BETTER HEALTH: Durham: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for BETTER HEALTH: Durham (NCT03052959), a randomized na study. The primary outcome is Composite Outcome Measure: Mean Percentage of CDPS Actions at Baseline That Are Subsequently Met at Follow-up.

2. Clinical Question

This is a question about reported completion of prevention and screening actions, not about diagnoses, disease events, or long-term health outcomes. The primary measure was assessed at the patient level, but the intervention was organized across clusters. The registry identifies the study as randomized and parallel, with no masking. It does not report the detailed allocation procedure or the number of clusters assigned to each group.

3. Trial Design

BETTER HEALTH: Durham is recorded as a randomized, parallel-group study with no masking. The registry describes an intervention delivered in designated clusters in Durham Region. Some clusters were to receive the BETTER intervention right away, while clusters assigned to the wait-list control group were to receive it six months later. This is therefore a comparison of immediate access with delayed access during the primary follow-up period.

The intervention involved supportive meetings with a specially trained prevention practitioner nurse. Participants reviewed recommended chronic disease prevention and screening activities and identified goals for accomplishing them over the following six months. Promotion, recruitment, and intervention delivery were intended to be adapted to residents’ needs through participatory research methods and community engagement strategies. The listed intervention types are behavioral for the immediate-intervention arm and other for the wait-list arm.

The primary outcome was measured six months after the baseline survey interview. Research assistants were to administer a follow-up survey about health and CDPS actions to participants in intervention and control clusters alike. The registry says all outcomes are self-reports of completion of CDPS actions. The study also lists completion of individual actions and number of self-referrals as secondary outcomes, but posted analyses for these outcomes are not reported in the registry record.

The registry reports enrollment of 126 and marks the study completed. It lists a start date of 2017-10-02 and primary completion date of 2020-01-28. It names Unity Health Toronto as sponsor and lists the phase as not applicable. The record does not report the original sample-size justification, the exact allocation of clusters, or a detailed statistical analysis plan. The posted arm-level outcome counts are 59 for immediate intervention and 66 for wait list; these arm counts are reported as recorded and should not be treated as an explanation of enrollment or participant flow.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Composite Outcome Measure: Mean Percentage of CDPS Actions at Baseline That Are Subsequently Met at Follow-up, measured over Six months.

The primary endpoint is the mean percentage of CDPS actions that were eligible at baseline and subsequently reported as met at the six-month follow-up. Eligibility was determined from baseline characteristics: different participants could be eligible for different prevention or screening actions. At follow-up, each participant was re-evaluated, and the number of eligible actions met was enumerated. The posted outcome is therefore a continuous measure of the proportion of a participant’s eligible actions completed, summarized as a group mean and standard deviation.

This composite measure gives a common summary across participants who may have different eligible actions. Rather than treating all participants as if they had the same set of recommended activities, it relates completion to the actions for which each person was eligible. The registry states that completion was self-reported, so the measure reflects participants’ reports rather than an independently verified record of each action.

The endpoint is relevant to the intervention’s stated goal of helping people identify and accomplish personal prevention and screening goals. It captures reported action completion over the specified follow-up period, but it does not directly measure changes in disease incidence, long-term health, or the quality of each action. Those outcomes are not reported as the primary measure in the registry record.

7. Results Posted in the Registry

• Wait List Intervention: mean 42.10 versus 64.50 on control; difference -22.40 (95% CI -31.47 to -13.33), standardized effect d = -0.87; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted primary outcome is continuous: each participant’s score is a percentage of eligible CDPS actions reported as completed, and the registry summarizes the score with a mean and standard deviation for each arm. A mean describes the group’s average outcome; a standard deviation describes how dispersed individual outcomes are around that group average. These are different summaries. A mean difference compares group averages, while the standard deviation describes within-group variation.

What does the mean difference represent?

The comparison reports the wait-list group relative to the immediate-intervention group. The means are 42.1 and 64.5, respectively, and the registry reports a mean difference of -22.4. This is a contrast in the outcome’s percentage-point scale, not a relative percentage change. Its sign depends on the order of subtraction: a negative result indicates that the first-named group in this contrast has a lower mean. The comparison is unadjusted and based on the posted arm-level summary statistics.

What does the confidence interval add?

