Enhancing Self Regulation as a Strategy for Obesity Prevention in Head Start Preschoolers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Enhancing Self Regulation as a Strategy for Obesity Prevention in Head Start Preschoolers (NCT01398358), a randomized na study. The primary outcome is Change in Body Mass Index Z-score.

2. Clinical Question

The study’s stated rationale was that strengthening emotional and behavioral self-regulation might contribute to obesity prevention. Its research objectives proposed comparing POPS with usual Head Start exposure, and assessing whether POPS combined with IYS would produce the greatest improvement in health behaviors and adiposity indices, potentially through improved self-regulation. Those are the questions the study set out to investigate, not conclusions established by the posted results. In particular, the registry record does not report an analysis of mediation or results for the broader set of health behaviors and adiposity measures described in the objectives.

3. Trial Design

The registry record describes a randomized, parallel-group trial with three groups: usual Head Start exposure, POPS, and POPS combined with IYS. It lists allocation as randomized and masking as none. The sponsor was the University of Michigan. The record gives a start date of 2011-03, primary completion in 2015-05, and status as completed. The study concerns obesity prevention among children attending Head Start in Michigan.

In the POPS group, children received classroom lessons about nutrition and obesity prevention, delivered by an Extension Educator in collaboration with the classroom teacher; parents were invited to related classes. The combined group received this POPS program and IYS. The usual-exposure group remained in its regular Head Start classroom and received the educational interventions usual to that classroom. The posted descriptions do not give further detail about the frequency or duration of program sessions, fidelity procedures, or implementation differences between sites.

The registry lists enrollment of 697. Its brief summary describes the proposed study as involving 600 children and their parents, so the proposed number and the listed enrollment are distinct figures. The primary outcome was change in BMI z-score over 9 months. The posted secondary outcome was change in teacher-reported self-regulation over the same timeframe. The registry record lists two outcome measures, one posted secondary outcome, and no reported analyses. Details such as the randomization procedure, allocation concealment, original sample-size justification, and prespecified statistical analysis plan are not reported in the registry record.

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.

Before assignment: 417 declined to enroll; 36 ineligible

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Seven of the 697 were excluded from analysis because they were missing both pre- and post- intervention BMI z-score. It was possible for a child to be included in the final analysis population even if they had not completed the follow-up portion of the study because the mixed model statistical method used could include their pre-intervention data.

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 Body Mass Index Z-score, measured over 9 months.

The primary endpoint is change in child BMI z-score, measured from height and weight at enrollment and at the end of the study period. A z-score expresses a measurement relative to a reference distribution, allowing a child’s BMI to be considered in relation to age- and sex-based expectations rather than as a raw BMI value alone. The registry identifies the unit as a z-score and the timeframe as 9 months.

Change is used to describe how the measured outcome moved over the study period, while comparisons between groups address whether the average change differed across trial assignments. The posted mean changes were -0.01 for usual Head Start exposure, 0.0 for POPS, and -0.04 for POPS + IYS. These are group summaries; they do not describe every child’s trajectory or establish why any change occurred. The registry description says that seven of the 697 participants were excluded from the primary analysis because both pre- and post-intervention BMI z-scores were missing. It also says a child could be included even without completing follow-up because of the mixed-model statistical method. Further details of that method and of the analysis population are not reported in the registry record.

BMI z-score is an adiposity-related measure, but it is not the same as a direct measure of every aspect of body composition or health. The study objectives included other health behaviors and adiposity indices; their results are not reported in the registry record.

7. Results Posted in the Registry

• Parents of Preschoolers Series (POPS): mean 0.00 versus -0.01 on control; difference 0.01 (95% CI -0.07 to 0.09), standardized effect d = 0.02; z-test p = 0.814.
• POPS + Incredible Years Series (IYS): mean -0.04 versus -0.01 on control; difference -0.03 (95% CI -0.11 to 0.05), standardized effect d = -0.07; z-test p = 0.480.

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.

