lists one group and no comparator. The registry record does not explain this difference, identify a control-arm sample, or report a revised protocol. It would therefore be inappropriate to interpret the reported study as a completed comparison between the two approaches.

The primary outcome is scheduled at 6 weeks, 9 weeks, and 12 weeks. The record gives analyzed counts of 40 at baseline, 30 at week 6, 30 at week 9, and 26 at week 12, attributing the decline to withdrawal by the study team or at participants' request. Four outcome measures and 25 analyses are listed, but the registry record does not describe how those analyses were organized or which analysis produced each result. No statistical analysis plan, interim-analysis plan, or detailed sample-size justification is reported in the registry record. The stated planning illustration assumptions are not enough to establish the original design rationale.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The overall number of participants analyzed was 40 at baseline, 30 at week 6, 30 at week 9, and 26 at week 12. The decline in participants analyzed at each time point is due to participant withdrawal by the study team or by their request.

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 Change in Clinically Significant Depressive Symptoms as Measured by the MADRS, measured over 6 weeks, 9 weeks, 12 weeks.

The primary outcome is titled change in clinically significant depressive symptoms measured by the Montgomery Asberg Depression Rating Scale (MADRS). The unit is units on a scale, and the outcome parameter is a mean with standard deviation as the dispersion measure. Measurements are scheduled at 6 weeks, 9 weeks, and 12 weeks. The record states that the values at these timepoints are absolute scores, even though the outcome title refers to change.

The registry describes score categories as 0 to 6 for normal or symptom absent, 7 to 19 for mild depression, 20 to 34 for moderate depression, and >34 for severe depression. These categories give context to the scale's numerical range as described in the record. The endpoint is relevant to the stated study question because it tracks depressive symptom severity over follow-up in a population described as experiencing elder mistreatment.

In general, repeated measurements of a continuous scale can help describe whether the group's average score shifts over time. A mean gives a central summary, while a standard deviation describes spread among participant scores. Neither statistic alone shows whether change was similar for all participants. A change score also differs from an absolute follow-up score: the former requires a baseline-to-follow-up contrast, while the latter reports the value at a specific measurement occasion.

The registry record does not report the MADRS means, standard deviations, change estimates, confidence intervals, or tests of statistical significance. It therefore does not show the size or precision of any symptom change, nor whether the average score crossed a category boundary.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as a continuous measure summarized by a mean and standard deviation. In general, a mean is the arithmetic center of a set of measurements. It can be useful for describing the average symptom score in a group, but it does not show the full distribution. A standard deviation summarizes how dispersed individual scores are around the mean. A larger standard deviation indicates more variability among participants, while a smaller one indicates scores are more concentrated. These are general interpretations; numerical MADRS summaries are not reported in the registry record.

What does a continuous outcome summary show?

For a continuous outcome such as a scale score, reporting the mean alongside a measure of spread makes it possible to understand both the group's central level and the variation among individuals. The mean is sensitive to unusually high or low scores, so the distribution and number of observations can affect how representative it is. The standard deviation is not the uncertainty of the mean. General statistical methods often quantify uncertainty in a mean with a standard error or confidence interval. Neither a confidence interval nor the underlying score summaries for the primary outcome are reported in the registry record.

How should repeated measurements be understood?

The MADRS outcome has scheduled assessments at 6 weeks, 9 weeks, and 12 weeks, in addition to the baseline count. Repeated observations allow researchers to describe how scores vary across follow-up occasions. Measurements from the same participant are generally related rather than independent: a participant's earlier score may be informative about their later score. In general, an analysis of repeated measurements should account for this within-person relationship if it estimates an overall trajectory or change. The registry record does not identify the model or method used for the primary analysis, so it is not possible to say whether a repeated-measures model, separate timepoint summaries, or another approach was used.

There is also a distinction between an absolute score and a change score. An absolute score refers to the measured scale value at a particular occasion. A change score compares that value with a reference, commonly a baseline measurement, using a defined calculation. The registry's outcome title refers to change, but its description says the presented scores are absolute values. It does not report the numerical values or a defined analysis contrast. The interpretation should therefore follow the stated measurement type and avoid treating an unreported contrast as an observed result.

Why do participant counts affect interpretation?

The number of participants analyzed declines over follow-up, from 40 at baseline to 30 at week 6 and week 9, and 26 at week 12. In general, fewer observations mean that estimates may be less precise, all else being equal. More importantly, a changing set of participants can affect comparability of timepoint summaries. If participants who leave differ systematically from those who remain, the later mean may not describe the same underlying mix of people as the earlier mean. The record attributes the declines to withdrawal by the study team or by participant request, but it does not report reasons by participant, the timing of withdrawal, or whether missingness was associated with MADRS scores.

General approaches to incomplete repeated data include analyzing available observations, modeling repeated outcomes under assumptions about missingness, or using imputation. Each approach requires choices and assumptions. For example, some methods assume that missingness can be explained by observed information, while others require stronger assumptions about unobserved outcomes. The registry record does not report a missing-data method or assumptions, so no particular approach can be attributed to this study.