A confidence interval is a range of values generated by a statistical procedure that expresses the estimate’s precision under the model and assumptions used. For the posted mean difference, the interval runs from -31.47 to -13.33. A narrower interval would indicate greater precision than a wider interval on the same scale. The interval is not a range containing a specified proportion of individual outcomes, and it does not account for sources of uncertainty that were not included in the calculation. In particular, the registry does not report a cluster-adjusted interval or enough detail to establish whether clustering was addressed in the trial’s own analysis.

How should the p-value be read?

The unadjusted p-value is <0.001. In general, a p-value assesses how unusual data at least as incompatible with a stated null model would be if that model and the test assumptions held. It is not the probability that the observed result arose by chance, nor the probability that the null model is true. The label “unadjusted” matters: the registry does not report adjustments for other outcomes, multiple comparisons, covariates, or cluster structure in this posted comparison.

Why does the design matter?

In a cluster-randomized study, assignment occurs at the group or cluster level, while outcomes may be measured on individual participants. People in one cluster can share characteristics, services, or local conditions, so their outcomes may be more alike than outcomes from unrelated clusters. This dependence is called within-cluster correlation. If it is ignored, an analysis may overstate the amount of independent information and understate uncertainty. This is a general statistical consideration, not a claim that the registry’s own analysis did or did not account for it. The registry reports the randomized, parallel design and cluster-based delivery, but the detailed randomization procedure and analysis method are not reported in the registry record.

Randomization is intended to make comparison groups similar on average at the point of assignment, but it does not guarantee exact balance in every characteristic. Cluster-level allocation also means the number of independent units relevant to inference is related to the clusters, not only to the participant counts. The record does not provide enough detail to assess balance, the number of clusters per arm, or any adjustment used to address cluster structure.

What does a pooled standard deviation mean?

The posted pooled standard deviation is 25.61. In general, pooling summarizes variability across groups into one scale, often for standardized comparisons or calculations that assume a common spread. The posted standardized effect size is not needed to interpret the means and confidence interval, and it is not repeated because the registry’s displayed value has more than three decimal places. The arm-specific standard deviations remain useful for understanding variability in the outcome within each group.

All primary outcomes are described as self-reports. Self-report is a measurement method, not a statistical adjustment. Differences in recall or reporting between groups could affect measured completion, while the registry does not report validation procedures or how missing survey responses were handled. These measurement and analysis details matter when interpreting a numerical contrast, but their presence or magnitude cannot be inferred from the posted means.

10. Confidence Intervals and Precision

The registry reports a mean difference of -22.4 for the wait-list group relative to the immediate-intervention group, with a confidence interval from -31.47 to -13.33. The interval lies on the negative side of zero, consistent with the reported direction of the contrast: the wait-list mean is lower than the immediate-intervention mean. Its endpoints express uncertainty about the mean difference under the calculation used; they do not identify every plausible source of uncertainty in the trial.

Precision is about how tightly an estimate is located, not whether the estimate is important or whether the study result applies beyond its setting. The interval is expressed on the outcome’s scale, so it can be read alongside the group means of 64.5 percent and 42.1 percent. It is not an interval for the average of each participant’s individual change, and it is not a prediction interval for an individual’s result.

The posted interval is associated with an unadjusted comparison. The registry does not report whether the trial’s own analysis used a cluster-adjusted variance estimate, adjusted for baseline factors, or handled missing outcome data in a particular way. Those choices could affect precision. With a cluster-based intervention, accounting for similarity within clusters is especially relevant to uncertainty; the information needed to determine how that was handled is not reported in the registry record.

The confidence interval should therefore be read as the precision of the posted comparison as calculated, rather than as a complete description of every design and measurement uncertainty. The unadjusted p-value is <0.001, but the interval is more informative about the range of mean differences compatible with the posted calculation. Neither statistic alone describes the spread of individual outcomes, which is reflected separately by the arm-specific standard deviations.

11. 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. They are illustrations based on the observed effect, not the trial’s original design assumptions; those original assumptions are not reported in the registry record. The stated planning approach uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is specified as a fraction of the observed effect.

