Treating Caregiver Depression to Improve Childhood Asthma: Impact and Mediators: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Treating Caregiver Depression to Improve Childhood Asthma: Impact and Mediators (NCT02809677), a na na study. The primary outcome is Hamilton Rating Scale for Depression (HRS-D) Change Scores.

2. Clinical Question

The study's stated rationale concerns the relationship between caregiver depression and a child's asthma, rather than the efficacy of an antidepressant against a control treatment. Its brief summary describes plans to examine asthma control, spirometry and quality of life alongside depressive symptoms. It also describes testing whether improved medication adherence or decreased child anxiety or depression mediates an association between caregiver improvement and child outcomes.

These are research questions and planned analyses, not findings. The registry record does not report numerical results for the primary endpoint or for the child asthma outcomes. It also does not report results establishing whether changes in caregiver depression came before changes in child outcomes. The study design can describe change and temporal relationships within the observed group, but without a randomized comparison it cannot by itself separate the effects of caregiver treatment from other changes occurring over time.

3. Trial Design

The registry identifies this as a completed, single-group, non-interventional longitudinal study. The arm description calls it a 52 week non-placebo controlled and non-randomized clinical research study. The brief summary describes a one-year repeated-measures, within-subject design. These descriptions indicate that participants were followed over time, rather than assigned to randomized treatment and control groups.

Caregivers may choose to receive antidepressant medication. The registry does not describe a study-assigned medication regimen, and the arm description does not provide further details about the medication considered. The study's stated rationale explains that investigators considered a randomized control trial but judged withholding medication from caregivers diagnosed with Major Depressive Disorder for the proposed study duration to be unacceptable. The stated aim was to study whether improvement in caregiver depression relates to child outcomes, not to test the efficacy of antidepressants for caregiver depression.

The registry lists allocation as NA and masking as NONE. There is no like-for-like control comparison. A participant's measurements at different times can be related to one another, but temporal change does not offer the same protection against confounding as randomized assignment. Other influences that change over time could coincide with improvements or worsening in either caregiver symptoms or child asthma outcomes.

The record lists 205 as enrollment and gives the study status as completed, with a start date of 2016-06 and primary completion of 2020-10. Those registry fields do not establish how many participants completed follow-up, had data available at each measurement, or were included in any analysis. Such analysis-population details are not reported in the registry record. The registry record also does not report a statistical analysis plan beyond the methods described in the brief summary.

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:

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 (HRS-D) Change Scores, measured over From date of Baseline until the date of first documented progression, assessed up to 52 weeks..

The registered primary endpoint is change from baseline in the caregiver's depression score, measured with the Hamilton Rating Scale for Depression (HRS-D). The measure is a 17 item scale with a range of 0-53. The registry describes score bands as normal at 0-7, mild at 8 - 13, moderate at 14 - 18, severe at 19-22, and very severe at > 22. The endpoint is a continuous score, so a change score expresses how a measured score differs from its baseline value.

Repeated measurement of caregiver symptoms is relevant to the study's central question because the investigators seek to examine whether improvement in caregiver depression precedes changes in child asthma outcomes. A longitudinal record of caregiver scores can be considered alongside child measures over time. The registry's primary endpoint, however, is caregiver depression change; it is not itself a measure of a child's asthma control, lung function, quality of life, or healthcare use.

The timeframe is described as baseline until the date of first documented progression, assessed up to 52 weeks. The registry record does not provide a numerical estimate of HRS-D change, its variability, or a confidence interval. It therefore does not show the direction, size, or precision of the observed primary endpoint result. It also does not report enough detail to determine how the first documented progression wording was applied in practice.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The brief summary describes two related methods: cross-lagged panel modeling (CLPM) and structural equation modeling (SEM), with maximum likelihood estimation for the SEM. These are methods for studying relationships among repeated measurements. Their appearance in the registry describes the proposed analytic approach; it does not establish that the analyses were carried out. The registry record lists zero analyses and reports no fitted model estimates.

What does a longitudinal model ask?

A longitudinal analysis uses measurements collected at different times to study patterns of change and relationships between variables over time. In this study, the intended variables include caregiver depressive symptoms and child asthma outcomes. Repeated observations can help address whether an earlier measurement is associated with a later measurement, while accounting for the fact that the variables are measured more than once. The registry record does not report the specific measurement schedule or the model's exact specification.

What is a cross-lagged panel model?

In general, a CLPM examines whether a measure at an earlier time is associated with another measure at a later time, while also considering how each measure relates to its own earlier value. In the study description, the intended question is whether caregiver improvement precedes child asthma improvement. That is a question about temporal ordering and association. A cross-lagged association alone does not establish causation: other factors may influence caregiver depression and child asthma over time, and the record does not describe a randomized intervention contrast that would isolate a causal effect.

