Intervention for IPV Perinatal Women- RCT: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Intervention for IPV Perinatal Women- RCT (NCT02370394), a randomized na study. The primary outcome is Composite Abuse Scale (CAS)-CAS Victimization Total Score.

2. Clinical Question

The study summary describes development and assessment of a computer-delivered intervention intended to address barriers to early identification and intervention for perinatal women with intimate partner violence who were seeking mental-health treatment. The clinical question is therefore about the intervention in this stated population and its relationship to the registered victimization measure. The available record identifies a randomized comparison, but it does not supply a single outcome contrast that can be interpreted as the estimated difference between the groups.

The reported endpoint description says that lower Composite Abuse Scale victimization scores represent less victimization. That direction helps explain the scale, but it does not establish that either group improved, nor does it establish a difference between groups. Without a reported estimate, uncertainty interval, or statistical test for a defined comparison, the magnitude and precision of any difference cannot be assessed from the registry record.

3. Trial Design

The study, titled “Intervention for IPV Perinatal Women- RCT,” is registered as NCT02370394. Its official title is “Computer-Based Intervention for Victimized Perinatal Women With Mental Illness (Phase 1).” The record lists randomized allocation, no masking, and a parallel-group model. The phase is listed as NA. Enrollment is reported as 53, and the study status is completed. The listed sponsor is Women and Infants Hospital of Rhode Island.

The registry dates the study start to 2015-03 and primary completion to 2016-09. Those dates describe the registered study timeline; they do not supply information about when individual participants were enrolled or assessed. The primary endpoint was assessed at baseline and again three months later. The ROSE arm description says participants completed baseline and follow-up assessment three months later. The control-condition description is incomplete after its account of baseline assessment, so the control follow-up procedure is not reported in the registry record.

Participants assigned to ROSE received the computer-delivered intervention and an in-person booster. The record places the booster within a month after the intervention. Participants in the control condition completed questions about television show preferences, watched brief videos of popular entertainers or shows, and rated subjective preferences. The record does not describe masking; it explicitly lists masking as none.

The study records the Composite Abuse Scale victimization total score as the primary outcome and lists Motivation Scale and the Readiness to Change Contemplation Ladder as secondary outcomes. Five outcome measures and ten analyses are reported, but the full mapping of analyses to outcomes, time points, and groups is not reported in the registry record. The registry record also does not report the allocation procedure, the method used to generate the random sequence, concealment procedures, or a detailed statistical analysis plan. The record describes a completed randomized study, but its design description alone does not establish the outcome findings.

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 number of participants analyzed differs from baseline to follow-up because 2 dropped out of the ROSE intervention group, and 1 dropped out and 1 withdrew (no longer interested) from the control condition.

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 Composite Abuse Scale (CAS)-CAS Victimization Total Score, measured over Assessed at baseline, and again three months later.

The primary endpoint is the Composite Abuse Scale (CAS) victimization total score, expressed in units on a scale and summarized as a mean with standard deviation. The description says the instrument is a widely used self-report of behaviors with 36 items, while the registered score is the CAS Victimization Total score. Items are scored from 0 to 5, with Never scored as 0 and Daily scored as 5; the scale range is 0-180. The registry record says the mean score was used to calculate differences between baseline and follow-up and between groups.

The endpoint is intended to quantify reported victimization rather than to report a diagnosis or a clinical event count. According to the record, a lower score corresponds to less victimization. Repeated assessment at baseline and three months later permits the study to consider scores over time, while a between-group comparison could address how the groups' scores differ. The registry record does not report a single reduced comparison, so the actual endpoint contrast and its estimate are not reported in the registry record.

A mean describes the center of the observed scores, while a standard deviation describes their spread. Those summaries can help characterize a continuous measure, but they do not by themselves show how precise a treatment contrast is. For that, an effect estimate and an uncertainty measure such as a confidence interval would be needed. These are not reported in the registry record. The record also notes that the number analyzed differed between baseline and follow-up because of dropout and withdrawal, which matters when interpreting a mean over time.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods help distinguish what a study set out to compare from what the observed data can establish. In this record, the design is randomized and parallel, and the primary measure is continuous. However, a single primary-outcome comparison and its numerical result are not reported in the registry record. The concepts below explain how such results are generally interpreted; they do not describe analysis procedures that the record does not document.

