Colorectal Cancer Screening Assessment Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Colorectal Cancer Screening Assessment Study (NCT04500171), a randomized na study. The primary outcome is Use of FOBT ColoCARE Home Kit.

2. Clinical Question

The study also describes process outcomes concerning development of suitable and comprehensible health-promotion materials and commitment to use the home kit. The registry record lists three primary outcome measures in total, but only one primary endpoint and its results are detailed in the posted outcome information. The other primary outcome results are not reported in the registry record.

3. Trial Design

The registry identifies this as a randomized, parallel study with no masking. It lists an enrollment of 264 and a completed status. The recorded study dates are 2020-09-21 for start and 2023-06-08 for primary completion. The sponsor is the University of Hawaii. The registry lists the study phase as NA.

Four outcome groups are reported: English Original Assessment Tools, Samoan Original Assessment Tools, English Modified Assessment Tools, and Samoan Modified Assessment Tools. The posted group sizes are 58.0, 90.0, 40.0, and 73.0, respectively. The descriptions identify English or Samoan assessment materials and original or modified tools. The registry’s listed behavioral interventions include versions of ColoCARE instruction sheets, colorectal cancer screening brochures, and an instruction video, with assessment approaches named in some intervention titles.

Random allocation means assignment is intended to reduce systematic differences between groups, on average. The registry record does not report the sequence-generation method, allocation concealment, whether assignment was stratified, or the implementation details of randomization. It also does not explain how the four groups were formed or specify a factorial or other design structure beyond the parallel model. The no-masking designation means the registry does not describe masking; it does not establish which participants or study personnel knew assignments at each stage.

The stated project involved a community-based and culturally relevant process, and the study’s purpose included assessing materials in English and Samoan for populations with varying health-literacy levels. The registry record does not provide further detail about community involvement, recruitment procedures, setting, eligibility criteria, or participant characteristics. Those omissions limit assessment of how closely the enrolled group reflects the population the project intended to address.

The posted outcome comparisons use English Original Assessment Tools as the reference and compare each of the other groups separately with it. This layout supports reading each reported contrast on its own. It does not, by itself, estimate a single overall effect of modification, language, or any one component of the materials. The registry record does not report a prespecified statistical analysis plan or the rationale for the particular contrasts.

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

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 Use of FOBT ColoCARE Home Kit, measured over 3 months.

The posted primary endpoint is the number of participants reporting results for the FOBT ColoCARE home kit at 3 months. It is expressed as a count of participants, not as a continuous measurement or a time-to-event outcome. The registry describes the measure as the number of participants reporting results for the home kit. The observed group counts are 14.0 for English Original Assessment Tools, 2.0 for Samoan Original Assessment Tools, 14.0 for English Modified Assessment Tools, and 0.0 for Samoan Modified Assessment Tools.

This endpoint is relevant to the project because its stated aims include assessing commitment to use the kit and use of the kit. A participant-reported kit result gives an observable outcome connected to that aim. It does not directly measure whether the educational materials were suitable or comprehensible, which are also listed among the project’s primary outcomes. Nor does the count alone show whether kit use was completed, whether a result was valid, or what happened to participants who did not report one; those definitions are not reported in the registry record.

Counts are useful for showing how many outcome events were recorded in each group. They are sensitive to the group size, so a count alone is not a like-for-like comparison when group sizes differ. The posted risk-difference comparisons account for the displayed group denominators by comparing the observed proportions. The registry record does not report the handling of missing outcome information or clarify whether every randomized participant was included in the endpoint denominator.

7. Results Posted in the Registry

• Samoan Original Assessment Tools: 2.2% (2/90) versus 24.1% (14/58) on control; risk difference -21.9 percentage points (95% CI -33.3 to -10.5), risk ratio 0.09 (0.02–0.39); unadjusted two-proportion z-test p = <0.001.
• English Modified Assessment Tools: 35.0% (14/40) versus 24.1% (14/58) on control; risk difference 10.9 percentage points (95% CI -7.6 to 29.3), risk ratio 1.45 (0.78–2.70); unadjusted two-proportion z-test p = 0.242.
• Samoan Modified Assessment Tools: 0.0% (0/73) versus 24.1% (14/58) on control; risk difference -24.1 percentage points (95% CI -35.2 to -13.1); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

In general, a risk difference compares the proportions of participants with an outcome in two groups by subtracting one group’s proportion from the reference group’s proportion. It is an absolute measure: a value of zero corresponds to equal observed proportions, and a negative or positive value indicates the direction of the contrast according to the stated group order. In these registry comparisons, the English Original Assessment Tools group is the reference. The reported risk differences are -0.219 for Samoan Original Assessment Tools, 0.109 for English Modified Assessment Tools, and -0.241 for Samoan Modified Assessment Tools.

