Shared Decision Making in Parents of Children With Head Trauma: Head CT Choice: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Shared Decision Making in Parents of Children With Head Trauma: Head CT Choice (NCT02063087), a randomized na study. The primary outcome is Assess Parents' Knowledge Regarding Their Child's Risk for a Significant Brain Injury.

2. Clinical Question

The trial asked whether giving parents a structured Head CT decision aid during an emergency department encounter could change what they understood about their child’s risk of significant brain injury and the choices around head CT and observation. It also assessed whether decision-making experiences differed between families offered the aid and those receiving usual care.

3. Trial Design

The registry record describes a randomized, parallel-group trial with single masking. It lists 971 participants enrolled and identifies Mayo Clinic as the sponsor. The record gives a start date of 2014-04, a primary completion date of 2016-10, and a completed status. The phase is listed as not applicable.

Participants were assigned to the Decision Aid or Usual Care group. In the decision-aid group, the intervention was the Head CT Decision Aid. In usual care, clinicians and patients did not have access to that aid. The registry record does not identify which people involved in the trial were masked, so the meaning of the single-masking label cannot be specified further.

The primary outcome was a continuous knowledge score assessed on Day 1, immediately after the clinical encounter. Its group summaries include 467 participants in the Decision Aid group and 468 in the Usual Care group. The outcome description says the primary analysis was limited to participants with complete data; the record does not describe how missing responses were handled beyond that statement.

The registry lists additional outcomes concerning engagement in decision-making, decisional conflict, trust, satisfaction, CT use, tests ordered within 7 days, clinically important traumatic brain injury, and fidelity to care options. Numerical group summaries are available for several decision-making measures. The registry record does not report the original trial sample-size justification, a detailed statistical analysis plan, or interim-analysis procedures. Those details cannot be inferred from the completed status or from the posted group summaries.

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.

Emergency Department Intervention

7 Day Follow-up

Analysis population for the primary outcome, as defined in the registry: Analysis limited to participants with complete 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 Assess Parents' Knowledge Regarding Their Child's Risk for a Significant Brain Injury, measured over Day 1 (immediately after the clinical encounter).

The primary endpoint measures parents’ knowledge about their child’s risk for a significant brain injury immediately after the emergency department encounter. It is scored as the number of correct answers out of 10. The questions cover the child’s quantitative risk, the pros and cons of head CT compared with active observation, and symptoms to watch for during the next 24-48 hours that could prompt a return to the emergency department.

Each question offers True, False, or Unsure as response options. A correct response receives a score of 1; an incorrect or Unsure response receives a score of 0. The outcome description says the overall score is calculated by summing correct responses and dividing by the number of questions asked. The posted results summarize this continuous measure with group means and standard deviations.

General statistical context: a knowledge score provides a way to compare the groups’ average performance on the same set of questions. It does not, on its own, show whether parents later made a particular choice, whether care was safer, or whether knowledge persisted beyond the assessment time. Those questions require outcomes and follow-up information beyond an immediate post-encounter score.

7. Results Posted in the Registry

• Decision Aid: mean 6.20 versus 5.26 on control; difference 0.94 (95% CI 0.68 to 1.20), standardized effect d = 0.47; z-test p = <0.001.

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.

Patient Engagement in the Decision-making Process

• 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:
  • Decision Aid: mean 25.00 versus 13.30 on control; difference 11.70 (95% CI 10.39 to 13.01), standardized effect d = 1.54; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Decision Aid than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Decisional Conflict

• 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:
  • Decision Aid: mean 14.80 versus 19.20 on control; difference -4.40 (95% CI -6.47 to -2.33), standardized effect d = -0.27; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Decision Aid than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Trust in the Physician

• 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:
  • Decision Aid: mean 91.50 versus 89.30 on control; difference 2.20 (95% CI 0.56 to 3.84), standardized effect d = 0.17; z-test p = 0.008.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Decision Aid than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes broaden the question beyond parents’ immediate knowledge. They include engagement, decisional conflict, trust, satisfaction, use of head CT, tests ordered within 7 days, clinically important traumatic brain injury, and fidelity to care options. The posted numerical summaries for engagement, decisional conflict, and trust concern related aspects of the encounter, but they are not interchangeable measures.