What can a single-group design estimate?

A single-group design can summarize outcomes observed among participants who received the intervention. It does not, on its own, estimate a causal contrast against a concurrent control group. General causal reasoning relies on comparing observed outcomes with a credible estimate of what would have occurred in the absence of the intervention or under an alternative intervention. Without a control arm, changes over time may also reflect external events, concurrent services, measurement variation, or the course of symptoms. This is a limitation of design, not a claim about what happened to participants.

The registry's brief summary describes an intended comparison, but the structured information identifies a single group, allocation as NA, and one arm. There are no comparator outcomes in the record. A comparative effect estimate therefore cannot be reconstructed from the posted design information.

What do significance tests and intervals contribute?

In general, a significance test evaluates how compatible observed data are with a specified null hypothesis under an analysis model. A p-value does not measure the probability that a treatment works, and it does not describe the size of an effect. A confidence interval expresses uncertainty around an estimate on the scale of the outcome. Together, estimates and intervals can convey magnitude and precision more clearly than a significance test alone. The registry record does not report primary-outcome p-values, confidence intervals, or numerical effect estimates.

General concerns about over-dispersion apply when observed variability is greater than a statistical model expects. If a model assumes less variation than the data contain, uncertainty can be understated and inference can become too confident. Whether over-dispersion is relevant depends on the outcome and model; it is not possible to assess it without the data and analysis details. The registry record does not describe an over-dispersion assessment.

How should the planning illustration be read?

The record describes a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. For time-to-event outcomes, the illustration is specified in total events; for other outcomes, it is specified as participants per arm. These are planning assumptions, not evidence that the original study used a two-arm design. The registry record lists a single-group study, and the specific planning outputs are not reported in the registry record. Any sample-size illustration based on an observed effect is a hypothetical design calculation, not observed power and not proof that the study was adequately powered.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures and 25 analyses. Those counts indicate that multiple outcomes or analyses are registered, but they do not reveal how the analyses were prioritized, whether they were prespecified in a statistical analysis plan, or whether any adjustment for multiple testing was used. The record identifies a primary outcome, but multiplicity procedures are not reported in the registry record.

In general, testing many outcomes or contrasts can increase the chance of at least one apparently unusual result under a collection of null hypotheses. Statistical approaches can address this in different ways, including adjusting significance thresholds or emphasizing estimation and uncertainty rather than treating each test as a separate definitive finding. Which approach is appropriate depends on the aims and analysis plan. No specific correction or testing hierarchy is reported in the registry record, so none should be assumed.

Missing outcome data are relevant because analyzed counts decrease from 40 at baseline to 30 at week 6, 30 at week 9, and 26 at week 12. The record attributes the decline to participant withdrawal by the study team or by participant request. It does not report a missing-data analysis, imputation method, sensitivity analysis, or the assumptions used to handle incomplete follow-up. In general, conclusions from incomplete data depend on whether participants with missing measurements differ from those observed and on how the analysis accounts for that possibility.

The analyzed counts describe how many participants contributed at each stated timepoint, but the registry record does not specify a formal analysis population such as intention-to-treat or per-protocol. Nor does it report whether the same participants were included in each analysis or how missing baseline or follow-up values were handled. These omissions limit reproducibility and make it difficult to judge how the changing follow-up population affects outcome estimates. The record's 25 analyses do not, by themselves, identify which population or missing-data method was used.

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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: psychiatric hospitalization (2.5%), hospitalization due to seizures (2.5%).

For the Video PROTECT group, the registry reports 40 participants at risk for serious adverse events, with 2 affected, and 40 at risk for other adverse events, with 4 affected. Deaths are reported as 0. The serious-event terms listed are Psychiatric Hospitalization at 2.5% and Hospitalization due to Seizures at 2.5%.

These are descriptive registry counts. The at-risk numbers are the group sizes recorded for the safety outcomes; the affected counts describe participants with the corresponding event category as posted. The record does not provide event-level timing, duration, severity details beyond the serious classification, or an assessment of whether events were related to the intervention. It also does not explain whether a participant could contribute to more than one event category. The listed serious-event terms should therefore be read as the registry's reported categories, not as a complete narrative of each event.

In general, adverse-event counts do not establish causation. Without a comparator, the observed counts cannot be contrasted with an untreated or alternative-treatment group to estimate excess risk. The reported counts also do not establish that no other events occurred beyond those captured in the registry. They describe the posted safety record for this group and should be interpreted alongside the study's single-group design and the limits of the information recorded.

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

The registry record establishes that NCT04258579 was completed as a single-group study of Video PROTECT, with a planned MADRS assessment schedule and a declining number of participants analyzed over follow-up. It also reports serious and other adverse-event counts for the intervention group. These details help characterize the study and the available follow-up, but they do not provide the numerical primary-outcome results.