The risk difference is useful because it expresses the contrast on the proportion scale and can be interpreted alongside the event counts and group sizes. The counts alone are not enough to compare groups with different denominators. A difference in counts is not the same quantity as a difference in proportions. The interpretation also depends on knowing which group is the reference and which group is being contrasted; reversing the order would reverse the sign.

In general, a confidence interval describes a range of values that are reasonably compatible with the observed data under the statistical procedure used. It communicates uncertainty around an estimate rather than a range containing a known true value with certainty. A narrower interval indicates greater precision than a wider interval, though precision does not resolve questions of bias, measurement quality, or generalizability. For a risk difference, an interval containing zero includes no difference as a compatible value under that procedure. An interval wholly below or above zero indicates that the interval does not include that null value. The registry record does not report the exact interval-construction method.

The posted p-values are unadjusted. In general, a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be if that hypothesis and the analysis assumptions held. It is not the probability that the null hypothesis is true, nor a measure of the size or importance of an effect. A small p-value does not establish that an observed difference is caused by the intervention, and a larger p-value does not prove equality. The registry record does not report the full testing procedure, whether the contrasts were prespecified, or whether any additional outcomes were analyzed.

For two comparisons, the record also gives risk ratios and odds ratios. In general, a risk ratio divides one group’s outcome proportion by the reference group’s proportion. A value of one corresponds to equal proportions; values below or above one indicate a lower or higher proportion relative to the reference, given the group order. An odds ratio compares odds rather than proportions. When an outcome is not common, these measures may be close, but they are not interchangeable, and the odds ratio can appear farther from one than the risk ratio. The posted risk and odds ratios should therefore be read according to their own scales, not as risk differences.

The third comparison has a risk difference and p-value but no posted risk ratio or odds ratio. Those measures are not reported in the registry record for that contrast. The relative measures for the first two comparisons do not replace the absolute risk differences: each answers a different descriptive question. It is also important not to treat the ratio estimates as an adjusted analysis. The registry record does not report covariate-adjusted estimates.

Randomization can support causal interpretation when assignment is implemented appropriately and the analysis preserves the benefits of assignment, but the registry record does not report enough detail to assess those conditions. It does not describe allocation concealment, analysis populations, adherence, crossovers, missing-outcome handling, or adjusted analyses. The posted contrasts are therefore best read as registry-reported group comparisons. The record does not specify whether the analysis followed intention-to-treat principles.

In general, over-dispersion is a concern when the variability in data exceeds what a statistical model assumes. It can make standard errors too small and confidence intervals too narrow if not addressed. For this binary count endpoint, the registry does not report a model-based dispersion parameter or an over-dispersion assessment. There is no basis in the record to claim that such a model was used or required.

The sample-size illustrations use a normal approximation, equal allocation between two arms, a two-sided significance level of 0.05, and a target effect expressed as a fraction of the observed effect. In general, these assumptions translate an effect size and a chosen power target into a planning sample size. They are not evidence that the original study was designed with those assumptions. The registry record does not report the original sample-size justification or statistical plan.

10. Confidence Intervals and Precision

The confidence intervals help show how precisely the registry estimates each risk difference. For Samoan Original Assessment Tools versus English Original Assessment Tools, the estimate is -0.219 and the interval runs from -0.333 to -0.105. The entire interval is below zero, so the reported range does not include equal observed proportions. The interval still spans a range of possible differences, which is a reminder that the point estimate is not exact.

For English Modified Assessment Tools versus the English original-tools group, the estimate is 0.109, with an interval from -0.076 to 0.293. This interval includes zero and extends in both directions. It therefore leaves uncertainty about whether the observed proportion is lower or higher in the modified-tools group. The p-value of 0.242 is consistent with that uncertainty under the unadjusted comparison.

For Samoan Modified Assessment Tools versus the reference group, the estimate is -0.241, with an interval from -0.351 to -0.131. This interval lies below zero. It communicates uncertainty around the size of the negative contrast while excluding zero under the reported interval procedure.

Intervals do not account automatically for every source of uncertainty. The registry record does not report whether the estimates account for multiple comparisons, missing outcomes, clustering, or covariate adjustment. It also does not give the interval method. Thus, the ranges describe precision as reported, not a complete accounting of possible bias or design uncertainty.

For the first two comparisons, the record also reports confidence intervals for risk ratios and odds ratios. Their interpretation is on relative scales and should be kept distinct from the risk-difference intervals. The registry does not report corresponding ratio intervals for the Samoan modified-tools comparison. Overall, the intervals provide more information than point estimates alone, but they do not identify why group outcomes differed or establish a causal mechanism.