Change in Teacher-reported Self-regulation as Measured by the General Adaptation T-score From the Social Competence and Behavior Evaluation Scale

• 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 errors.
• Result:
  • Parents of Preschoolers Series (POPS): mean 1.28 versus 0.93 on control; difference 0.35 (95% CI -1.19 to 1.89), standardized effect d = 0.04; z-test p = 0.656.
  • POPS + Incredible Years Series (IYS): mean 3.92 versus 0.93 on control; difference 2.99 (95% CI 1.53 to 4.45), standardized effect d = 0.37; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Parents of Preschoolers Series (POPS) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for POPS + Incredible Years Series (IYS) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcome is teacher-reported change in self-regulation, measured with the General Adaptation T-score from the Social Competence and Behavior Evaluation Scale. Across the three groups, the broad pattern is that the POPS + IYS group has a larger posted mean change than either usual Head Start exposure or POPS alone. The posted POPS-versus-usual comparison is much less distinct. This pattern differs from the primary BMI z-score comparisons, which do not show a clear separation between either intervention group and usual exposure in the unadjusted estimates.

That contrast in patterns is relevant to the study’s rationale: the combined program was intended to address self-regulation as well as obesity-related behaviors. But a secondary outcome cannot substitute for the primary endpoint, and the self-regulation result does not demonstrate that a change in self-regulation caused, or mediated, a change in BMI z-score. The registry record does not report a formal mediation analysis. It also does not report results for the health behaviors and other adiposity indices named in the study objectives.

Unadjusted secondary results generally carry less weight than a prespecified primary comparison because many outcomes or contrasts can create more opportunities for a chance finding. General statistical principle: when multiple hypotheses are tested, the probability of at least one apparently notable result can increase unless the analysis accounts for that multiplicity. The registry record does not report whether a correction or hierarchy was planned. The secondary result should therefore be viewed as a posted finding that may inform further study, rather than conclusive evidence about the full intervention pathway. Its teacher-reported nature is also distinct from a direct measurement of BMI z-score and may reflect the information available to teachers in the classroom.

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

• Change in Teacher-reported Self-regulation as Measured by the General Adaptation T-score From the Social Competence and Behavior Evaluation Scale

9. Statistical Interpretation

10. Statistical Methods Explained

The registry lists the primary outcome as a mean of a continuous measure: change in BMI z-score. A continuous outcome can take many values along a scale rather than being limited to event counts. The group summaries are means, accompanied by standard errors. A mean describes the center of the observed group values; a standard error describes uncertainty in estimating that mean, rather than the spread of individual children’s measurements. General statistical principle: standard errors tend to be smaller when estimates are more precise, but they do not show the full range of individual outcomes.

What does a between-group mean difference represent?

A mean difference compares the average outcome in one group with the average in a reference group. For the primary outcome, the reference is usual Head Start exposure. A positive difference means the intervention-group mean is numerically higher than the usual-exposure mean; a negative difference means it is numerically lower. Interpretation still depends on the outcome’s direction and on the uncertainty around the estimate. The posted primary comparisons are POPS versus usual exposure and POPS + IYS versus usual exposure. They are unadjusted recalculations from the posted summary statistics. They should not be described as adjusted effects or as estimates from a fully documented trial model.

How should a p-value be read?

General statistical explanation: a p-value describes how unusual data at least as inconsistent with a specified null hypothesis would be if that null hypothesis and the analysis assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of an effect. The posted primary p-values are 0.814 and 0.48. They are unadjusted, and the registry record does not report a multiplicity procedure. The secondary comparison for POPS + IYS reports p < 0.001; that result also needs to be interpreted in light of its secondary status and the unreported multiplicity plan.

What does an interval estimate add?

General statistical explanation: an interval estimate represents uncertainty around an estimated effect, with narrower intervals indicating greater precision under the method used. If an interval includes zero for a difference, values on both sides of zero remain compatible with the reported interval. The primary comparison intervals extend on both sides of zero. The registry record does not identify the confidence level or provide a full account of how these intervals were produced, so their coverage properties cannot be verified from the record alone.

What does randomization contribute?

