Promotion of Successful Parenting: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Promotion of Successful Parenting (NCT03808987), a randomized na study. The primary outcome is Parental Sensitivity (Measured With the Parental Bonding Questionnaire).

2. Clinical Question

The trial’s stated rationale is that home visitation by community health workers is used to support families, while some families may face trauma, depression, or other challenges for which more intensive intervention may be considered. The evaluation therefore examines whether adding Child-Parent Psychotherapy to community health worker outreach is associated with differences in positive parenting and parent-child relationships. The comparison is not simply between receiving and not receiving home visitation: every described arm includes community health worker services, while the arms differ in whether and when Child-Parent Psychotherapy is added.

The posted primary results concern parental sensitivity at the child’s age of 9 months. The registry record also describes assessments of parenting, maternal sensitivity, representational models, cognitions, physiological reactivity, and physical health indicators. The results available for interpretation include the primary outcome and two secondary outcome entries at the child’s age of 15 months. The record does not report results for every assessment described in the study summary, nor does it provide a complete account of analyses addressing how or for whom the intervention may be effective.

3. Trial Design

The registry record describes a randomized, single-masked, parallel-design study of Promotion of Successful Parenting, officially titled TRANSFORM: Translational Research That Adapts New Science FOR Maltreatment Prevention. The recorded enrollment is 222, and the trial status is completed. The record lists the start date as 2019-02-21 and the primary completion date as 2024-12-30. The phase is listed as NA. The sponsor is the University of Rochester.

Four intervention groups are described. The CHW+CPP brief prenatal-onset group received Child-Parent Psychotherapy for 6 months beginning prenatally, in addition to community health worker services. The CHW+CPP brief postnatal-onset group received Child-Parent Psychotherapy for 6 months beginning postnatally, also in addition to community health worker services. The CHW+CPP 12-month group received Child-Parent Psychotherapy for 12 months beginning prenatally, alongside community health worker services. The CHW-only group received community health worker services without Child-Parent Psychotherapy.

The record calls the study randomized and single-masked, but the allocation ratio and the identity of the masked parties are not reported in the registry record. The recruitment setting, eligibility details beyond the listed population, and procedures for assigning participants to the four groups are not reported in the registry record. The enrollment figure describes the trial record; the posted outcome analyses use smaller, outcome-specific group sizes, consistent with the record’s note about attrition or missing data at the third data-collection wave. The original design assumptions, including any planned sample-size justification, are not reported in the registry record.

The study description places assessments prenatally and at children’s ages of 3, 9, and 12 months. The posted primary result uses the child’s age of 9 months, while the secondary outcome entries use the child’s age of 15 months. The registry record does not explain the relationship between those stated assessment schedules and the 15-month results in detail. It also does not report the timing of every intervention relative to each posted outcome assessment beyond the arm descriptions.

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: The number analyzed on the Postpartum Boding Questionnaire at baby age 9 months differs from the number recruited because of attrition or missing data at our third wave of data collection

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 Parental Sensitivity (Measured With the Parental Bonding Questionnaire), measured over Assessment at child's age of 9 months.

The primary endpoint is the mean score on the Parental Bonding Questionnaire at the child’s age of 9 months. The registry description says the measure captures improvements in maternal reports of positive emotional responses to babies. It includes 25 items, with each item ranging from 0-5, from never to always. The total score is the sum of item scores and ranges from 0-125; higher scores indicate greater bonding or a stronger mother-reported emotional bond with her baby.

A continuous score retains information about the measured level of the construct rather than reducing responses to a simple event count or yes/no classification. Reporting a group mean describes the average score among those analyzed, while the standard deviation describes how spread out participant scores are around that mean. These are general interpretations of continuous outcomes. The score is a measure of reported bonding as defined by the questionnaire, not a direct measure of every aspect of parenting or of child development.

The registry record identifies the number analyzed as different from the number recruited because of attrition or missing data at the third wave of data collection. The numbers analyzed for the primary outcome are reported by group, but reasons for missingness, the amount of missing outcome data for each participant, and the method used to address missing data are not reported in the registry record. The record also does not report a clinically important difference for this scale or a threshold for interpreting a particular score. Thus, the result describes questionnaire scores and their statistical uncertainty; it does not by itself define the practical importance of a score difference.