For planning for 100% of the observed effect, the listed requirement is 21 participants per arm for 80% power and 28 participants per arm for 90% power. For planning for 75% of the observed effect, the corresponding figures are 37 and 49 participants per arm. For planning for 50% of the observed effect, they are 83 and 110 participants per arm. These are planning illustrations, not observed power.

In general, power is the chance that a planned study will detect an effect of a specified size under the assumptions used, if that effect is present. The 80% column represents a design intended to have an 80% chance of detecting the specified effect under those assumptions; the 90% column represents a higher chance. Smaller effects are harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants per arm. These figures do not establish that the original trial was designed around the observed result or that its actual analysis had the assumptions used for the illustration.

12. 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 = -22.400; pooled standard deviation σ = 25.611.
2. n = 2 × (1.96 + 0.842)² × 25.611² / -22.400² = 20.5, rounded up to 21 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 <- -22.4                      # observed difference in means
sigma <- 25.6114018054269                     # 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(3052959)
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 = 21 (80% power), 28 (90% power)
 75% of observed effect: n per arm = 37 (80% power), 49 (90% power)
 50% of observed effect: n per arm = 83 (80% power), 110 (90% power)
power.t.test: 21.5 
Simulated power at n = 21 per arm: 0.791 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists two secondary outcomes: completion of individual actions and number of self-referrals. It reports three outcome measures, but posted results include a primary outcome comparison and no posted analyses. Results for the secondary outcomes are not reported in the registry record. The record does not describe a prespecified hierarchy among outcomes or a procedure for controlling the chance of false-positive findings when examining multiple outcomes.

Multiplicity is a general statistical concern when several outcomes, time points, subgroups, or analyses are tested. The more comparisons made, the greater the opportunity for at least one apparently notable result to arise even when there is no underlying difference. Adjustments or a prespecified testing strategy can address this issue, but the registry does not report whether either was used. The posted primary comparison is labelled unadjusted, and its p-value should be interpreted in that context rather than as evidence that multiplicity was handled.

Missing-data handling is also not reported in the registry record. The primary outcome depended on a follow-up survey at six months, and the result is based on reported action completion. The registry does not give the number of participants with missing baseline or follow-up measurements, explain how missing responses were treated, or describe any sensitivity analysis. It is therefore not possible to determine from the posted summary how missing outcome data may have affected the group means or comparison.

The analysis population is not described in enough detail to identify whether the results follow an intention-to-treat approach, a complete-case approach, or another rule. The registry’s arm-level outcome counts are reported, but participant flow and the relationship between enrollment and the analysis sets are not explained. The detailed statistical plan, including covariate adjustment and methods for the cluster design, is not reported in the registry record. These omissions limit assessment of whether the posted unadjusted comparison reflects the study’s planned primary analysis.

14. 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 serious adverse events for the immediate-intervention group as zero affected among 59 at risk, and for the wait-list group as zero affected among 66 at risk. Other adverse events are also reported as zero affected in both groups. The deaths count is zero in each group. No serious adverse event terms are listed.

These figures describe the adverse-event fields posted in the registry record. They do not establish that no adverse experiences of any kind occurred, nor do they show how events were sought, classified, or monitored. The record does not provide narrative details about safety surveillance or explain whether events outside the listed categories were collected. A zero count in a posted field should be read as the registry’s reported count for that field, not as proof of absence of every possible harm.

The at-risk counts are group-specific and are not a measure of the primary outcome. The safety figures also do not indicate whether an event was related to the intervention. Relatedness assessments and further safety analysis are not reported in the registry record.

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

The posted data show that the immediate-intervention group had a higher mean percentage of baseline-eligible CDPS actions reported as met at six months than the wait-list group. The means were 64.5 percent and 42.1 percent, respectively. The registry’s unadjusted contrast, defined as wait list relative to immediate intervention, is -22.4, with a confidence interval from -31.47 to -13.33 and an unadjusted p-value of <0.001. These statistics describe the posted comparison for this outcome and follow-up period.

The result is not a count of people who completed all their recommended actions. It is a group mean of participant-level percentages, and participants could be eligible for different CDPS actions based on baseline characteristics. The measure relies on self-reported completion. It does not establish that actions were independently verified or that reported actions led to changes in health.