In broad terms, these outcomes address whether the decision process felt participatory, whether parents experienced uncertainty, and how they viewed the physician. The primary knowledge score addresses what parents answered correctly after the encounter. A pattern across these measures can help describe several dimensions of the decision-making experience, but it does not demonstrate that one measure caused another or that changes translated into later healthcare use or clinical outcomes.

The registry record reports unadjusted p-values for the posted secondary comparisons and does not report a multiplicity adjustment. General statistical context: when several outcomes are examined, individual unadjusted tests do not account for the possibility of chance findings across the full set. This makes secondary results less conclusive than a single prespecified primary comparison, especially when the analysis plan and outcome hierarchy are not reported in detail.

Secondary outcomes also differ in their measurement scales and observation windows. A scale-based measure collected on Day 1 captures an immediate experience, while utilization or injury outcomes address different events and time frames. Readers should consider each outcome on its own terms and avoid treating a consistent direction across measures as proof of a unified underlying effect. The primary result is specifically about immediate knowledge; the other measures provide context, not a replacement for that endpoint.

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

• Patient Engagement in the Decision-making Process
• Decisional Conflict
• Trust in the Physician
• Parental Satisfaction
• Proportion of Children Who Undergo Head CT
• Healthcare Utilization - Number of Tests Ordered Within 7 Days
• Rate of Clinically Important Traumatic Brain Injury (ciTBI)
• Fidelity - Options for Care

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: it is a score summarizing the number of correct knowledge responses. The registry reports a mean and standard deviation for each group. A mean describes the group’s average score; a standard deviation describes how spread out individual scores are around that average. For the Decision Aid group, the mean was 6.2 and the standard deviation was 2.0. For Usual Care, the mean was 5.26 and the standard deviation was 2.0.

The reported comparison is a mean difference, with Usual Care as the reference and Decision Aid as the other group. The estimated difference is 0.94. A positive value means the Decision Aid group’s posted mean is higher. The confidence interval for this contrast has a positive lower bound and an upper bound of 1.196, so the range does not cross zero. The unadjusted p-value is <0.001.

What does a mean difference tell the reader?

A mean difference expresses the contrast in the original units of the outcome. Here, those units are knowledge-score points. This makes the result easier to relate to the score than a standardized measure would be. It also means the interpretation depends on the questions, scoring approach, and how meaningful a score-point change is to families. The registry describes the questions and scoring rules but does not report a threshold for a meaningful difference.

General statistical context: a confidence interval is a range produced by a statistical procedure that, under its assumptions, reflects uncertainty in an estimated effect. Its width depends on factors such as the amount of information and variation in the outcome. A confidence interval is not a range containing most individual participants’ scores, and it does not account for every possible source of bias or uncertainty.

Why does the analysis population matter?

The registry says the primary analysis was limited to participants with complete data. The group summaries include 467 participants assigned to Decision Aid and 468 assigned to Usual Care. That is a complete-data comparison, rather than an analysis whose membership is described as including every randomized participant regardless of whether an outcome was observed. The record does not report how many participants lacked primary outcome data, why data were missing, or whether a missing-data method was used.

General statistical context: excluding incomplete observations can affect a comparison if the people with missing data differ from those whose outcomes are observed. The direction and size of any such effect cannot be determined from the group means alone. A clear accounting of missing outcomes and the analysis approach helps readers judge how representative the analyzed groups may be.

What does an unadjusted comparison imply?