7. Results Posted in the Registry

• CHW+CPP Brief Postnatal Onset: mean 116.40 versus 120.38 on control; difference -3.98 (95% CI -6.18 to -1.78), standardized effect d = -0.80; z-test p = <0.001.
• CHW+CPP 12 Months: mean 118.70 versus 120.38 on control; difference -1.68 (95% CI -3.92 to 0.56), standardized effect d = -0.34; z-test p = 0.141.
• CHW Only: mean 118.65 versus 120.38 on control; difference -1.73 (95% CI -3.66 to 0.20), standardized effect d = -0.37; z-test p = 0.079.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Parental Sensitivity (Measured With the Parental Bonding Questionnaire)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • CHW+CPP Brief Postnatal Onset: mean 116.46 versus 118.80 on control; difference -2.34 (95% CI -5.20 to 0.52), standardized effect d = -0.39; z-test p = 0.109.
  • CHW+CPP 12 Months: mean 116.86 versus 118.80 on control; difference -1.94 (95% CI -5.18 to 1.30), standardized effect d = -0.28; z-test p = 0.240.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CHW+CPP Brief Postnatal Onset and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CHW+CPP 12 Months and the control. The interval gives the range of effects that remain compatible with the data.

Parental Acceptance (Measured With the Parental Acceptance and Rejection Questionnaire)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • CHW+CPP Brief Postnatal Onset: mean 68.03 versus 64.03 on control; difference 4.00 (95% CI 0.30 to 7.70), standardized effect d = 0.49; z-test p = 0.034.
  • CHW+CPP 12 Months: mean 64.09 versus 64.03 on control; difference 0.06 (95% CI -3.02 to 3.14), standardized effect d = 0.01; z-test p = 0.970.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for CHW+CPP Brief Postnatal Onset than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between CHW+CPP 12 Months and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results extend the outcome picture to the child’s age of 15 months and include parental sensitivity and parental acceptance. Taken together, they do not present a consistent pattern of strong evidence across the outcomes and comparisons. The sponsor-reported overall analyses have p-values of .47 for parental sensitivity and .10 for parental acceptance. One unadjusted pairwise recalculation for parental acceptance is below .05, while the other listed pairwise comparisons are not. These results should be read as exploratory context alongside the primary outcome, not as independent confirmation of it.

The secondary sensitivity measure is related by name and construct to the primary parental-sensitivity measure, but it is reported at a different follow-up time. Results at the child’s age of 15 months do not directly replace or resolve the primary result at the child’s age of 9 months. The acceptance measure is a different outcome, so its result addresses a related but distinct aspect of parenting. A difference in one measure does not establish that the same pattern holds for the other.

Unadjusted secondary results carry less weight because they arise among multiple outcomes and comparisons. When many statistical tests are considered, some may appear unusual by chance alone even when there is no underlying effect. An adjustment procedure, prespecified hierarchy, or other multiplicity strategy can help contextualize that risk, but none is reported in the registry record. The record also does not report whether the secondary analyses were prespecified or whether they were considered exploratory.

The secondary results should therefore be understood as descriptive and hypothesis-generating unless supported by a clearly reported analysis plan and appropriate treatment of multiplicity. They do not establish a broad pattern across parenting outcomes, and they do not change the fact that the primary sponsor-reported analysis is the principal result for the stated 9-month endpoint. Missing-data methods and analysis populations are also not described in enough detail to determine how those issues affect the secondary findings.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Parental Sensitivity (Measured With the Parental Bonding Questionnaire)
• Parental Acceptance (Measured With the Parental Acceptance and Rejection Questionnaire)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry record reports a continuous primary outcome summarized as a mean and standard deviation. It also reports an ANOVA across the four groups. The recalculated pairwise comparisons use posted summary statistics and are explicitly unadjusted. Those different summaries answer related but nonidentical statistical questions, so their results should not be treated as interchangeable.

What does a group mean describe?

A mean is the arithmetic average of the observed scores in a group. For this outcome, the group means describe the average Parental Bonding Questionnaire score among participants included in each group’s analysis at the child’s age of 9 months. A mean is a useful summary of the central level of a continuous measure, but it does not show whether all participants had similar scores. The standard deviation provides information about spread: a larger standard deviation indicates more variability around the group mean. These are general statistical explanations.