Interpretation of a cross-lagged model depends on which observations and time points are included, how the measures are represented, and what assumptions are made about the relationships. Those implementation details and model estimates are not reported in the registry record. The stated method should therefore be read as an analytic plan, not as evidence that a temporal association was found.

What does structural equation modeling add?

In general, SEM represents relationships among measured variables in a system of equations. It can evaluate whether data are consistent with a proposed structure linking caregiver symptoms, potential mediators and child outcomes. The registry describes testing whether improved adherence and/or decreased child anxiety or depression mediated an effect. A mediation analysis asks whether a proposed intermediate variable may help account for an association between an exposure or change and an outcome. It does not, by itself, establish that the intermediate variable caused the outcome. Strong causal interpretation requires assumptions about temporal order and confounding that cannot be confirmed just by fitting a model.

The record does not report mediation estimates, uncertainty intervals, model fit information, or evidence that the proposed pathways were supported. It also does not report how adherence, child anxiety or depression, or the child asthma outcomes were operationalized in the models. Consequently, the named methods clarify the research questions but do not yield an interpretable result from the posted record.

What is maximum likelihood estimation?

In general, maximum likelihood estimation selects model parameters that make the observed data most compatible with the specified statistical model. In SEM, this approach depends on the model being specified and on assumptions about the data and their distribution. Model estimates are usually interpreted with measures of uncertainty and, where relevant, model-fit information. The registry record names maximum likelihood but does not report the assumptions used, estimation diagnostics, fit measures, or resulting parameters.

How should the planning illustration be read?

The stated planning assumptions describe a two-arm illustration using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect. These assumptions are not a description of the actual single-group, non-randomized study, and the illustration is not evidence of observed power. The registry record does not report a numerical observed effect or resulting planning values, so no specific planning requirement can be interpreted from this record.

In general, a confidence interval expresses uncertainty around an estimated quantity under the analysis assumptions. A narrow interval indicates greater precision than a wide interval, while neither interval width nor a statistical test resolves limitations in study design. No confidence interval or statistical test result for the primary endpoint is reported in the registry record.

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("Caregiver Child Dyad")
n   <- c(197)
m   <- c(-8.39)
sp  <- c(7.37)
sd  <- sp      # posted standard deviations
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  Caregiver Child Dyad  197  -8.390  7.370  -9.426  -7.354

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 8 outcome measures and names secondary outcomes that include unscheduled asthma-related service utilization visits, Pediatric Asthma Quality of Life Questionnaire change scores, and percent change in prescribed inhaler adherence. The primary endpoint is caregiver HRS-D change. Considering multiple outcomes can create a multiplicity issue: when many hypotheses are tested, the chance of at least one apparently notable result may increase. In general, an analysis plan can specify which outcomes are primary, how tests will be interpreted, and whether adjustment is planned. The registry record does not report a multiplicity strategy or the results of tests across outcomes.

The registry lists zero analyses. It does not report results by analysis population, the number contributing data to each outcome, or how missing measurements were handled. Details such as the analysis population, assumptions about missingness, imputation, and sensitivity analyses are not reported in the registry record. Without those details, it is not possible to assess whether missing follow-up could affect the estimates or how the reported enrollment relates to the observations analyzed.

The longitudinal methods described in the brief summary may require data at multiple measurement times. In general, missing measurements can reduce information and can bias an estimate if the absence of data is related to participant characteristics or outcomes. The extent and consequences depend on the pattern of missingness and the analytic method. Neither the pattern nor the handling of missing data is reported in the registry record, so those issues cannot be evaluated.

12. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists the Caregiver Child Dyad group with 205 at risk for serious adverse events and 205 at risk for other adverse events. It records 0 affected for serious adverse events, 46 affected for other adverse events, and 0 deaths. These are registry counts, not comparative estimates. The record lists no serious adverse event terms.

The category of other adverse events should not be treated as equivalent to serious adverse events. The registry does not list the terms for those other events in the record, so their nature and clinical importance cannot be characterized from the posted figures. The counts also do not describe event timing, duration, severity within the other-event category, or how events were collected.

Because this was a single-group, non-randomized study, these figures do not show how event experience compares with a control group or establish that events were caused by a study-related treatment. The at-risk counts do not establish that every person contributed the same duration of safety follow-up. No event rate, risk difference, or comparative safety estimate is reported in the registry record.

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

The study is designed to examine whether caregiver depression improvement is related over time to child asthma outcomes. Its within-subject longitudinal approach can frame questions about change and temporal relationships among repeated measurements. The listed methods also describe an interest in whether adherence or child anxiety or depression may be intermediate pathways.