The record does not report MADRS means, standard deviations, changes, confidence intervals, or primary-outcome significance tests. Consequently, it does not show whether scores moved in either direction, how large any change was, how variable scores were, or how precise an estimate might be. The score categories in the outcome description provide scale context but cannot be used to infer where participants' scores fell without the actual measurements.

The design also defines what can be concluded. A single-group study can describe observed outcomes among participants receiving the intervention. It cannot establish what would have happened to the same participants under another service or without the intervention. The brief summary mentions an intended comparison, but the structured record reports one group and no comparator results. A comparative treatment effect is therefore not available from the registry record.

The changing analyzed counts add uncertainty about who is represented at later follow-up. The record attributes fewer participants at each later timepoint to withdrawal by the study team or by participant request, but does not report whether withdrawal was related to symptoms or other factors. Without missing-data methods and sensitivity analyses, it is not possible to assess how incomplete follow-up may affect the outcome summaries.

These limitations do not mean that the intervention had no effect, nor do the posted safety counts prove that the intervention caused the events. They mean that the registry information is insufficient to make either conclusion. A sound reading separates the study's documented design and participant counts from outcome claims that require numerical estimates, uncertainty measures, and a suitable comparison.

13. Why This Trial Matters Statistically

This study raises several important statistical issues in a clinically complex setting. The target population is described as older adults experiencing mistreatment and depression, and the intervention is intended to operate alongside services addressing safety and support. That context makes the distinction between describing participant outcomes and attributing changes to a particular intervention especially important. A single-group follow-up can document what happened among those enrolled, but causal conclusions require a credible comparison that the structured registry record does not identify.

The record also illustrates why outcome labels and measurement details should be read together. The primary outcome is titled as a change in depressive symptoms, while the description says that values at each timepoint are absolute scores. These quantities answer different questions. Clear reporting of baseline values, follow-up values, change estimates, and uncertainty would allow readers to understand the trajectory without conflating absolute status with change.

Follow-up completeness is another central statistical concern. The analyzed count decreases across the scheduled assessments, and the record attributes this to withdrawal. The pattern matters because incomplete follow-up can affect both precision and representativeness. Reporting reasons for missingness, the analysis population, and methods for handling missing values would help readers judge the robustness of any outcome estimate.

Finally, the registry describes multiple analyses and a planning illustration but does not report the primary numerical results or the detailed analysis plan. Transparent reporting of estimates, confidence intervals, multiplicity decisions, and missing-data methods supports reproducibility and balanced interpretation. The record's design and safety information remain useful, but the statistical case for an outcome conclusion cannot be assessed without those elements.

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 limitations are specific to the information recorded for this study. First, the structured design is single-group, with allocation listed as NA and masking as NONE. Although the brief summary describes a comparison that investigators intended to make, the registered arm information identifies Video PROTECT alone. The registry record does not explain this discrepancy or provide comparator data, which prevents a comparative treatment-effect interpretation.

Second, the primary outcome is described as change in depressive symptoms, but the registry says the scores at each timepoint are absolute values. The numerical MADRS results are not reported in the registry record. There are no reported means, standard deviations, change estimates, confidence intervals, or primary-outcome tests from which to evaluate the direction, magnitude, or precision of symptom outcomes.

Third, the number analyzed declines from baseline through follow-up. The registry attributes this to withdrawal by the study team or by participants, but does not give further detail about timing, reasons, symptom status at withdrawal, or the handling of missing measurements. The record also does not report whether analyses used a defined intention-to-treat or per-protocol population.

Fourth, the registry lists four outcomes and 25 analyses without describing a multiplicity strategy or a detailed statistical analysis plan. It also does not report an interim-analysis plan or a specific sample-size justification. The planning assumptions described for an illustration do not resolve these omissions or establish the original study's design rationale.

Finally, the safety information is limited to event counts and listed serious-event terms. The record does not provide detailed event narratives or attribution to the intervention. These limitations concern what can be learned from the registry record and do not establish that unreported analyses or information did not exist elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Trial to Test Effectiveness of Depression Intervention for Mistreated Older Adults — NCT04258579.
• Registry record last updated 2023-08-28; status last verified 2023-08.
• Results first posted 2023-08-28.
• 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

NCT04258579 is a completed, single-group study sponsored by Weill Medical College of Cornell University. It evaluated PROTECT with Technology Augmentation, described in the arm information as Video PROTECT, among older adults in the context of elder mistreatment and depression. Enrollment is 40; the brief summary describes an intended enrollment of 50 and a comparison, while the structured record identifies one group and no comparator.

The primary outcome is MADRS depressive symptom scores at 6 weeks, 9 weeks, and 12 weeks. Analyzed counts are 40 at baseline, 30 at week 6, 30 at week 9, and 26 at week 12. The registry says these are absolute scores, but numerical primary-outcome results are not reported in the registry record. Serious adverse events affected 2 participants, other adverse events affected 4, and deaths are reported as 0. The record does not report a detailed statistical analysis plan, missing-data method, or multiplicity procedure.