At the primary time point, the posted group means are 120.4 for brief prenatal onset, 116.4 for brief postnatal onset, 118.7 for the 12-month course, and 118.7 for CHW only. The corresponding standard deviations are 3.458, 5.817, 5.734, and 5.188. The score’s possible range is 0-125, and the record defines higher values as greater reported bonding. The meaning of a mean difference depends on that scale definition and on the population and time point being compared.

What does the ANOVA address?

ANOVA is a method for assessing whether group means, considered together, are consistent with a common mean or show evidence of variation among groups. The registry record reports an ANOVA F-value of 3.41 and p = .019 for the primary outcome. The F-value is a test statistic, and the p-value summarizes how compatible the observed test result is with the null model used for that analysis. The record labels the analysis as an equivalence type, but it does not provide an equivalence margin or further details of the analysis. The record does not report the full model specification, covariates, degrees of freedom, or whether the analysis was adjusted for baseline measurements.

An omnibus ANOVA does not identify which specific group pairs account for an overall pattern. That requires separate comparisons or a planned contrast approach. The registry record’s reported result is the sponsor’s own analysis, and it takes precedence because the record flags a discrepancy with the recalculated pairwise results. The details needed to reproduce or explain that discrepancy are not reported in the registry record.

What do the recalculated mean differences represent?

The pairwise recalculations compare each listed group with the brief prenatal-onset group. A mean difference expresses the separation between the two group means in the questionnaire’s score units. A negative estimate for a comparison group indicates a lower mean in that group than in the brief prenatal-onset group; a positive estimate indicates a higher mean. The recalculations are simple, unadjusted comparisons based on the posted summary statistics, not adjusted estimates from a participant-level model. They do not account for any covariates or for the full four-group analysis unless the record explicitly says so; it does not.

The posted pooled standard deviation is used to describe the spread for a pairwise comparison, and the standardized effect size expresses the mean difference relative to that pooled spread. Standardization can help describe the size of a contrast in relation to variability, but it does not establish clinical importance. The record does not report a threshold for a meaningful difference on the parental-sensitivity measure. Some recalculated statistics are shown with different precision in the posted calculations; values with more than three decimal places are not repeated in this explanation.

How should p-values be read?

A p-value is a measure of how unusual data at least as incompatible with a stated null model would be if that model and the test assumptions held. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. A p-value below a conventional threshold does not resolve concerns about multiplicity, missing data, model choice, or the relevance of the measured outcome. The registry record does not report the sponsor’s prespecified significance threshold or full statistical analysis plan.

The recalculated primary pairwise p-values are unadjusted. When multiple comparisons are examined, the chance of obtaining at least one small p-value can increase. Adjustment methods can address that issue, but the record does not report whether an adjustment was planned or applied. The sponsor’s ANOVA is an overall group analysis; the pairwise results are narrower and should not be treated as though they were the sponsor’s complete analysis.

Why does variability matter?

Variation affects how precisely a mean difference can be estimated. A difference between group means is more uncertain when scores are variable or when the analyzed groups contain fewer participants. Standard errors summarize sampling uncertainty under a statistical model; confidence intervals combine an estimate with its uncertainty to give a range of values compatible with the calculation. These are general explanations. The recalculated standard errors and intervals are based on the posted group summaries and do not account for unreported sources of uncertainty, such as missing-data assumptions or adjustments.

Over-dispersion is particularly important for outcomes analyzed as counts: if the observed variation exceeds what a simple count model assumes, standard errors may be too small and inference may be misleading unless the model accounts for the extra variability. The posted primary outcome is continuous and summarized with means and standard deviations, not an event count, so over-dispersion is not the central issue for this endpoint. This general point should not be mistaken for a claim about the trial’s analysis plan.

What analysis details remain uncertain?

The record does not report the original sample-size justification, interim analyses, detailed randomization procedures, the identity of masked parties, or a complete statistical analysis plan. It also does not report how missing outcome data were handled, whether the analysis followed an intention-to-treat population, or whether model assumptions were checked. These omissions limit independent assessment of how the sponsor’s result was generated and how robust it is to reasonable alternative analysis choices.