Random assignment is intended to make trial groups comparable, on average, at the time of assignment, reducing confounding by factors that influence both group membership and outcome. It does not guarantee identical groups, eliminate missing-data concerns, or make every subsequent comparison unbiased regardless of analysis choices. The registry identifies the allocation as randomized, but the specific randomization sequence and allocation-concealment procedures are not reported in the registry record. Masking is listed as none. The record does not describe whether outcome assessors or analysts were masked.

Why can a mixed model include incomplete follow-up?

General statistical explanation: mixed models can use available repeated measurements rather than requiring every participant to have a complete set, under assumptions about the data and the model. This can be useful when some follow-up measurements are missing. The registry description says that a child could remain in the final analysis population despite not completing follow-up because a mixed-model statistical method was used. It also says seven children were excluded because both their pre- and post-intervention BMI z-scores were missing. The specific model, assumptions, covariates, handling of missingness, and sensitivity analyses are not reported in the registry record. The description alone does not establish that missing data could not bias results.

Why does over-dispersion matter?

General statistical explanation: over-dispersion arises when the variability in count data is greater than a model assumes. If ignored, uncertainty estimates can be too small and statistical evidence can look stronger than it is. The primary endpoint is a continuous mean, not a count outcome, and the registry record does not identify over-dispersion as an issue. It is therefore not a reason to reinterpret the posted BMI z-score comparisons. More generally, matching the statistical model to the outcome and its variation matters; the registry does not provide enough model detail to assess that choice fully.

The registry lists no reported analyses. No original statistical plan, interim analyses, or sample-size justification is reported in the registry record. Consequently, the posted summaries support descriptive comparisons and simple unadjusted recalculations, but not a reconstruction of all planned or sponsor-conducted analyses. Any broader inference should retain that distinction.

11. Confidence Intervals and Precision

Precision concerns how tightly an estimated group difference is pinned down by the data and the analysis. The posted primary comparison for POPS versus usual exposure is 0.01, with an interval from -0.073 to 0.093. For POPS + IYS versus usual exposure, the difference is -0.03, with an interval from -0.113 to 0.053. Each interval includes zero and also allows values on either side of zero. Thus the point estimates alone do not capture the range of effects compatible with the posted intervals.

The intervals are relatively broad in relation to the small mean changes, so the primary estimates do not precisely distinguish modest differences from no difference. This is a statement about statistical precision, not proof that the groups are identical. Conversely, an estimate’s position on one side of zero is not by itself evidence of a reliable difference when the interval includes zero. The unadjusted p-values, 0.814 for POPS and 0.48 for POPS + IYS, align with the uncertainty reflected in these intervals.

The registry record does not report the confidence level for these intervals or the full analysis procedure used to generate them. The comparisons are unadjusted recalculations from posted summaries, so the intervals should not be assumed to incorporate covariate adjustment, clustering, or other features of a prespecified trial analysis. These limitations matter when interpreting how precisely the trial estimates the primary contrasts.

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 planning figures are participants per arm, not total events. The registry’s stated illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers detecting the stated fraction of the observed effect: 100%, 75%, or 50%. The trial’s own original design assumptions and sample-size justification are not reported in the registry record; these figures are planning illustrations based on the observed effects, not a reconstruction of the original design.

The 80% power column gives the number of participants per arm that the illustration estimates would be needed for an 80% chance of detecting the specified effect under its assumptions, if that effect were present. The 90% power column gives the corresponding per-arm planning number for a 90% chance. Power is a property of a proposed study under specified assumptions, not a measure of whether a completed trial found a real effect. These figures are not observed power.

Smaller effects require more participants to distinguish them from random variation under the same assumptions. For example, the planning rows for a smaller fraction of an observed effect require more participants per arm than the rows for the full observed effect. The estimates also depend on the observed effect used as their starting point. Because an observed effect can be uncertain, especially when the estimated group difference is small, these illustrations should not be read as definitive sample-size recommendations. They do not establish what the trial originally planned, and they do not replace a design calculation based on a prespecified clinically or scientifically meaningful effect and an appropriate analysis model.

