Mailed FIT Outreach 2022: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Mailed FIT Outreach 2022 (NCT05341622), a randomized na study. The primary outcome is FIT Completion.

2. Clinical Question

3. Trial Design

The registry identifies this as a completed, randomized, double-masked, factorial pragmatic trial. It evaluated behavioral approaches within a centralized mailed screening outreach program at Penn Medicine. The trial enrollment is listed as 5460. Its recorded start was 2022-05-27, and primary completion was 2022-10-13.

The intervention factors were differentiated packaging versus standard mail, text versus no text, and personalized reminder versus no reminder. The registry lists eight combinations of these approaches. One arm description for Standard Mail + No Text + Reminder is blank, so the details of that description are not reported in the registry record; the arm title identifies the combination. The primary results group the combinations into factor-level comparisons: Box (Arms 1+2+3+4) versus Envelope (Arms 5+6+7+8), Text versus No Text, and Reminder versus No Reminder. These factor-level summaries overlap: a participant assigned to one packaging group also belongs to a text group and a reminder group. The groups should not be treated as separate, mutually exclusive arms.

The registry describes the study as pragmatic and states that its aim was to increase response to mailed FIT kits among eligible patients. The masking procedures, allocation sequence details, and original statistical analysis plan are not reported in the registry record. The phase is listed as NA. The registry does not report the rationale for the enrollment target or the methods used to determine it.

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.

Only the first 6 of 8 groups are shown.

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.

Only the first 6 of 9 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as FIT Completion, measured over 4 months.

The primary endpoint is FIT completion, measured as the count of participants who completed FIT within 4 months. The registry describes the measure as the percent of patients who complete FIT, but the posted outcome values are participant counts alongside group denominators. The count is a binary-outcome summary: each participant either completed FIT by the stated time or did not.

FIT completion is used because it directly reflects response to the mailed kit, the behavior the outreach strategies were designed to influence. A participant count alone does not indicate the rate without its denominator, so interpretation depends on considering the counts alongside the group sizes. The comparisons express the difference between group risks, with confidence intervals describing uncertainty around those differences.

This endpoint measures completion of the mailed test, not all colorectal cancer screening activity, follow-up after a positive FIT, or a health outcome. Those are separate outcomes in the registry. The result therefore addresses the stated outreach response question at the specified time point; it does not, by itself, answer broader questions about screening pathways or subsequent health effects.

7. Results Posted in the Registry

• Envelope (Arms 5+6+7+8): 18.0% (470/2613) versus 17.8% (469/2631) on control; risk difference 0.2 percentage points (95% CI -1.9 to 2.2), risk ratio 1.01 (0.90–1.13); unadjusted two-proportion z-test p = 0.879.
• Text (Arms 1+2+5+6): 21.2% (558/2628) versus 17.8% (469/2631) on control; risk difference 3.4 percentage points (95% CI 1.3 to 5.5), risk ratio 1.19 (1.07–1.33); unadjusted two-proportion z-test p = 0.002.
• No Text (Arms 3+4+7+8): 14.6% (381/2616) versus 17.8% (469/2631) on control; risk difference -3.3 percentage points (95% CI -5.3 to -1.3), risk ratio 0.82 (0.72–0.92); unadjusted two-proportion z-test p = 0.001.
• Reminder (Arms 1+3+5+7): 20.3% (532/2617) versus 17.8% (469/2631) on control; risk difference 2.5 percentage points (95% CI 0.4 to 4.6), risk ratio 1.14 (1.02–1.28); unadjusted two-proportion z-test p = 0.021.

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.

FIT Completion

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Envelope (Arms 5+6+7+8): 18.8% (492/2613) versus 19.2% (506/2631) on control; risk difference -0.4 percentage points (95% CI -2.5 to 1.7), risk ratio 0.98 (0.88–1.09); unadjusted two-proportion z-test p = 0.710.
  • Text (Arms 1+2+5+6): 22.6% (593/2628) versus 19.2% (506/2631) on control; risk difference 3.3 percentage points (95% CI 1.1 to 5.5), risk ratio 1.17 (1.06–1.30); unadjusted two-proportion z-test p = 0.003.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Envelope (Arms 5+6+7+8) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Text (Arms 1+2+5+6) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Colorectal Cancer (CRC) Screening Completion (Any)