The posted results do not include numerical primary endpoint estimates or child outcome estimates. They therefore do not show whether caregiver depression scores improved, whether child asthma measures changed, or whether any change in one preceded change in another. The registry record also does not report model estimates for the proposed mediation pathways. No conclusion about the direction, size, or precision of these relationships can be drawn from the reported results.

The design has no randomized control group. As a result, even an association between changes over time would not by itself establish that caregiver treatment caused a change in child asthma. Changes in circumstances, care or other factors could coincide with changes in both caregiver and child outcomes. The study's stated rationale is to examine the impact of caregiver improvement, rather than to test antidepressant efficacy against a control treatment; the design distinction remains important when interpreting any reported association.

Enrollment of 205 and completion status do not provide the missing endpoint estimates, or information about the completeness of repeated measurements. The safety counts are descriptive and are not a comparative assessment. The registry record supports describing the study's aims, design and recorded safety counts, but it does not provide the outcome analyses needed to evaluate the central research questions.

14. Why This Trial Matters Statistically

Statistically, this study illustrates the distinction between a question about temporal association and a question about treatment effects. Repeated measurements can be useful for examining whether changes in caregiver symptoms and child outcomes occur in an ordered pattern. Cross-lagged models and SEM offer ways to represent such relationships, but a model cannot remove the design limitations of a non-randomized, single-group study.

The design also highlights how a study's primary endpoint may differ from its broader scientific motivation. The registered primary endpoint measures caregiver depression, while the broader aim concerns child asthma outcomes and possible pathways linking caregiver and child change. Clear reporting of estimates for both caregiver and child measures would be needed to judge how closely the observed data address that broader aim. Such estimates are not reported in the registry record.

Multiplicity and missing-data reporting matter in a repeated-measures study with several outcomes. Readers need to know which analyses were conducted, how incomplete follow-up was handled, and how multiple outcomes were interpreted. The registry record lists outcome measures but does not report these analytic details or results. This limits assessment of both statistical precision and the robustness of any potential findings.

Finally, planning assumptions should be distinguished from the design that actually generated the data. A two-arm planning illustration does not convert a single-group study into a randomized comparison, and planning calculations are not observed power. Here, the record gives assumptions for an illustration but no numerical result to interpret. These distinctions help keep conclusions aligned with what the design and registry reporting can support.

15. 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 registry record does not report numerical results for the primary endpoint or child asthma outcomes. It lists zero analyses, so the described longitudinal and mediation methods cannot be assessed against model estimates, uncertainty measures or diagnostics. The central questions remain unanswered in the posted results.

The study is non-randomized, non-placebo controlled and single-group. Caregivers may choose whether to receive antidepressant medication, so treatment receipt is not randomly assigned. Changes over time may reflect factors other than caregiver depression improvement, and no concurrent control group is available to distinguish those influences. The record does not report an analysis designed to address those potential sources of confounding.

The registry lists enrollment of 205 but does not report how many participants contributed data to each outcome, completed follow-up, or were included in analyses. The schedule of repeated measurements, handling of missing data, and analysis populations are not reported in the registry record. The description of the primary endpoint timeframe also refers to first documented progression, but practical details of that timing are not reported.

Several outcomes are listed, but the record does not report how multiplicity was handled, which secondary analyses were performed, or estimates for the proposed mediators. Safety reporting is limited to aggregate counts, without listed serious event terms or details about other events. These limitations constrain evaluation of the magnitude, precision, completeness and interpretation of the study's findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Treating Caregiver Depression to Improve Childhood Asthma: Impact and Mediators — NCT02809677.
• Registry record last updated 2025-09-24; status last verified 2025-09.
• Results first posted 2022-02-15.
• 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.

19. Record Summary

The registry record describes NCT02809677, Treating Caregiver Depression to Improve Childhood Asthma: Impact and Mediators, sponsored by the University of Texas Southwestern Medical Center. It is listed as completed, with a start date of 2016-06, primary completion of 2020-10, and enrollment of 205.

The study is a single-group, non-interventional, non-randomized longitudinal study, described as lasting 52 weeks. Its primary endpoint is change in caregiver depression on the Hamilton Rating Scale for Depression. The broader aims concern child asthma outcomes and possible relationships with adherence and child anxiety or depression. The brief summary describes cross-lagged panel modeling and SEM using maximum likelihood.

The registry lists 8 outcome measures and zero analyses. Numerical primary and child outcome results, analysis details and missing-data methods are not reported in the registry record. Safety counts list 0 serious adverse events, 46 affected for other adverse events, and 0 deaths in the Caregiver Child Dyad group.