The study design was randomized and parallel, with intervention delivery organized in clusters. The registry does not report the detailed allocation procedure, the number of clusters per arm, or whether the posted comparison accounted for cluster dependence. The reported estimate is therefore best read as the unadjusted comparison displayed in the record, not as a substitute for a documented cluster-adjusted analysis. Missing-data methods, analysis populations, and the original statistical plan are also not reported.

The result should not be taken to mean that every participant experienced the group-average outcome, that the result applies automatically to populations outside the study setting, or that the intervention changed disease outcomes. The registry’s stated focus is completion of prevention and screening activities, and the primary result pertains to reported actions over six months. Evidence about longer-term outcomes and secondary outcomes is not reported in the registry record.

16. Why This Trial Matters Statistically

This trial illustrates a common challenge in evaluating public-health interventions: the intervention is delivered in community settings, while outcomes are measured for individuals. The registry describes delivery across clusters and individual-level assessment of reported prevention and screening actions. That structure makes the unit of randomization and the unit of outcome measurement important to statistical interpretation. A participant-level summary alone does not reveal whether the analysis properly reflected the cluster-based allocation.

The endpoint also raises a useful measurement question. Participants were eligible for different actions according to baseline characteristics, and the primary measure expressed completion in relation to each person’s eligible actions. This creates a composite summary that can compare groups despite differing sets of eligible activities. At the same time, interpretation depends on how eligibility and completion were determined, and the registry states that completion was self-reported.

The posted mean difference is accompanied by a confidence interval and an unadjusted p-value. Reading these together helps distinguish the estimated direction and precision from the broader question of study quality. Statistical significance does not resolve concerns about clustering, missing outcomes, self-report, or whether multiple outcomes were considered. The registry does not report enough analysis detail to assess these issues fully.

Finally, the sample-size illustrations show how planning changes when the target effect is smaller or the desired power is higher. They are distinct from the trial’s original design rationale, which is not reported in the registry record. Keeping planning illustrations separate from observed results is important: a calculation based on an observed effect cannot establish that a study was adequately planned or that the observed effect is unbiased.

17. 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 is based on self-reported completion of CDPS actions. The registry does not describe whether reports were checked against clinical or administrative records, or whether the way actions were reported differed between groups. Self-report can be affected by recall and reporting, so the posted measure may not be identical to independently verified completion.

The intervention was organized through clusters, but the registry does not report the number of clusters assigned to each arm or the statistical method used to account for clustering. The posted comparison is unadjusted. Without the trial’s analysis details, it is not possible to determine whether the reported confidence interval and p-value reflect the cluster-based design, baseline characteristics, or other prespecified adjustments.

The registry does not report missing-data procedures, participant flow, or the analysis population. Because the primary outcome was collected through a follow-up survey, missing responses or incomplete reporting could affect the available outcome data. The posted arm-level counts alone do not explain how participants were included in the analysis or how missing measurements were treated.

There are also limits to the scope of the endpoint. The measure captures reported completion of eligible prevention and screening actions at six months; longer-term health outcomes are not the primary endpoint. Secondary outcomes are listed, but their posted results are not reported in the registry record. The original sample-size justification, detailed statistical plan, and any multiplicity strategy are also not reported. These limitations do not change the posted numerical comparison, but they constrain how fully it can be evaluated and generalized.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: BETTER HEALTH: Durham — NCT03052959.
• Registry record last updated 2021-11-01; status last verified 2021-10.
• Results first posted 2021-11-01.
• 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.

21. Record Summary

The registry record describes BETTER HEALTH: Durham, a completed randomized, parallel study with no masking. It lists Unity Health Toronto as sponsor and enrollment of 126. The study population was people aged 40–64 years in designated areas of Durham Region in Oshawa and Whitby. The intervention used supportive meetings with a specially trained prevention practitioner nurse, while the comparator was a wait-list group scheduled to receive the intervention later.

The primary outcome was the mean percentage of baseline-eligible CDPS actions reported as met at six months. Posted means were 64.5 for immediate intervention and 42.1 for wait list. The registry reports an unadjusted mean difference of -22.4 for wait list relative to immediate intervention, with a confidence interval from -31.47 to -13.33 and p < 0.001. The measure relied on self-report. Detailed analysis methods, missing-data handling, and cluster-adjustment procedures are not reported in the registry record.