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 = 0.010; pooled standard deviation σ = 0.443.
2. n = 2 × (1.96 + 0.842)² × 0.443² / 0.010² = 30,873.7, rounded up to 30,874 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 <- 0.01                      # observed difference in means
sigma <- 0.443481523931906                     # 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(1398358)
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 = 30,874 (80% power), 41,332 (90% power)
 75% of observed effect: n per arm = 54,887 (80% power), 73,478 (90% power)
 50% of observed effect: n per arm = 123,495 (80% power), 165,325 (90% power)
power.t.test: 30874.7 
Simulated power at n = 30,874 per arm: 0.800 (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. Multiplicity, Missing Data and Analysis Populations

The trial has a three-group design, and the posted primary comparisons assess each intervention group against usual Head Start exposure. The record also lists a secondary self-regulation outcome and broader objectives involving health behaviors and adiposity indices. The registry record reports no analyses and does not describe a multiplicity adjustment, a hierarchy among outcomes, or which comparisons were prespecified. This makes it difficult to determine how the posted comparisons fit within the complete set of analyses considered by the investigators.

General statistical explanation: testing multiple outcomes or contrasts can increase the chance of observing a low p-value even when none of the underlying effects is real. Adjustments or prespecified testing strategies can address this issue, but whether they were used is not reported in the registry record. The unadjusted p-values should be read as individual comparisons rather than as evidence that the family of outcomes was controlled for multiplicity.

For the primary BMI z-score outcome, the record says seven of the 697 participants were excluded because both pre- and post-intervention scores were missing. It also says that children could be included without completing follow-up because a mixed-model method was used. The specific model and its missing-data assumptions are not reported in the registry record. The record does not give the number missing at each time point, reasons for missingness, or sensitivity analyses that test whether missing data could change the results. The primary arm counts are 216, 221, and 253, while the adverse-event groups have different at-risk counts; these are reported for distinct purposes and should not be treated as interchangeable analysis populations.

The registry record does not report the full analysis population definition, covariate adjustments, clustering procedures, or an intention-to-treat analysis. It also does not report how protocol deviations or incomplete intervention participation were handled. These omissions limit assessment of whether the posted estimates reflect the originally assigned groups and whether the analysis adequately addresses missing outcomes.

15. 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 registry’s adverse-event record lists zero serious adverse events, zero other adverse events, and zero deaths in each of the three groups. The listed numbers at risk are 218 for usual Head Start exposure, 224 for POPS, and 255 for POPS + IYS. These denominators differ from the primary outcome analysis counts, because the adverse-event figures refer to the groups listed in the safety record.

These entries describe what was reported in the registry record; they do not prove that no event occurred outside the recorded definitions, monitoring period, or reporting process. No serious adverse-event terms are listed. The registry record does not provide further detail about event collection, event definitions, severity grading, or the period over which events were monitored. A zero count should therefore be read as a registry report, not as a general guarantee of safety or as a complete account of every possible adverse experience.

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

The posted primary results describe average change in BMI z-score over 9 months for children assigned to usual Head Start exposure, POPS, or POPS + IYS. The unadjusted intervention-versus-usual differences are small in the posted estimates, and both intervals include zero. This means that the registry results do not clearly distinguish either intervention group from usual exposure on this primary outcome. It does not establish that the groups are exactly alike, because the intervals also allow differences in either direction.

The results do not establish that the interventions reduced obesity, changed health behaviors, or produced a broader benefit. The BMI z-score endpoint is one measure, and results for other health behaviors and adiposity indices mentioned in the objectives are not reported in the registry record. The secondary self-regulation pattern is not proof of mediation, and the record does not report the analysis needed to test that pathway. The posted self-regulation outcome is teacher-reported and should be understood in that context.

The comparisons are unadjusted recalculations from posted summary statistics. The registry lists no reported analyses, and details of the original analysis plan, confidence level for the intervals, missing-data procedures, and multiplicity strategy are not reported. Those gaps limit what can be concluded from the posted estimates. In particular, a non-small p-value is not proof of no effect, just as a smaller p-value on a secondary measure would not establish an effect on the primary outcome.

