Helping Improve Pediatric Patient Outcomes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Helping Improve Pediatric Patient Outcomes (NCT03471221), a randomized na study. The primary outcome is Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Total Distress Score.

2. Clinical Question

The randomized comparison focuses on the immediate assessment after the first visit, rather than a longer-term average across repeated visits. The intervention could be offered repeatedly during hospitalization, but the primary endpoint timeframe was immediately after the first visit. This distinction matters: the primary result describes the groups at that specified time, not the accumulated effect of receiving visits over several weeks.

The record also describes a second, observational activity intended to measure microbial levels before a dog visit, after the visit but before hand cleaning, and after hand cleaning. Whether that activity was conducted and its results are not reported in the registry record. The reported randomized comparison therefore addresses the trial's stated immediate outcome question, while the additional microbial sampling question cannot be evaluated from the reported results.

3. Trial Design

This was a randomized, double-masked, parallel-group trial sponsored by Kaiser Permanente. The registry lists an enrollment of 26 and a completed status. Participants were assigned to therapy dog visits or a control group receiving usual medical care. The intervention was behavioral. The registry record does not report the method used to generate the random sequence, how allocation was concealed, or which individuals were masked.

Participants assigned to the intervention could receive a visit from a therapy dog and handler team up to one time per week for up to four weeks. The schedule depended on the length of hospitalization and the therapy dog team's capacity. A visit could last up to about 20 minutes. The control group received usual medical care. The number of visits actually received by each participant is not reported in the registry record.

The primary outcome was assessed immediately after the first visit and was designated as an intention-to-treat analysis. The recorded assessment populations were 11 in the therapy-dog group and 12 in the control group. The registry notes that two of 14 participants randomized to control did not have a first visit and one of 12 randomized to the intervention group did not complete the survey. The record does not provide further detail about the handling of missing observations.

The study description sets out two distinct activities: a randomized trial and a proposed observational microbial-sampling activity after the main trial aims. The observational activity's conduct and findings are not reported in the registry record. The original sample-size justification and the trial's full statistical analysis plan are also 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Two out of 14 patients randomized to the control arm did not have a visit 1; one out of 12 patients randomized to the intervention group did not complete the survey.

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.

Population: Randomized participants and their parents (note that the Enrolled Number is 26, as parents are not randomized or enrolled)

6. Primary Endpoint

The registry lists the primary outcome as Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Total Distress Score, measured over Immediately after first visit (Intention to Treat).

The primary endpoint was the Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales Total Distress Score, measured immediately after the first visit. It is intended to capture present or in-the-moment distress, rather than distress recalled over a longer period. The total is computed as the average of six scales: afraid/scared, sad/blue, angry, worry, tired, and pain/hurt. Each scale ranges from 0-100, and lower overall scores indicate lower total distress.

A brief, present-focused measure is suited to a question about the immediate experience surrounding a visit. Averaging the six scales creates an overall distress measure while drawing on several kinds of unpleasant feelings or symptoms. The individual scales and emotional distress summary are also listed among the trial's outcomes, so the total score is one part of a wider set of psychosocial measures rather than a complete description of every response.

The endpoint is a continuous score, and the reported group summaries are means and standard deviations. The primary comparison concerns the difference in mean scores between the randomized groups at the specified assessment time. The scale's direction is important when reading that difference: lower values correspond to lower distress. The result does not, by itself, describe how participants felt at other times or establish whether any difference persisted beyond the immediate post-visit assessment.

7. Results Posted in the Registry

• Therapy Dog Visits: mean 11.00 versus 13.60 on control; difference -2.60 (95% CI -12.94 to 7.74), standardized effect d = -0.21; z-test p = 0.622.

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.

Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Total Distress Score

• 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:
  • Therapy Dog Visits: mean 11.00 versus 13.60 on control; difference -2.60 (95% CI -12.94 to 7.74), standardized effect d = -0.21; z-test p = 0.622.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Therapy Dog Visits and the control. The interval gives the range of effects that remain compatible with the data.

Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Emotional Distress Summary Score.

• 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:
  • Therapy Dog Visits: mean 8.20 versus 8.00 on control; difference 0.20 (95% CI -10.34 to 10.74), standardized effect d = 0.02; z-test p = 0.970.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Therapy Dog Visits and the control. The interval gives the range of effects that remain compatible with the data.