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Envelope (Arms 5+6+7+8): 24.3% (640/2631) versus 24.9% (650/2613) on control; risk difference -0.6 percentage points (95% CI -2.9 to 1.8), risk ratio 0.98 (0.89–1.08); unadjusted two-proportion z-test p = 0.644.
  • Text (Arms 1+2+5+6): 28.1% (739/2628) versus 24.9% (650/2613) on control; risk difference 3.2 percentage points (95% CI 0.9 to 5.6), risk ratio 1.13 (1.03–1.24); unadjusted two-proportion z-test p = 0.008.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Envelope (Arms 5+6+7+8) and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically higher for Text (Arms 1+2+5+6) than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary results available in detail show a broadly consistent pattern across FIT completion at 6 months and colorectal cancer screening completion of any type at 6 months: the text comparisons have positive estimated risk differences, while the packaging comparisons have estimates near zero with intervals that include zero. This general pattern aligns with the direction of the primary outcome's text and packaging comparisons. It offers a description across outcomes, not independent confirmation that the primary pattern is definitive.

The registry lists additional secondary outcomes concerning scheduling and completion of follow-up after a positive FIT, but results for those outcomes are not reported in the registry record. The primary and secondary FIT completion outcomes are related measures observed at different time points, while any-screening completion captures a broader screening outcome. They should not be treated as interchangeable endpoints.

The secondary comparisons are unadjusted. When several outcomes and contrasts are examined, the chance of observing an apparently notable result by chance can increase. The registry record does not report a multiplicity adjustment, a hierarchy of secondary hypotheses, or sponsor-reported analyses. Consequently, secondary findings carry less weight than a prespecified primary analysis with a documented plan, and apparent consistency across related measures should be interpreted cautiously. These results can describe patterns in the posted data, but they do not establish that a strategy caused the same effect across all outcomes or follow-up times.

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

• FIT Completion
• Colorectal Cancer (CRC) Screening Completion (Any)
• FIT Positive Follow-up Scheduling
• FIT Positive Follow-up Completion

9. Statistical Interpretation

10. Statistical Methods Explained

The outcome is binary: FIT was either completed within the specified period or it was not. The registry reports counts of participants with completion and group denominators. A risk is the proportion of participants in a group who experienced the outcome. Comparing risks gives an absolute difference on the probability scale, called a risk difference. A risk difference of zero corresponds to equal observed risks; a positive or negative estimate indicates the direction of the comparison as it is defined.

What does the risk difference describe?

For the packaging comparison, the estimated risk difference is 0.002 for Envelope relative to Box. For Text, the estimate is 0.034 relative to the reference group, while the No Text estimate is -0.033. The Reminder estimate is 0.025, and the No Reminder estimate is -0.023. These are absolute contrasts, rather than relative measures such as a risk ratio or odds ratio. The displayed comparisons are unadjusted and arise from the posted counts and denominators; they do not account for possible covariates or the factorial structure in an adjusted model.

How should the confidence intervals be read?

A confidence interval gives a range of values compatible with the data under the method used to calculate it. It conveys uncertainty around an estimate, rather than identifying a single known effect. An interval that includes zero does not rule out a difference in either direction at the corresponding confidence level. An interval entirely on one side of zero indicates that the range calculated by that method does not include zero. This description is general statistical interpretation; it does not make an interval a guarantee about the true effect.

What does a p-value contribute?

A p-value describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure the importance of an effect. The p-values listed for the primary comparisons are unadjusted. Because multiple factor-level contrasts are presented, they should not be interpreted as if each were the only analysis undertaken. The registry record does not report the sponsor's statistical plan, any adjustment for multiple testing, or a set of sponsor-reported analyses.

How does the factorial structure matter?

In a factorial design, intervention components are assigned in combinations, and results can be summarized for each factor by grouping across levels of the other factors. Here, the packaging groups, text groups, and reminder groups overlap. The component summaries can help describe the average contrast associated with a factor across those combinations, but they do not show whether the effect of one component depends on another. Interaction estimates and a model addressing such dependence are not reported in the registry record. The arm-level combination results are not available in the posted primary outcome summaries.

The registry identifies the trial as randomized and double-masked, but it does not describe the masking procedure or the statistical methods used for the sponsor's own analysis. The comparisons are unadjusted summaries; they should not be mistaken for a reported adjusted factorial model. Details about missing outcomes, the analysis population, and handling of incomplete follow-up are also not reported in the registry record.