What does randomization contribute?

In general, random allocation is intended to make treatment assignment independent of participants' measured and unmeasured characteristics, on average across repeated implementations of a trial. It supports a causal comparison when assignment is preserved and outcome data are analyzed appropriately. Randomization does not guarantee that groups will be identical in a particular study, especially in a study with limited enrollment. The registry record does not report the sequence-generation method, allocation concealment, or baseline group summaries, so those details cannot be evaluated.

A parallel-group design assigns participants to a group and compares outcomes across groups. The record identifies ROSE and a control condition, but it does not specify a single endpoint contrast to use for interpretation. For a repeated measure, an analysis might compare follow-up values adjusted for baseline, compare change from baseline, or model measurements across time. These approaches answer related but not identical questions and rely on different modeling choices. Which approach was used is not reported in the registry record.

What do means and standard deviations describe?

For a continuous measure, the mean is an average that describes the center of the observed scores. The standard deviation describes how dispersed the scores are around that mean. Neither summary is the same as an estimated treatment effect. An effect estimate would define a contrast, such as a between-group difference at a specified time or a difference in change over time. That definition matters because different contrasts can produce different answers, even when based on the same scale.

For the CAS, lower scores are described as better or less victimization. Interpreting a contrast therefore requires knowing which groups and time points are being compared and the direction in which the contrast is calculated. The registry record does not report those details as a single primary comparison. It also does not report the underlying means and standard deviations in the information available for this explanation.

How does uncertainty differ from an effect estimate?

In general, an effect estimate gives the size and direction of a contrast, while a confidence interval expresses uncertainty about that estimate under the statistical model and sampling process. A narrow interval indicates greater precision than a wide interval, but precision alone does not establish clinical importance. A p-value, when used, addresses compatibility of the data with a null hypothesis under specified assumptions; it does not give the probability that the null hypothesis is true or measure the size of an effect. No confidence interval or p-value for a single primary contrast is reported in the registry record.

A statistically significant result, if reported, would not by itself establish that a difference is meaningful to participants. Conversely, an estimate with substantial uncertainty would not establish that groups are equivalent. Equivalence or noninferiority requires an appropriately specified design and margin; such a margin is not reported in the registry record. Because the primary comparison is not reduced to one result, these interpretations cannot be made for the registered endpoint.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model assumes. A model that understates variability can produce uncertainty intervals that are too narrow and tests that are too optimistic. The appropriate response depends on the outcome distribution and analysis model. The registered primary endpoint is a mean of a continuous scale, but the registry record does not report a model or diagnostics, and it does not report whether over-dispersion was assessed. It should not be assumed that a particular model was used or that this issue affected the study.

How should a planning illustration be read?

Sample-size calculations are planning tools, not results from participants. The record states that its planning illustration is computed from the observed effect, assumes a two-sided significance level of 0.05, equal allocation between the groups, and a normal approximation, and uses fractions of the observed effect as the effects to detect. It also states that such figures are not the trial's own design. Thus, a planning figure would describe a hypothetical study under those assumptions, not the power observed in this completed trial. It cannot establish that the trial was adequately powered, and it should not be called observed power.

For outcomes other than time-to-event outcomes, the planning convention is participants per arm for 80% power or other stated power levels. The record's general convention for time-to-event planning is total events. The primary endpoint is a continuous scale, but the numerical planning outputs and the fractions used are not reported in the registry record. The planning assumptions should not be confused with a reported statistical analysis plan or with evidence that the study was designed around the observed effect.

What does missingness change?

When measurements are missing, a complete-case summary may describe only participants with observed data at the relevant assessment. If loss to follow-up is related to outcomes or participant characteristics, the observed group summaries may not represent all participants assigned. Statistical methods can make different assumptions about missing data; those assumptions are important because they affect estimates and uncertainty. The registry record reports dropouts and a withdrawal but does not report the analysis method for missing outcomes, the amount missing for each endpoint assessment, or whether any sensitivity analyses were performed.