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 sample-size illustrations are planning calculations, not observed study results. The figures are participants per arm, not total participants and not total events. Their assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The record gives illustrations for the Samoan Original Assessment Tools and English Modified Assessment Tools comparisons; a corresponding planning table for Samoan Modified Assessment Tools is not reported in the registry record.

For each available comparison, the 80% power column gives the planned number of participants per arm for a design with an 80% chance of detecting the assumed effect under the stated calculation assumptions. The 90% power column gives the number per arm for a higher target chance of detection, with the other stated assumptions held in the calculation. These are planning targets, not guarantees that a future study will produce a particular result.

The rows labeled 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect represent progressively smaller effects to detect. Smaller effects generally require more participants to achieve the same target power. For the Samoan original-tools comparison, the listed requirements range from 37 and 48 participants per arm at 100% of the observed effect to 198 and 264 at 50% of the observed effect, for the 80% and 90% targets respectively. For the English modified-tools comparison, the corresponding figures are 276 and 369 at 100%, and 1,045 and 1,398 at 50%.

These calculations use the observed effects as planning inputs and are not observed power. The registry record does not report that the trial’s original design used these assumptions; its own sample-size justification and original design assumptions are not reported in the registry record. Planning results based on observed effects can vary substantially with the effect chosen, especially when the observed estimate is uncertain.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.241; treatment proportion p₁ = 0.022; average p̄ = 0.132.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4784 = 0.9376.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4526 = 0.3809.
4. n = (0.9376 + 0.3809)² / (-0.2192)² = 36.2, rounded up to 37 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
p0 <- 0.241379310344828                          # control proportion
p1 <- 0.0222222222222222                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4500171)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 37 (80% power), 48 (90% power)
 75% of observed effect: n per arm = 77 (80% power), 103 (90% power)
 50% of observed effect: n per arm = 198 (80% power), 264 (90% power)
power.prop.test: 36.2 
Simulated power at n = 37 per arm: 0.885 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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 posted results include three pairwise comparisons with English Original Assessment Tools as the reference. Each comparison addresses a separate contrast. When several outcomes or contrasts are examined, the chance of obtaining at least one apparently unusual result can increase. In general, multiplicity procedures can control that issue by adjusting the testing threshold or estimates, but they can also change how results are interpreted.

The p-values are explicitly unadjusted. Whether the comparisons were prespecified, whether other analyses were conducted, and whether any multiplicity strategy was planned are not reported in the registry record. The values should therefore be read as unadjusted evidence for the individual posted comparisons, not as results corrected for a family of tests.

The record lists three primary outcome measures in the study description, but only the kit-result endpoint is detailed among the posted primary results. Results for the material-development and commitment outcomes are not reported in the registry record. It also lists no secondary outcomes posted. The absence of results in the record does not establish whether an outcome was not measured, was measured but omitted, or was reported elsewhere.

Missing-data handling is not reported in the registry record. The endpoint description refers to participants reporting kit results, but does not state whether participants without a reported result were counted as non-events, excluded, or handled in another way. The record does not provide an analysis population definition, such as whether all randomized participants were included, or describe exclusions, withdrawals, adherence, or protocol deviations.

The displayed group denominators are not identical to every adverse-event denominator. In particular, the Samoan Modified Assessment Tools outcome group is listed with 73.0 participants, while its safety group is listed with 76 at risk. The registry record does not explain this difference. That discrepancy reinforces the need to distinguish outcome-specific populations and denominators rather than assume the same analysis set for every measure.

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 serious adverse event figures report zero affected participants and zero deaths in each listed safety group. The groups are listed with 58, 90, 40, and 76 participants at risk, respectively. The registry also reports zero participants affected by other adverse events in each group. No serious adverse event terms are listed.

These figures describe what is recorded in the registry; they do not show that an event was impossible or that no events occurred outside the reporting definitions or observation period. The safety timeframe, event definitions, collection methods, and completeness of reporting are not reported in the registry record. A zero count also gives limited information about how frequently an uncommon event might occur.

The Samoan Modified Assessment Tools safety group has 76 participants at risk, whereas its primary-outcome group is listed with 73.0 participants. The registry record does not explain the difference. Safety denominators should therefore be read separately from the primary endpoint denominators. The posted figures are descriptive counts, not a comparative safety analysis.

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

The registry results describe differences in the proportion of participants reporting a FOBT ColoCARE home-kit result at 3 months across the four assessment-tool groups. Relative to English Original Assessment Tools, the reported risk difference is negative for both Samoan groups and positive for English Modified Assessment Tools. The unadjusted intervals for the two Samoan comparisons do not include zero; the interval for the English modified-tools comparison does include zero.