Microbial Load (Colony Forming Units; Cfu) on Hands That Touched the Dog Versus Hands of Medically Eligible Controls

• 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:
  • Therapy Dog Visits: mean 2.74 versus 3.39 on control; difference -0.65 (95% CI -1.62 to 0.32), standardized effect d = -0.58; z-test p = 0.188.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Therapy Dog Visits and the control. The interval gives the range of effects that remain compatible with the data.

The reported secondary outcomes extend beyond the primary total distress score to emotional distress and microbial load, while the registry lists additional psychosocial and microbial measures. Across the detailed secondary findings, the estimated directions and sizes are not a single consistent pattern across all kinds of outcomes. These measures address different constructs and use different scales, so their estimates should not be combined into one overall effect.

The per-protocol total distress result relates closely to the primary result because it measures the same outcome at the same immediate post-visit timeframe, but it carries a different analysis label. The emotional distress summary is another psychosocial measure, while microbial load addresses a biological sampling question rather than a reported feeling or quality-of-life score. These secondary outcomes may provide context for the primary comparison, but they do not replace it or independently establish its interpretation.

Unadjusted secondary results deserve less weight because the trial considered multiple outcomes. When many comparisons are made, some estimates can appear unusual by chance, even if there is no underlying difference. Whether multiplicity was addressed through an adjustment or a prespecified hierarchy is not reported in the registry record. Secondary analyses can therefore be useful for describing patterns and generating questions, while remaining less conclusive than a clearly prespecified primary analysis.

The registry reports 11 secondary outcomes, but the analysis method, missing-data handling, and complete numerical findings for every measure are not reported in the registry record. The brief study description also refers to a separate observational microbial activity. Its conduct and findings are not reported in the registry record, so it cannot be used to extend the randomized comparison's conclusions.

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

• Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Total Distress Score
• Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Individual Scales (Afraid/Scared, Sad/Blue, Angry, Worry, Tired, Pain/Hurt)
• Pediatric Quality of Life Inventory Present Functioning Visual Analogue Scales (PedsQL VAS) Emotional Distress Summary Score.
• Number of Participants With the Presence of Clinically Important Organisms
• Pediatric Quality of Life Inventory (PedsQL) Cancer Module: Treatment Anxiety Dimension
• Patient Feelings When Thinking About Being in the Hospital (Afraid/Scared, Sad/Blue, Angry; Worried)
• The Positive and Negative Affect Schedule (PANAS): Parent Measure of Child's Positive Affect
• The Positive and Negative Affect Schedule (PANAS): Parent Measure of Child's Negative Affect

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each participant has a score on a numerical scale, and the registry reports a mean and standard deviation for each group. A difference in means expresses the average score in the intervention group relative to the average in the control group. In this comparison, a negative difference means the intervention group's mean score was lower. Since lower scores indicate lower distress, the direction of the point estimate is toward a lower mean distress score in the intervention group.

What does a confidence interval describe?

A confidence interval gives a range of values compatible with the observed data under the method used to calculate it. It conveys uncertainty around an estimate; it is not a range containing a fixed proportion of individual participants' scores. For the recalculated primary mean difference, the 95% confidence interval extends from -12.94 to 7.745. Because it includes zero, the interval is compatible with no mean difference as well as with differences in either direction. Its breadth is also relevant: the estimate is not precise enough to pin down a narrow range of possible group differences.

How should the p-value be read?

The unadjusted p-value for the recalculated primary comparison is 0.622. A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that hypothesis and the model assumptions were true. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The registry's reported primary analysis does not list a p-value; the stated p-value belongs to the unadjusted re-calculation.

Why distinguish the registry analysis from the re-calculation?

The registry reports a median difference for the primary outcome, whereas the simple re-calculation compares means using posted group summary statistics. Means and medians are different summaries. A mean uses the numerical values throughout the distribution and can be affected by unusually high or low observations. A median describes the middle of the ordered observations and is less affected by extreme values. The reported estimates need not be numerically identical because they do not target precisely the same summary. The method for the registry's reported analysis is not stated in the registry record.

The re-calculated mean comparison is unadjusted: it uses the group summaries without covariate adjustment. It should not be mistaken for a more fully specified model or for the trial's original analysis plan. The registry record does not report whether the original analysis adjusted for baseline measures or other covariates, how the distribution of scores was assessed, or how any departures from model assumptions were handled.