11. Confidence Intervals and Precision

Precision concerns how tightly the data constrain the estimated contrast. The risk-difference confidence intervals for the primary outcome range from -0.019 to 0.022 for Envelope versus Box; from 0.013 to 0.056 for Text versus the reference group; from -0.052 to -0.013 for No Text; from 0.004 to 0.046 for Reminder; and from -0.043 to -0.003 for No Reminder. These intervals differ in both width and position, so the estimates do not all have the same degree or direction of uncertainty.

The packaging interval includes zero and extends in both directions. It therefore does not pin down the direction of the contrast. The intervals for the text and reminder factor comparisons lie on one side of zero under the unadjusted calculation, but their widths still indicate uncertainty about the magnitude. Statistical evidence about whether an interval includes zero is distinct from precision: an estimate may be on one side of zero while still covering a range of possible effect sizes.

Confidence intervals are calculated from the observed group counts and denominators under a statistical method. They do not account for all possible sources of uncertainty, such as unreported missing-data handling, model choices, or multiplicity. The registry does not report the sponsor's analysis method or adjusted intervals. Accordingly, the posted intervals describe the unadjusted contrasts and should not be read as a complete assessment of uncertainty for all outcomes and comparisons in the trial.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning illustrations give participants per arm for 80% power and participants per arm for 90% power; they are not total sample sizes or total event counts. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for each illustration is the stated fraction of the observed effect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

For example, planning for 100% of the observed Text contrast gives 2,125 participants per arm for 80% power and 2,844 participants per arm for 90% power. Planning for 75% of that observed effect gives 3,715 participants per arm for 80% power and 4,973 participants per arm for 90% power. The illustrations apply separately to the comparison and outcome from which the observed effect was taken; they are not recommendations for a future study.

The 80% column represents the participant count per arm estimated to give an 80% chance of detecting the stated effect under the planning assumptions, if that effect and those assumptions hold. The 90% column has the analogous meaning for a 90% chance. Because smaller effects are harder to distinguish from random variation, planning for a smaller fraction of the observed effect requires more participants. These are planning illustrations based on observed results, 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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.178; treatment proportion p₁ = 0.180; average p̄ = 0.179.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5422 = 1.0627.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5422 = 0.4563.
4. n = (1.0627 + 0.4563)² / (0.0016)² = 889,499.3, rounded up to 889,500 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 <- 0.178259217027746                          # control proportion
p1 <- 0.179869881362419                          # 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(5341622)
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 = 889,500 (80% power), 1,190,787 (90% power)
 75% of observed effect: n per arm = 1,579,943 (80% power), 2,115,095 (90% power)
 50% of observed effect: n per arm = 3,551,742 (80% power), 4,754,774 (90% power)
power.prop.test: 889499.3 
Simulated power at n = 889,500 per arm: 0.799 (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.

14. Multiplicity, Missing Data and Analysis Populations

The trial evaluates multiple behavioral components and reports several outcome measures. The primary results include contrasts for packaging, text messages, and reminders, while the registry also lists secondary outcomes. Each additional comparison creates another opportunity for an apparently unusual result to arise by chance. A multiplicity plan can address how to interpret a family of comparisons, for example by adjusting inference or identifying a hierarchy of primary and secondary questions. Such a plan is not reported in the registry record.

The p-values and confidence intervals discussed for the primary contrasts are unadjusted. No adjustment for the set of component comparisons or secondary outcomes is reported. The record lists zero reported analyses, and the posted outcome summaries do not include a sponsor-reported adjusted model. It is therefore not possible to determine from the registry record whether these contrasts were prespecified, how interactions were considered, or how the sponsor intended to control multiplicity.

Missing-data methods and analysis populations are also not reported in the registry record. The posted outcome summaries include group denominators, but the record does not explain how incomplete outcome ascertainment, withdrawals, or other missing observations were handled. It does not state whether the analyses followed intention-to-treat principles, used a per-protocol population, or applied another definition. The registered enrollment and the denominators in the outcome summaries are not enough to infer those choices.

These omissions limit the ability to assess whether the posted comparisons represent the full planned analysis and how robust the inferences are to missing outcomes or multiple testing. They do not change the descriptive counts and unadjusted comparisons, but they matter when deciding how much inferential weight those results can carry.