11. Confidence Intervals and Precision

The recalculated intervals describe uncertainty around the unadjusted pairwise mean differences. For the brief postnatal-onset group compared with the brief prenatal-onset group at the child’s age of 9 months, the estimated difference is -3.98, with an interval from -6.179 to -1.781. The interval is entirely below zero, consistent with a lower mean score in the postnatal-onset group in this pairwise calculation. Its p-value is < 0.001.

The recalculated intervals for the 12-month group and CHW-only group comparisons include zero, as do the intervals for both posted pairwise comparisons at the child’s age of 15 months. An interval that includes zero is compatible with no mean difference under that calculation, as well as with some differences in either direction. It does not prove that the groups are identical. The registry record does not report the confidence level used for the recalculated intervals, so a confidence percentage should not be inferred from the posted interval endpoints.

Precision is distinct from the point estimate. A point estimate gives the calculated difference, while an interval conveys how much uncertainty remains around it under the calculation’s assumptions. The intervals depend on the group sizes and variability, and they reflect only the unadjusted summary-statistic calculation. They do not incorporate uncertainty about missing data, the sponsor’s analysis model, or multiplicity unless those issues are explicitly built into the method; the registry record does not report that they are.

The intervals should be read alongside the sponsor-reported omnibus ANOVA, not as a substitute for it. The record identifies a discrepancy, so the sponsor’s own analysis takes precedence. The sponsor-reported ANOVA result does not include a confidence interval. The registry record also does not report a clinically important difference threshold for the scale. As a result, the intervals speak to statistical precision in score units, but not by themselves to whether a difference would matter in practice.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size planning figures are participants per arm, not total participants or total events. The calculations use two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. For each comparison, they use the observed effect or a stated fraction of that observed effect. The original trial’s own design assumptions and sample-size justification are not reported in the registry record, so these figures are planning illustrations based on the posted results rather than a reconstruction of the trial’s design.

The 80% power column gives the estimated participants per arm for a future two-arm study to have an 80% probability of detecting the specified effect under the stated assumptions, if that effect is real and the model assumptions hold. The 90% power column expresses the corresponding planning target for a 90% probability. These are design calculations, not guarantees: actual power depends on the true effect, variability, missing data, analysis method, and other features of a future study.

The rows for 100%, 75%, and 50% of the observed effect illustrate how planning changes when the target effect is set at different fractions of the observed pairwise estimate. A smaller target effect requires more participants per arm to distinguish it from random variation at the same significance level and power target. The figures are calculated separately for the listed comparisons, so they should not be treated as a single recommended sample size for a future study. They are not observed power for this completed trial.

Some recalculated effects are small relative to the observed variability, which leads to very large planning figures when the calculation targets only a fraction of the observed effect. Such outputs are sensitive to the estimate used as a planning input, especially when that estimate is uncertain. The figures do not account for multiplicity, attrition, clustering, covariate adjustment, or any design feature beyond the assumptions stated in the registry record’s planning description. They should therefore be read as conditional illustrations rather than as a complete sample-size proposal.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -3.980; pooled standard deviation σ = 4.969.
2. n = 2 × (1.96 + 0.842)² × 4.969² / -3.980² = 24.5, rounded up to 25 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -3.97999999999999                      # observed difference in means
sigma <- 4.96927794839031                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(3808987)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 25 (80% power), 33 (90% power)
 75% of observed effect: n per arm = 44 (80% power), 59 (90% power)
 50% of observed effect: n per arm = 98 (80% power), 132 (90% power)
power.t.test: 25.5 
Simulated power at n = 25 per arm: 0.792 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor reports an ANOVA for the four groups at the primary time point, with F-value 3.41 and p = .019. The registry record flags a discrepancy between that reported analysis and the posted recalculations. Under the interpretation rules for these results, the sponsor’s own analysis takes precedence over the simple unadjusted recalculations.

The recalculations compare the brief prenatal-onset group separately with each of the other groups using the posted means and standard deviations. They are not a reproduction of the sponsor’s full analysis and do not include information beyond the posted summary statistics. One pairwise recalculation has an interval excluding zero, while the other primary pairwise intervals include zero. The reported ANOVA is an overall test across the groups and does not identify which pairwise contrasts account for its result. Differences in analysis methods, assumptions, or data handling could produce different results, but the registry record does not explain the discrepancy, so its cause should not be guessed.