What do standard deviations and standard errors convey?

A standard deviation describes variation among participant scores within a group. The primary standard deviations were 12.5 in the therapy-dog group and 12.8 in the control group. A standard error instead describes the uncertainty in an estimate such as a group mean; it is used when forming an interval around an estimate. The posted re-calculation reports standard errors of 3.769 and 3.695 for the intervention and control means, respectively. They are not alternate descriptions of the spread of individual scores.

For continuous outcomes, unequal or highly variable observations can affect the precision of estimated group means. In general, a larger standard deviation means more variation in participant scores and, all else equal, less precision for a group mean. Small assessment groups can also leave estimates sensitive to a few observations. The registry record does not report diagnostic checks or details of any robust or alternative analysis.

How do the intention-to-treat and per-protocol labels differ?

The primary outcome is labeled intention to treat, and a secondary result for the same total distress score is labeled per protocol. In general, an intention-to-treat approach analyzes participants according to randomized assignment, while a per-protocol approach analyzes participants according to specified adherence or eligibility criteria. The exact per-protocol criteria and the handling of missing observations are not reported in the registry record. The posted group counts for the two total-distress summaries are the same, so the labels alone do not explain the analysis procedures used.

Three participants are described as lacking the first visit or survey: two assigned to control did not have a visit, and one assigned to the intervention group did not complete the survey. The exact reason for each missing outcome, any imputation, and any sensitivity analysis are not reported in the registry record. Missing outcome data can affect both the composition of the analyzed groups and the uncertainty of a result, particularly when the observed groups are small.

The study lists multiple psychosocial and microbial outcomes. Testing multiple outcomes can increase the chance that at least one result appears noteworthy by chance. Whether the trial used any adjustment or a prespecified hierarchy for multiple comparisons is not reported in the registry record. Thus, unadjusted secondary analyses should be read as exploratory evidence rather than as independent, definitive tests.

11. Confidence Intervals and Precision

The recalculated primary mean difference is -2.6, with a 95% confidence interval from -12.94 to 7.745. This interval is wide relative to the point estimate and includes zero. It therefore leaves substantial uncertainty about the direction and size of the mean difference. The data are compatible with lower distress scores in the therapy-dog group, but also with higher scores. The interval should not be read as proof that there is no difference simply because it crosses zero.

The registry's reported primary median difference is -3.0, with a 95% confidence interval from -12.4 to 6.4. This interval also includes zero. Its estimate is median-based, while the re-calculated interval concerns a difference in means, so the two intervals summarize different comparisons. The method used for the reported interval is not stated in the registry record.

Precision depends on the amount of information in the data and their variability. The primary assessment included 11 participants in the therapy-dog group and 12 in control, with standard deviations of 12.5 and 12.8. These small observed groups provide limited information for estimating a group difference, and the outcome summaries indicate variation among participants. The missing first-visit assessments noted in the registry further describe why the randomized counts and analyzed counts are not the same.

A confidence interval communicates statistical uncertainty, not whether a difference would matter to patients. The registry record does not specify a threshold for a clinically important difference on this endpoint. Consequently, the interval cannot be classified against such a threshold from the registry information alone. The posted estimate is best read as an uncertain result from a small comparison, not as a precise measurement of an effect.

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 figures are planning illustrations based on the observed effect, not the trial's original design assumptions. For the continuous outcomes in this record, the figures are participants per arm; the time-to-event convention of total events does not apply to these continuous comparisons. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario targets a specified fraction of the observed effect.

The 80% power column gives the number of participants per arm that the planning calculation estimates would be needed for an 80% chance of detecting the assumed effect under those assumptions, if the effect and model assumptions were correct. The 90% power column gives the corresponding per-arm count for a 90% chance. These are planning quantities, not a measure of the probability that the observed trial result is true.

The rows planning for 100%, 75%, and 50% of the observed effect illustrate how the assumed target changes the required enrollment. Smaller effects are harder to distinguish from random variation, so planning to detect a smaller effect requires more participants. The figures are highly dependent on the observed effect in this small trial; they should not be treated as a definitive recommendation for a future study. The trial's own original sample-size justification is not reported in the registry record. These figures are not observed power.