The posted primary contrast is unadjusted. It compares the group summaries without a reported adjustment for other variables. The record does not provide a detailed statistical model, covariate list, or analysis plan. Random assignment is the design feature intended to make groups comparable on average, but the registry information does not establish that every possible source of imbalance or bias was absent.

The registry lists several secondary outcomes, and the posted summaries include engagement, decisional conflict, and trust. Each additional outcome can raise a separate statistical question. If many outcomes are tested, the chance of at least one apparently notable result due to random variation can increase. The record does not report a multiplicity-adjustment procedure, so the secondary p-values should be read as unadjusted evidence for their individual comparisons, not as a family of results with a controlled overall error rate.

General statistical context: over-dispersion refers to variability greater than a statistical model expects. It is relevant when a model’s assumptions about variability do not fit the observed data, because uncertainty estimates can then be too small if the extra variability is ignored. The primary outcome is reported as a continuous score with a standard deviation, and the registry does not describe a model or an over-dispersion assessment. There is therefore no basis in the registry record to say that such a model or correction was used.

The registry record does not report interim analyses, an original sample-size justification, or a detailed prespecified statistical analysis plan. The absence of those details in the record should not be replaced with assumptions about what the investigators planned. The planning illustrations associated with the posted effects use stated assumptions, but they are not the trial’s original design justification.

11. Confidence Intervals and Precision

The primary mean difference is 0.94 points, with a confidence interval extending from a positive lower bound to 1.196. Because the interval does not include zero, the posted unadjusted comparison is compatible with a positive difference in mean knowledge scores for the Decision Aid group under the analysis used. The interval also shows that the estimate is not a single exact value: a range of effects remains compatible with the data and assumptions.

The interval is expressed in knowledge-score points, the same units as the outcome. This supports direct interpretation of the contrast, but the registry record does not report a threshold for a difference that would be considered important to parents or clinicians. Statistical precision and practical importance are separate questions. A result can be estimated with a relatively narrow interval without establishing what size of difference matters in decision-making.

General statistical context: confidence intervals quantify sampling uncertainty for a specified estimate and method. They do not automatically include uncertainty from incomplete data, measurement limitations, deviations from the intended intervention, or selective outcome reporting. The primary analysis’s complete-data restriction is relevant to that broader assessment, and the registry does not provide enough information to quantify any effect of missing data.

The secondary comparisons also have confidence intervals, but each concerns a different measure and should be interpreted in that measure’s own units. Intervals and p-values do not correct for testing several outcomes unless an adjustment is specified. The registry record does not report such an adjustment. Precision should therefore be considered alongside the number of outcomes, the analysis population, and the limits of what was prespecified, which are not fully described in the record.

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 event counts. The accompanying assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. These assumptions describe the planning illustration, not the trial’s original sample-size design, which is not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding number for a 90% chance. Greater target power generally requires more participants for the same effect. For a smaller target effect, the required number increases; the rows for 75% and 50% of the observed effect illustrate that pattern.

These figures are planning illustrations based on the observed effect, not observed power. They should not be read as the number needed to reproduce the trial, nor as evidence that the original trial was planned using these assumptions. The figures are conditional on the model assumptions and chosen target effect. For continuous outcomes such as the knowledge score, the displayed values refer to participants per arm; the event-count convention is not relevant to this outcome.

The sample-size curve should be read as a visual expression of the same planning relationship: smaller effects require more participants to attain the same target power, while larger effects require fewer. A curve cannot resolve whether an effect is clinically meaningful, whether the observed effect is a stable estimate for future planning, or whether the original design had different assumptions. Those original design details are not reported in the registry record.