The record does not provide the sponsor’s full model specification, adjusted estimates, confidence intervals for the ANOVA, or enough detail to verify the analysis independently. The recalculated values are useful as a transparent, limited reading of the posted summaries, but they should not be presented as the sponsor’s findings or as a replacement for them. The discrepancy also makes it especially important to distinguish the sponsor-reported overall result from the recalculated pairwise comparisons in any interpretation.

15. Multiplicity, Missing Data and Analysis Populations

The record reports several outcomes and more than one group comparison. The primary outcome is parental sensitivity at the child’s age of 9 months. The posted secondary outcomes include parental sensitivity and parental acceptance at the child’s age of 15 months. The record also describes a broader range of assessments, but does not provide results for every measure mentioned in the study summary. Considering multiple outcomes, time points, and contrasts creates a multiplicity concern: the chance of seeing at least one apparently notable result can rise as more tests are considered.

The registry record does not report a multiplicity adjustment, a testing hierarchy, or whether the pairwise comparisons were prespecified. It also does not report a complete statistical analysis plan. The reported overall ANOVA and the recalculated pairwise comparisons should therefore be distinguished, and the unadjusted secondary comparisons should be treated cautiously. Without information about planned contrasts and adjustment, it is not possible to determine how the sponsor intended to control the overall risk of false-positive findings.

Missing outcome data are a further concern. The primary outcome description says the number analyzed differs from the number recruited because of attrition or missing data at the third wave. The posted group sizes show that analysis counts vary across outcomes and follow-up times. The reasons for missingness, whether participants who left the study differed from those retained, and how missing measurements were handled are not reported in the registry record. These details matter because a complete-case comparison can be biased if the availability of an outcome is related to the outcome itself or to group assignment.

The analysis population is also not fully specified. The record does not say whether analyses followed intention-to-treat principles, whether participants were analyzed according to randomized assignment, or whether any additional exclusions were applied. The randomization procedure and allocation concealment details are not reported in the registry record. Without these details, the effect of missingness and analysis choices on the reported estimates cannot be assessed fully. These are limitations in interpretation, not evidence that the reported results are necessarily biased.

16. 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 adverse-event record lists zero serious adverse events and zero other adverse events for each of the four groups. It also lists zero deaths in every group. The numbers at risk are reported separately for each event category and group: 46 for the brief prenatal-onset group, 56 for brief postnatal onset, 57 for the 12-month group, and 63 for CHW only. These figures are the registry’s safety denominators; they are not the same as the outcome-analysis group sizes reported for parental sensitivity.

A zero count means that no event in the listed category was recorded in the registry results for the corresponding group. It does not establish that an intervention has no risks, nor does it describe events that were not captured or reported. The registry record lists no serious adverse-event terms. It does not provide additional detail about how events were solicited, defined, monitored, or adjudicated, or about the observation period used for the safety summary.

These safety figures should be read as a report of recorded events, not as proof of safety or as a comparison with a broader population. The record does not describe a statistical safety analysis or provide precision estimates for the event counts.

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

The posted results describe group averages on specified questionnaire measures at particular follow-up times. At the primary time point, group means range from 116.4 to 120.4 on a scale with a stated range of 0-125, where higher scores indicate greater reported bonding. The sponsor’s reported overall ANOVA has p = .019, and the registry record flags a discrepancy with the unadjusted recalculations. The sponsor’s own analysis takes precedence, but the available information does not identify which group differences account for its overall result.

The results do not by themselves establish whether Child-Parent Psychotherapy works or does not work. Statistical evidence about score differences is not the same as evidence about clinical importance. The registry record does not report a threshold for a meaningful difference on the questionnaire, and it does not link a particular score change to a specific change in parenting or child outcomes. A group mean also does not describe every participant’s experience.

The findings do not establish which timing or duration is best. The comparisons are limited to the listed groups, outcomes, and assessment times. The record does not report a complete analysis of how or for whom the intervention may be effective, despite naming those as study questions. It also does not provide enough information about missing data, analysis populations, multiplicity, or the sponsor’s full model to settle how robust the findings are to alternative analysis choices.