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 = -2.600; pooled standard deviation σ = 12.658.
2. n = 2 × (1.96 + 0.842)² × 12.658² / -2.600² = 372.1, rounded up to 373 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 <- -2.6                      # observed difference in means
sigma <- 12.6580296367845                     # 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(3471221)
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 = 373 (80% power), 499 (90% power)
 75% of observed effect: n per arm = 662 (80% power), 886 (90% power)
 50% of observed effect: n per arm = 1,489 (80% power), 1,993 (90% power)
power.t.test: 373 
Simulated power at n = 373 per arm: 0.795 (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 registry's reported primary result is a median difference of -3.0, with a 95% confidence interval from -12.4 to 6.4. The unadjusted re-calculation based on posted group means is a mean difference of -2.6, with a 95% confidence interval from -12.94 to 7.745 and an unadjusted p-value of 0.622. Both confidence intervals include zero, but the estimates are not identical because the reported analysis and the re-calculation summarize the outcome differently.

The registry record does not name the method used for its reported median difference and does not give a p-value for that analysis. The re-calculation uses the posted group means, standard deviations, and sample sizes to form an unadjusted mean comparison. It is a supplementary check based on summary statistics, not a reconstruction of the trial's full statistical analysis plan. Details of that plan are not reported in the registry record.

The same distinction applies to the per-protocol result for total distress: the registry reports a mean difference of -3.0, with a 95% confidence interval from -12.4 to 6.4, while the posted summary statistics yield an unadjusted re-calculated mean difference of -2.6. The analysis populations have the same reported group counts, but the criteria defining the per-protocol population are not reported in the registry record. The registry's own reported estimates should be kept distinct from the simple re-calculations rather than treated as interchangeable outputs.

15. Multiplicity, Missing Data and Analysis Populations

The trial has a primary distress outcome and a broader set of secondary measures, including multiple distress scales, treatment anxiety, affect measures, and microbial outcomes. Examining many outcomes raises a multiplicity concern: even when no true differences exist, repeated testing can make at least one result appear noteworthy by chance. The registry record does not report a multiplicity adjustment, a hierarchy for testing secondary outcomes, or which secondary analyses were designated confirmatory.

That absence of reporting does not establish that no adjustment was made; it means the registry record does not describe one. The unadjusted p-values from re-calculations should therefore not be treated as though they account for the number of outcomes assessed. Secondary findings are most appropriately read as supporting context or exploratory evidence unless further analysis-plan details are available.

Missing assessments are relevant to the primary comparison. The record states that two of 14 participants randomized to control did not have a first visit and one of 12 randomized to the intervention group did not complete the survey. The analyzed counts were 12 for control and 11 for intervention. The reason each observation was missing, whether outcomes were imputed, and whether sensitivity analyses were conducted are not reported in the registry record.

The primary outcome is labeled intention to treat, but the registry record does not explain the operational handling of participants without an observed outcome. A per-protocol analysis is also listed for the same total distress score, but its criteria are not reported. Without those details, the labels alone cannot establish how missing data or protocol deviations were managed. The original statistical analysis plan, including model specifications, covariate adjustments, and interim analyses, is not reported in the registry record.

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 registry lists zero serious adverse events affecting participants in the therapy-dog group among 12 at risk and zero among 14 at risk in the control group. It also lists zero other adverse events affecting participants in both groups. These counts describe what was recorded during the trial, rather than proving that no adverse events of any kind could occur.

The registry separately lists two deaths in the therapy-dog group and zero in the control group. The causal relationship of these deaths to the intervention is not reported in the registry record. The counts alone do not establish that the deaths were caused by, or unrelated to, therapy dog visits. The registry does not list serious adverse-event terms for either group.

Safety figures should be read with the small group sizes and the limits of event reporting in mind. A count of zero in a small study cannot rule out uncommon events, and the record does not provide narrative details about the deaths or other safety monitoring. These figures are descriptive registry entries, not a complete causal assessment of safety.

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

The posted primary results show a small observed difference in mean distress scores, with the therapy-dog group having a lower mean score than the control group at the immediate assessment. The confidence interval is wide and includes zero, leaving uncertainty about whether the underlying mean difference is lower or higher in the intervention group. The registry's median-based result also has a confidence interval that includes zero. Together, the results do not precisely establish the direction or magnitude of a group difference.

This does not mean the groups have been shown to be identical. An imprecise estimate can be compatible with several possible differences. Nor does a point estimate in one direction establish a reliable effect. The trial's small observed assessment groups and missing assessments limit the amount of information available for drawing a firm conclusion.