10. Multiplicity, Missing Data and Analysis Populations

The record lists 5 outcome measures and 10 analyses. It names Motivation Scale and the Readiness to Change Contemplation Ladder as secondary outcomes, in addition to the primary CAS victimization score. Multiple outcomes, time points, or comparisons can make interpretation more complex. In general, examining many hypotheses can increase the chance of a nominally notable finding, and adjustments or a prespecified hierarchy may help control that risk. Whether such procedures were used is not reported in the registry record. The analysis count does not identify which analyses were primary, secondary, or exploratory.

The registry explicitly notes that the primary outcome has several categories or time points and was not reduced to one comparison. Consequently, a single primary estimate and its uncertainty are not reported in the registry record. A reader cannot infer which time point or contrast was treated as definitive, or whether there was a prespecified strategy for choosing among possible contrasts.

The CAS description says that the number analyzed differs between baseline and follow-up. It attributes this to 2 dropouts in the ROSE intervention group and 1 dropout plus 1 withdrawal, no longer interested, in the control condition. These are reported reasons for the change in analysis numbers, but the exact endpoint-specific analysis populations are not reported in the registry record. The registry record also does not report how missing outcome values were handled, whether analyses followed randomized assignment, or whether sensitivity analyses tested alternative assumptions about missingness.

Randomization identifies how groups were assigned, but it does not by itself resolve missing-data or multiplicity questions. Those issues affect how confidently results can be interpreted. In the absence of a reported analysis strategy and a single primary comparison, the registered design and outcome descriptions do not establish the robustness of any outcome finding.

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

Serious events with the highest arm-level rate: ipv-no medical care (16.0%), hospitalization-general medical issue (14.3%), hospitalization-inpatient-psyc (10.7%), ipv related emergency room visit (7.1%).

The registry record lists serious adverse events separately by group. In the ROSE Program group, 28 participants are listed as at risk and 9 as affected by serious adverse events. In the control condition, 25 are listed as at risk and 6 as affected. Deaths are listed as 0 in each group, and other adverse events affected are also listed as 0 in each group. These are descriptive counts as recorded; they are not a formal comparison between groups.

The record lists the following serious-event terms and percentages: IPV-no medical care, 16.0%; hospitalization-general medical issue, 14.3%; hospitalization-inpatient-psyc, 10.7%; and IPV related emergency room visit, 7.1%. The denominator and calculation method for these percentages are not reported in the registry record. The categories may describe event terms rather than a distinct participant-level tally, so they should not be summed or treated as mutually exclusive based on the information recorded.

Serious-event entries do not alone establish whether there is a difference in safety between groups. The registry record does not report a statistical test, uncertainty interval, attribution assessment, or detailed event narratives. A zero entry for deaths or other adverse events affected is limited to the categories and reporting captured in this record; it does not provide a broader assessment of risk.

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

The registry record describes a randomized study of a behavioral, computer-delivered intervention for perinatal women with intimate partner violence seeking treatment for mental illness. It identifies the ROSE intervention, a control condition, the CAS victimization total score as the primary endpoint, and two named secondary measures. It also provides a description of serious adverse event entries. These registry entries characterize the registered study, not its effectiveness.

The endpoint is a self-reported continuous measure, and the record explains that a lower score corresponds to less victimization. That scale direction can guide interpretation if a score contrast is reported. But the record does not reduce the primary outcome's categories or time points to a single comparison. The corresponding effect estimate, its uncertainty, and a statistical test are not reported in the registry record. It is therefore not possible to say from these details whether the intervention group had higher or lower scores, or whether the groups differed.

The record also does not establish why any observed changes may have occurred. A before-and-after pattern, if available, would not by itself separate intervention effects from other changes over time. A group comparison would need a clearly defined contrast and appropriate handling of missing data. The registry record notes dropout and withdrawal but does not describe the missing-data method or analysis population in detail.

The planning assumptions refer to illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such illustrations are not evidence about outcomes, and they do not represent observed power. The numerical planning outputs are not reported in the registry record. Similarly, the serious adverse event counts are descriptive and do not establish a statistically tested difference in safety.