Finally, the reported absence of recorded adverse events should not be interpreted as proof that no risks exist. It is a summary of the registry’s event counts and denominators, without detailed methods for event monitoring or adjudication. The most appropriate reading is therefore bounded: these are posted statistical results for questionnaire outcomes in the analyzed participants, with important design and analysis details not reported in the registry record.

18. Why This Trial Matters Statistically

This trial raises statistical questions that are common in evaluations of complex behavioral interventions. The arms vary not only in whether Child-Parent Psychotherapy is added to community health worker services, but also in its timing and duration. That structure creates several possible comparisons, while the record also describes a range of parenting and health assessments. A clear distinction between the primary outcome, secondary outcomes, and exploratory questions is important when interpreting multiple results.

The difference between an overall group test and separate pairwise comparisons is central to reading the posted findings. The sponsor reports an ANOVA across the four groups, while the recalculations compare the prenatal-onset group with each other group using summary statistics. Because the record flags a discrepancy, those calculations should not be allowed to displace the sponsor’s analysis. The record’s limited explanation of the discrepancy illustrates why transparent reporting of models, estimands, and analysis populations matters.

Follow-up results also depend on who remains available for measurement. The primary outcome description explicitly notes attrition or missing data at the third wave, and analyzed group sizes differ by outcome. Reporting participant flow, missingness patterns, and the assumptions used to address missing data would help readers assess the reliability of the estimates.

Finally, a sample-size illustration based on an observed effect is not a substitute for the original design justification. Planning figures depend on assumptions about effect size, variability, allocation, and the target operating characteristics. Showing those assumptions makes the calculations interpretable, while keeping them separate from observed power avoids overstating what the completed trial established. Together, these issues show why statistical interpretation depends on more than a p-value: design transparency, precision, missing-data handling, multiplicity, and the distinction between statistical and practical importance all matter.

19. 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 trial-specific limitation is the discrepancy between the sponsor’s reported primary ANOVA and the recalculated pairwise results. The sponsor’s analysis takes precedence, but the registry record does not explain the discrepancy or provide the detailed model information needed to reconcile the results. The pairwise calculations are unadjusted, so they should be read as limited comparisons rather than a complete account of the sponsor’s analysis.

The primary outcome analysis includes fewer participants than the recorded enrollment, and the registry record attributes the difference to attrition or missing data at the third wave. Group sizes also vary across the posted outcome entries. Reasons for missingness, participant flow by group, and the missing-data method are not reported in the registry record. Those omissions limit assessment of possible selection effects and the precision of the outcome estimates.

The record does not report the original sample-size justification, allocation ratio, randomization procedures, identity of masked parties, or complete statistical analysis plan. It does not identify whether the reported analyses were adjusted for baseline scores or other covariates, nor does it provide a multiplicity strategy. Confidence levels for the recalculated intervals are not reported. The record also does not report a clinically important difference threshold for the questionnaire scales.

The study description names additional assessments and questions about how and for whom the intervention is effective, but the posted results summarized in the record do not resolve those questions. The 15-month outcome entries are later than the 9-month primary endpoint, and their relationship to the assessment schedule described in the summary is not explained in detail. Safety reporting gives event counts but limited information about event definitions and monitoring. These limitations restrict interpretation; they do not establish that the results are invalid.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Promotion of Successful Parenting — NCT03808987.
• Registry record last updated 2026-05-14; status last verified 2026-03.
• Results first posted 2026-05-14.
• 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.

23. Record Summary

The completed randomized, single-masked, parallel trial compares community health worker services alone with three approaches that add Child-Parent Psychotherapy: brief prenatal onset, brief postnatal onset, and a 12-month prenatal-onset course. The recorded enrollment is 222.

The primary outcome is mean parental-sensitivity score on the Parental Bonding Questionnaire at the child’s age of 9 months. The sponsor reports an overall ANOVA with F-value 3.41 and p = .019. The registry record flags a discrepancy with unadjusted pairwise recalculations, so the sponsor’s own analysis takes precedence. The posted secondary outcomes assess parental sensitivity and parental acceptance at the child’s age of 15 months.

Interpretation is limited by unreported details about missing-data handling, multiplicity, analysis populations, and the full statistical plan. The registry lists zero serious adverse events, zero other adverse events, and zero deaths in each group. These are reported registry counts, not proof that no risk exists.