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 = 0.940; pooled standard deviation σ = 2.000.
2. n = 2 × (1.96 + 0.842)² × 2.000² / 0.940² = 71.1, rounded up to 72 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 <- 0.94                      # observed difference in means
sigma <- 2                     # 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(2063087)
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 = 72 (80% power), 96 (90% power)
 75% of observed effect: n per arm = 127 (80% power), 170 (90% power)
 50% of observed effect: n per arm = 285 (80% power), 381 (90% power)
power.t.test: 72 
Simulated power at n = 72 per arm: 0.802 (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 registry record lists a primary knowledge outcome and multiple secondary outcomes. It does not report a detailed statistical analysis plan describing outcome hierarchy, prespecified contrasts, or multiplicity control. The posted p-values are unadjusted. Without a reported adjustment, each secondary p-value addresses its own comparison and should not be interpreted as controlling the chance of a false-positive finding across all outcomes considered together.

The primary outcome population is explicitly limited to participants with complete data. The record does not report the number of people with missing primary scores, the reasons for missingness, or how partial responses were handled. It also does not describe an imputation method, sensitivity analysis, or other missing-data procedure. For the secondary outcomes, analysis-population details are not reported in the registry record.

These omissions limit what can be assessed about selection into each analysis. If outcome availability is related to group assignment, participant characteristics, or the unobserved score, a complete-data comparison may not represent the full randomized groups. The registry summaries alone do not show whether such a pattern occurred or how much it could affect the estimates.

The record does not report interim analyses, stopping boundaries, or a detailed account of the original analysis plan. It also does not report the trial’s original sample-size justification. These details matter when assessing how outcomes and analyses were selected, but their absence from the registry record does not establish that no procedures existed. They remain unknown based on the registry record.

General statistical context: multiplicity is the challenge of interpreting results when many hypotheses are tested. Missing-data methods address incomplete outcomes but rely on assumptions about why data are missing. Reporting the planned approach, the amount and pattern of missingness, and sensitivity analyses helps readers understand both sources of uncertainty. Such information is not reported in the registry record for this trial.

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 reports zero serious adverse events in the Decision Aid group among 493 participants at risk and zero in the Usual Care group among 478 participants at risk. It also reports zero other adverse events in each group and zero deaths in each group. No serious adverse-event terms are listed.

These figures describe the events recorded in the registry. They do not, by themselves, establish that no events occurred outside the registry’s reporting, or that the two approaches have equivalent safety. The record does not provide additional detail about event definitions, collection procedures, follow-up for adverse events, or how attribution was assessed. The at-risk counts used for these safety summaries are not the same as the complete-data counts listed for the primary knowledge outcome.

General statistical context: when no events are recorded, the data offer limited information about the range of event rates that could still be compatible with the observations. A zero count is a report about the observed registry data, not proof that the underlying risk is zero. Safety interpretation also depends on duration and completeness of monitoring, which are not described in detail in the registry record.

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

The posted primary results show a higher average immediate knowledge score among parents in the Decision Aid group than among parents in Usual Care. The unadjusted mean difference is 0.94 points, with a confidence interval whose lower bound is positive and upper bound is 1.196. The unadjusted p-value is <0.001. These results describe the measured score among participants with complete data and support a difference between the posted group means under the analysis used.

They do not establish how meaningful a difference of this size is for families. The registry does not report a threshold for important change in knowledge, and a score collected immediately after the encounter does not show how much information parents retained later. The trial’s stated aims also included decision-making engagement and healthcare utilization, but the primary endpoint itself measures knowledge rather than use of CT, later testing, or injury outcomes.

The design is randomized, which supports a comparison between assigned groups, but the record does not provide enough information to assess every feature of trial conduct or analysis. In particular, the primary analysis was limited to complete data, while details about missing outcomes, the original analysis plan, and multiplicity control are not reported in the registry record.

The findings should not be expanded to settings, populations, or outcomes beyond those described in the record. They do not show that the decision aid changed a particular parent’s decision, that it reduced healthcare utilization, or that it altered the rate of clinically important traumatic brain injury. Those are distinct outcomes requiring their own reported results and careful interpretation. The safety counts report zero recorded adverse events and deaths in the listed groups, but do not prove an absence of risk.