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

The posted primary outcome data describe FIT completion within 4 months among groups defined by packaging, text, and reminder assignment. The unadjusted comparisons show estimates in different directions across the factors: the packaging estimate is close to zero, the text contrast is positive, the No Text contrast is negative, and the reminder and No Reminder contrasts point in opposite directions. Confidence intervals communicate the uncertainty around these estimates, and the unadjusted p-values summarize evidence under their respective null comparisons.

These findings do not establish that one outreach strategy is universally better, nor do they establish that the observed contrasts would recur in another setting. The study is described as pragmatic and conducted among eligible primary care patients at Penn Medicine; the registry record does not provide detailed participant characteristics or information sufficient to judge how results might apply elsewhere. The factorial summaries also combine participants across other intervention components, so they are not direct estimates for each specific intervention combination.

FIT completion is a process outcome. It is not itself a measure of cancer diagnosis, health benefit, or completion of all possible screening and follow-up steps. Although the registry lists secondary outcomes addressing broader screening and follow-up, not all listed outcomes have detailed results in the record. No medical conclusion should be drawn from these behavioral outreach comparisons alone.

The appropriate interpretation is therefore limited to the posted outcomes, time points, groupings, and unadjusted methods. The results describe observed differences and their calculated uncertainty; they do not resolve the unreported analysis choices, missing-data handling, or multiplicity plan.

16. Why This Trial Matters Statistically

This trial is statistically informative because it evaluates several outreach components within a factorial design rather than treating each combination as an unrelated intervention. Factor-level summaries can make comparisons of packaging, text messages, and reminders easier to interpret across combinations. At the same time, the overlap between those summaries means that they are not independent comparisons, and the design's structure needs to be considered when interpreting uncertainty and multiple testing.

The primary endpoint is a directly observed behavioral response to mailed FIT outreach. Reporting counts with denominators, absolute risk differences, and confidence intervals makes it possible to assess both the direction and uncertainty of the component comparisons. The contrast between estimates that are close to zero and estimates farther from zero also illustrates why point estimates should be read together with their intervals rather than as definitive answers.

The record also illustrates the distinction between trial outcomes and trial analysis. The registry contains outcome summaries and unadjusted comparisons, but does not report the original statistical plan, multiplicity handling, missing-data methods, or sponsor-reported analyses. Those details are essential for understanding how a factorial trial's many potential contrasts were selected and interpreted. The study is therefore useful for statistical reading not only because of its design and posted outcomes, but also because it shows where registry summaries leave important inferential questions unresolved.

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 registry record leaves several trial-specific details unreported. It does not give detailed eligibility criteria, participant characteristics, masking procedures, allocation sequence methods, or the original sample-size justification. Without those details, the precision of the estimates can be assessed from the posted summaries, but the full context for recruitment, implementation, and analysis cannot be reconstructed.

The primary outcome results are grouped by intervention factor rather than reported for each of the eight combinations. The factor-level groups overlap, and no interaction estimates are reported. This means the summaries do not show whether the effect associated with text messages, packaging, or reminders varied depending on the other components. The analyses are unadjusted, and the registry does not report a multiplicity plan, adjusted model, or sponsor analysis.

Missing-data handling and analysis-population definitions are not reported in the registry record. The outcomes for FIT-positive follow-up scheduling and completion are listed, but their results are not reported in the detailed secondary results. The adverse-event record lists zero serious and other affected counts in the arms, with one death recorded in Arm 5 and one in Arm 8; the record does not provide details establishing any relationship between those deaths and the interventions.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Mailed FIT Outreach 2022 — NCT05341622.
• Registry record last updated 2024-07-24; status last verified 2024-07.
• Results first posted 2024-07-24.
• 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, double-masked, factorial pragmatic trial of behavioral interventions for mailed FIT outreach at Penn Medicine. Enrollment is listed as 5460. The primary endpoint is FIT completion within 4 months, with posted summaries for packaging, text, and reminder groups. Secondary result details include FIT completion and any colorectal cancer screening completion at 6 months; other listed follow-up outcomes are not reported in the registry record. The posted contrasts are unadjusted. The original statistical plan, sample-size justification, missing-data methods, analysis populations, and multiplicity procedures are not reported.