The result concerns an immediate PedsQL VAS distress assessment after the first visit. It does not establish what participants experienced after repeated visits, whether any difference persisted, or whether other psychosocial outcomes changed in the same way. The trial's wider outcome set includes measures that address distinct questions, and secondary comparisons should be interpreted with caution because multiplicity handling is not reported.

The registry also records safety counts, including deaths for which causal relationship is not reported. Those entries do not by themselves determine whether an event was caused by the intervention. In addition, the planned observational microbial activity's conduct and results are not reported in the registry record. The randomized microbial comparison and the separate proposed sampling sequence therefore should not be conflated.

Overall, the registry record supports a description of the observed groups and their uncertainty, not a broad claim about effects in other hospitals or populations. The original sample-size justification, full statistical plan, and several analysis details are not reported, which further limits how far the results can be interpreted.

18. Why This Trial Matters Statistically

This trial illustrates why an estimate should be read together with its uncertainty. The primary point estimate is in the direction of lower distress scores in the therapy-dog group, but the confidence interval includes zero and spans possible differences in either direction. A single point estimate cannot convey that uncertainty by itself.

It also illustrates that the reported analysis and a simple recalculation can summarize the same outcome in different ways. The registry's primary analysis is expressed as a median difference, while the recalculation uses a mean difference derived from group summaries. Naming the estimand and the analysis population is essential when comparing results; apparently similar numbers do not make different summaries interchangeable.

The small analyzed groups are an important statistical feature. Limited data can make estimates imprecise and sensitive to individual observations. Missing assessments further distinguish the randomized group sizes from the numbers contributing to the primary outcome. The registry's intention-to-treat label does not, on its own, explain how those missing outcomes were handled.

The trial also has multiple secondary measures, which raises the need to distinguish confirmatory questions from exploratory ones. Without a reported multiplicity plan, secondary findings should not be treated as a set of independent confirmations. Finally, planning figures derived from the observed effect show how strongly future sample-size requirements depend on the assumed target effect, but they cannot stand in for the trial's original design justification or be interpreted as observed power.

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 primary outcome was assessed immediately after the first visit, so the result does not describe longer-term distress or the effect of receiving repeated visits over a hospitalization. The actual number of visits received by participants is not reported in the registry record. The intervention schedule depended on hospitalization length and therapy dog team capacity, which defines the intended delivery but does not document individual exposure.

The analyzed groups were small: 11 in the therapy-dog group and 12 in control for the primary endpoint. The record notes three participants without the relevant first-visit assessment, but does not explain the reasons or describe any imputation or sensitivity analysis. The per-protocol criteria are also not reported. These omissions constrain assessment of how missingness and analysis population affected the estimates.

The registry calls the trial double masked but does not report who was masked or how masking was implemented. It also does not report the randomization procedure, allocation concealment, original sample-size justification, or full analysis plan. These details would help readers assess the design and analysis beyond the high-level labels.

The study included a broad set of psychosocial and microbial outcomes, but a multiplicity strategy is not reported. The posted secondary results therefore need cautious interpretation. The registry describes a planned observational microbial activity, but its conduct and results are not reported in the registry record. Finally, the registry lists deaths in the intervention group without reporting their causal relationship, limiting interpretation of safety entries.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Helping Improve Pediatric Patient Outcomes — NCT03471221.
• Registry record last updated 2026-06-15; status last verified 2026-06.
• Results first posted 2026-06-15.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT03471221, titled “Helping Improve Pediatric Patient Outcomes,” is a completed randomized, double-masked, parallel-group trial sponsored by Kaiser Permanente. The registry lists an enrollment of 26. The intervention was therapy dog visits, compared with usual medical care for hospitalized pediatric oncology patients.

The primary endpoint was the PedsQL VAS Total Distress Score immediately after the first visit. The recalculated unadjusted mean difference was -2.6, with a 95% confidence interval from -12.94 to 7.745 and an unadjusted p-value of 0.622. The registry's reported primary analysis was a median difference of -3.0, with a 95% confidence interval from -12.4 to 6.4.

The primary assessment groups included 11 intervention and 12 control participants. The registry lists zero serious adverse events in either group and two deaths in the intervention group, with causal relationship not reported. The original design assumptions, full analysis plan, and results of the proposed observational activity are not reported in the registry record.