The appropriate interpretation is therefore bounded: the record documents a completed randomized trial, its registered outcome framework, participant loss between assessments, and reported serious-event figures. It does not provide enough information to draw a conclusion about the primary outcome contrast, its precision, or the robustness of analyses. Lack of a reported comparison is not evidence that groups were the same, just as the intervention's stated purpose is not evidence that it produced a particular benefit.

13. Why This Trial Matters Statistically

This trial is statistically informative as an example of the gap between a registered question and an interpretable outcome estimate. The record gives a recognizable randomized parallel-group design and describes a continuous primary measure with repeated assessment. It also records secondary measures, multiple analyses, and losses between baseline and follow-up. Those features make clarity about the estimand, analysis population, and missing-data approach especially important.

For a repeated continuous outcome, several reasonable contrasts may be considered: group differences at follow-up, within-group changes, or differences in change between groups. These do not answer precisely the same question. The registry record says the primary outcome has several categories or time points and is not reduced to a single comparison, leaving the principal contrast unclear. Reporting a defined contrast with an effect estimate and uncertainty would make the result more interpretable than a scale description alone.

The recorded analysis count also makes multiplicity relevant. When several outcome measures or analyses are considered, readers need to know which were prespecified and how the evidence was interpreted. The registry record does not report a multiplicity strategy. Likewise, because the analyzed number differed between baseline and follow-up, explaining how missing values were handled would help readers assess whether the conclusions depend on who remained observed.

Finally, the planning assumptions illustrate a general distinction between design calculations and study results. Planning illustrations based on observed effects are not the same as a prospective design rationale, and they are not observed power. The numerical outputs and their effect fractions are not reported in the registry record. These distinctions matter when interpreting what a completed study can support. The record is useful for understanding the trial's registered structure, but the absence of a single primary contrast prevents a statistical reading of its principal outcome.

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.

The main limitation for interpretation is that the primary outcome is not presented as one comparison. The registry record identifies the CAS victimization total score and its assessment at baseline and three months later, but it does not report a single primary contrast, an effect estimate, or an uncertainty measure. As a result, the magnitude, direction, and precision of the primary outcome findings are not reported in the registry record.

Information about analysis procedures is limited. Randomized allocation is recorded, but the method for generating and concealing assignment is not reported in the registry record. The record does not report a detailed statistical analysis plan, the model used for the primary outcome, the exact analysis population, or whether analyses followed randomized assignment. It also does not report whether multiplicity adjustments, interim analyses, or sensitivity analyses were used.

Missingness is a specific concern because the record states that the number analyzed changed between baseline and follow-up. It attributes this to 2 dropouts in the ROSE group and 1 dropout and 1 withdrawal in the control group. The timing of those losses, endpoint-specific denominators, and method for handling missing data are not reported in the registry record. Without those details, the potential effect of incomplete follow-up cannot be assessed.

The control condition description ends after stating that participants completed baseline assessment, so its follow-up procedures are not reported in the registry record. The record also does not report detailed eligibility criteria, baseline outcome summaries, or contextual information needed to judge how broadly the results might apply. The safety information consists of listed counts and terms, without a reported formal group comparison or details on event assessment. These omissions do not imply that procedures were absent; they indicate that they are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Intervention for IPV Perinatal Women- RCT — NCT02370394.
• Registry record last updated 2018-06-04; status last verified 2018-05.
• Results first posted 2018-01-05.
• 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

NCT02370394 is a completed, randomized, unmasked, parallel-group study sponsored by Women and Infants Hospital of Rhode Island. Enrollment is reported as 53; the phase is listed as NA. The record gives a study start of 2015-03 and primary completion of 2016-09.

The intervention was the behavioral ROSE Program, including a 35-40-minute Tablet PC intervention and an in-person 10-15-minute booster within a month. The control condition used television preference questions, videos, and ratings. The primary endpoint was the CAS victimization total score, assessed at baseline and three months later; lower scores represent less victimization.

The registry record notes differing analyzed numbers between assessments, including 2 dropouts from ROSE and 1 dropout plus 1 withdrawal from control. It lists serious adverse event affected counts of 9 among 28 at risk for ROSE and 6 among 25 at risk for control, with 0 deaths and 0 other adverse events affected in each group. A single primary-outcome comparison and its estimate are not reported in the registry record.