These findings do not establish that language or modification alone caused the differences. The comparisons are each against the same English original-tools reference, and the registry does not report direct comparisons between the two Samoan groups or a combined contrast of modified versus original materials. It does not report enough detail about assignment procedures, analysis populations, missing outcomes, or adjustment to assess all threats to interpretation.

The endpoint concerns reporting a kit result, not screening accuracy, detection of colorectal cancer, completion of follow-up care, or health outcomes. Although the project description includes assessment of material suitability, comprehension, and commitment to use the kit, results for those other stated outcomes are not reported in the registry record. The counts therefore address a specific behavioral reporting measure rather than the full set of project aims.

The results also should not be interpreted as evidence that a screening approach works or does not work clinically. They are descriptive trial outcome comparisons with uncertainty, and the unadjusted p-values are not adjusted for multiple contrasts. The registry record does not provide enough information to judge whether the observed patterns generalize beyond the participants and study setting or to explain the mechanisms behind them.

16. Why This Trial Matters Statistically

This trial is statistically interesting because it compares materials that vary by language and by original or modified form, while the posted outcome is a concrete participant-level count. That combination makes it important to distinguish a comparison between named groups from a claim about one isolated feature of the materials. The posted analyses compare each group with English Original Assessment Tools; they do not independently disentangle language and modification.

The outcome is also reported as both group counts and comparative estimates. Counts are readily understood but depend on the number of participants in each group. Risk differences address the comparison of proportions on an absolute scale, while risk ratios and odds ratios offer relative scales for two of the contrasts. Reading these measures together makes clear that the same evidence can be represented in different ways, and that the scales are not interchangeable.

The confidence intervals highlight how interpretation depends on precision as well as the point estimate. The English modified-tools contrast has an interval that includes zero, while the intervals for the two Samoan contrasts lie below zero. Those distinctions describe the reported estimates, but they do not settle causality, multiplicity, or generalizability. Unadjusted p-values should be interpreted with the number of contrasts and the unreported analysis plan in mind.

Finally, the record illustrates why outcome definitions and denominators matter. A report of kit results is not the same as a measure of comprehension or clinical outcome, and the safety denominator for the Samoan modified-tools group differs from its primary-outcome group. Statistical interpretation requires keeping those endpoints and populations separate.

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.

Several trial-specific limitations follow from what the registry does not report. Participant eligibility criteria, demographic and health-literacy characteristics, recruitment details, and the number approached are not reported in the registry record. Without those details, readers cannot assess how the study population relates to the intended population or how representative the results may be.

The registry identifies randomization but does not report the allocation sequence, concealment, implementation, or any stratification. It also does not explain whether the four groups were planned as a factorial design or describe the prespecified comparisons. The record does not report the statistical analysis plan, covariate adjustment, or whether the posted analyses were specified before the data were examined.

The endpoint counts concern participants reporting a kit result, but the criteria for a report, the collection procedure, and the treatment of participants without a report are not described. Missing-data handling and analysis populations are not reported in the registry record. The record also does not provide results for the other stated primary outcomes, including process outcomes about material development and commitment to kit use.

There is a difference between the listed primary-outcome group size of 73.0 for Samoan Modified Assessment Tools and the safety count of 76 at risk. The registry record does not explain this discrepancy. Safety observation periods, event definitions, and adverse-event collection methods are also not reported.

Finally, the reported p-values are unadjusted, and the record does not report a multiplicity plan or the original sample-size justification. These omissions constrain interpretation of both statistical uncertainty and the intended scope of inference.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Colorectal Cancer Screening Assessment Study — NCT04500171.
• Registry record last updated 2025-06-24; status last verified 2025-06.
• Results first posted 2025-06-24.
• 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 describes a completed randomized, parallel study with no masking, sponsored by the University of Hawaii. The study assessed English- and Samoan-language colorectal cancer screening materials in original and modified forms. The posted primary endpoint is the number of participants reporting results for the FOBT ColoCARE home kit at 3 months.

The reported counts are 14.0 for English Original Assessment Tools, 2.0 for Samoan Original Assessment Tools, 14.0 for English Modified Assessment Tools, and 0.0 for Samoan Modified Assessment Tools. Each comparison is against English Original Assessment Tools. The two Samoan contrasts have negative risk-difference intervals that do not include zero; the English modified-tools interval includes zero. The p-values are unadjusted.

The registry record does not report the analysis plan, missing-data handling, analysis populations, or results for the other stated primary outcomes. Serious adverse events, other adverse events, and deaths are recorded as zero in the listed safety groups.