General statistical context: statistical evidence for a group difference is not the same as evidence of clinical importance, persistence, or a causal pathway through other outcomes. The most direct conclusion is limited to the immediate knowledge scores reported for the two randomized groups.

17. Why This Trial Matters Statistically

This trial illustrates how a clinical intervention can be evaluated through an outcome that is not a conventional laboratory measure or disease event. The primary endpoint is a parent knowledge score, while the secondary outcomes address participation in decisions, decisional conflict, and trust. Making the endpoint explicit helps readers distinguish an immediate educational or decision-process result from later care use or clinical outcomes.

The comparison also shows why analysis populations matter. The primary outcome is reported for participants with complete data, whereas the safety summaries use different at-risk group counts. These denominators answer different questions. A single enrollment total does not imply that every outcome was observed for every participant, and understanding each denominator is part of interpreting the estimate.

Confidence intervals add useful context to a mean difference by showing uncertainty around the estimated contrast. In this trial, the primary interval stays on the positive side of zero, while its interpretation still depends on measurement quality, completeness, and practical meaning. The result illustrates that statistical evidence and importance to patients are related but separate considerations.

The secondary outcomes also demonstrate the interpretive challenge created by multiple measures. A collection of unadjusted comparisons can describe a broad set of experiences, but does not have the same confirmatory force as a single prespecified primary comparison with a clearly reported analysis plan. The registry record does not report multiplicity procedures or the original sample-size justification, so those uncertainties should remain visible when reading the findings.

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.

The primary analysis includes participants with complete data, but the registry record does not report how many primary outcomes were missing, why they were missing, or whether missingness differed by group. This limits assessment of whether the analyzed participants represent everyone assigned to the trial. The record also does not describe the handling of partially completed questionnaires.

The endpoint is an immediate post-encounter knowledge score. The registry describes its subject matter and scoring, but does not report evidence about the score’s validity, reliability, or threshold for a meaningful difference. The immediate timing also means the result cannot establish longer-term retention.

The registry record does not report the original sample-size justification, detailed statistical analysis plan, interim analyses, or multiplicity adjustment. The posted primary and secondary comparisons are unadjusted calculations from the group summaries, and no sponsor inferential analysis is reported in the registry record. These limitations make it harder to assess whether the analyses follow a prespecified hierarchy or how the investigators planned to interpret multiple outcomes.

The record labels masking as single but does not identify who was masked. It also does not give enough detail to assess intervention fidelity, the conduct of usual care across settings, or potential variation in how the decision aid was used. The outcome list includes satisfaction, CT use, testing, clinically important traumatic brain injury, and fidelity, but the registry information described for this comparison does not provide enough numerical detail to assess those outcomes alongside the primary estimate.

The safety summaries show zero recorded events in the listed categories, but event definitions and monitoring details are not reported. These counts therefore should be read as registry reports rather than as complete evidence that risk was absent.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Shared Decision Making in Parents of Children With Head Trauma: Head CT Choice — NCT02063087.
• Registry record last updated 2019-05-09; status last verified 2019-05.
• Results first posted 2019-05-09.
• 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

NCT02063087 is a completed randomized, parallel-group trial sponsored by Mayo Clinic. It compared the Head CT Decision Aid with usual care for parents of children presenting to the emergency department with blunt head trauma. The registry lists 971 enrolled participants and a primary completion date of 2016-10.

The primary outcome was parents’ knowledge immediately after the encounter, scored as correct answers out of 10. The Decision Aid group mean was 6.2 and the Usual Care group mean was 5.26; the unadjusted mean difference was 0.94, with p < 0.001. The primary analysis was limited to participants with complete data.

The registry also lists decision-making and healthcare outcomes. Numerical summaries are reported for several decision-making measures, while details of the original statistical plan, missing-data handling, and multiplicity procedures are not reported in the registry record. Safety summaries list zero serious adverse events, other adverse events, and deaths in both groups.