These findings are statistical descriptions of the registry record, not medical advice. They do not determine how the programs should be used in other settings or whether a particular child or family should receive an intervention. The record also does not report enough implementation detail to judge how closely the programs were delivered across classrooms. Any conclusions should remain specific to the outcomes and comparisons actually posted.

17. Why This Trial Matters Statistically

This trial is statistically informative because it compares two related behavioral approaches with usual Head Start exposure, while distinguishing the POPS curriculum from POPS combined with a program focused on self-regulation. That three-group structure connects a primary adiposity-related outcome with a proposed behavioral pathway. The registry’s broad objectives make clear why both BMI z-score and teacher-reported self-regulation were of interest.

The posted results also illustrate why a point estimate should not be read alone. The primary mean differences are small, but their intervals span zero and values on either side. The secondary outcome has a different pattern, yet its unadjusted status and the absence of a reported mediation analysis constrain what it can support. In this setting, precision, prespecification, and the relationship between primary and secondary outcomes are as important as whether an individual p-value crosses a conventional threshold.

The record additionally highlights the importance of transparent reporting. It lists enrollment and outcome summaries, but does not report the original sample-size justification, detailed statistical plan, missing-data sensitivity analyses, or multiplicity procedure. Without these details, readers cannot fully evaluate how the posted comparisons relate to the intended analysis. The record’s statement that a mixed model could include children without completed follow-up is useful context, but does not by itself reveal the model’s assumptions or robustness.

For biostatisticians and clinical researchers, the central lesson is to separate descriptive summaries, unadjusted recalculations, and a prespecified trial analysis. Those are not interchangeable forms of evidence. The registry results permit a cautious reading of the posted estimates, while leaving important design and analysis questions unresolved.

18. Limitations

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

Several trial-specific limitations affect interpretation. The registry identifies a randomized design but does not report how the random sequence was generated, whether allocation was concealed, or whether any outcome assessors were masked. Masking is listed as none. The record does not provide enough information to assess possible effects of unmasked delivery or assessment on the teacher-reported secondary outcome.

The record lists no reported analyses and gives no original sample-size justification or full statistical analysis plan. It therefore does not establish which comparisons were prespecified, how the investigators planned to handle multiplicity, or whether the posted unadjusted comparisons match the primary model used by the sponsor. The confidence level for the posted intervals is also not reported. As a result, the recalculated comparisons should be read as limited summaries rather than a complete account of trial inference.

Missing data are another concern. Seven children were excluded from the primary analysis because both pre- and post-intervention BMI z-scores were missing, while the record says a mixed-model method allowed some children without completed follow-up to remain in the analysis. The detailed model, assumptions about missingness, amount and pattern of missing data, and sensitivity analyses are not reported. The posted analysis counts and safety denominators differ, so those counts represent different reporting contexts.

The reported outcomes cover only part of the study’s stated objectives. Results for obesity-related health behaviors and other adiposity indices are not reported in the registry record, and no mediation analysis is reported. Further, the interventions were delivered in Head Start classrooms with educators and teachers, but implementation and fidelity results are not posted in the outcome information. These limitations restrict both the breadth of conclusions and assessment of how the programs were carried out.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Enhancing Self Regulation as a Strategy for Obesity Prevention in Head Start Preschoolers — NCT01398358.
• Registry record last updated 2018-02-05; status last verified 2017-07.
• Results first posted 2018-02-05.
• 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.

22. Record Summary

The registry record describes a completed randomized, parallel-group study sponsored by the University of Michigan, with enrollment of 697 and primary completion in 2015-05. It compares usual Head Start exposure with POPS and POPS + IYS. The primary outcome is change in BMI z-score over 9 months; the posted secondary outcome is teacher-reported change in self-regulation.

The unadjusted primary comparisons do not clearly distinguish either intervention group from usual exposure. The secondary outcome has a different pattern, but no mediation analysis or multiplicity plan is reported. Seven participants were excluded from the primary analysis because both BMI z-scores were missing; the registry also describes use of a mixed-model method that could include children without completed follow-up. The record lists no reported analyses and does not report the original sample-size justification or detailed analysis plan. Safety entries list zero serious events, other events, and deaths in all groups.