Getting to Yes, Michigan! (G2YMI): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Getting to Yes, Michigan! (G2YMI) (NCT05096260), a randomized na study. The primary outcome is Self-reported COVID-19 Vaccine Uptake.

2. Clinical Question

3. Trial Design

Getting to Yes, Michigan! (G2YMI), also titled Community-Centered Interventions for Improved Vaccine Uptake for COVID-19 (CIVIC): Getting to Yes, Michigan!, is a randomized, parallel-group study. The registry lists no masking. It records the study as completed, with enrollment of 1180, a start date of 2022-07-05, and a primary completion date of 2025-09-01. The listed phase is NA. The sponsor is the University of Michigan.

The study description places the work in four Michigan counties: Wayne, Genesee, Kent, and Washtenaw. It describes a community-based participatory research approach and states that the project aimed to understand barriers and drivers of vaccine uptake and hesitancy, test a multicomponent intervention, and sustain and evaluate the CIVIC partnership. The stated focus was communities experiencing disproportionate COVID-19 burdens, including African American and Latinx communities.

The intervention group received motivational interviewing-based SMS/MMS messages and completed baseline and follow-up surveys. The control group received a simple website and completed the same survey occasions; the description states that control participants would receive the motivational interviewing-based messages after those surveys. Although the intervention name also mentions web content, the registry record does not report additional operational details about that content.

The primary outcome was assessed 6 months after the start of the intervention. The primary results include 417 intervention participants and 459 control participants. These analysis counts do not add up to the registry's total enrollment, but the record does not explain the difference or give a participant-flow account. Eligibility criteria, the original sample-size justification, and detailed allocation procedures are 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: We've added the number of participants analyzed.

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: Participants could choose not to disclose demographics.

6. Primary Endpoint

The registry lists the primary outcome as Self-reported COVID-19 Vaccine Uptake, measured over 6 months after the start of the intervention.

The primary endpoint is self-reported COVID-19 vaccine uptake, measured as a binary participant-level outcome. The registry describes it as participant-reported vaccine uptake and gives the assessment timeframe as 6 months after the start of the intervention. A binary endpoint classifies each participant according to whether the event was reported, then compares the number or proportion with the event between groups.

The posted results count 117 participants with reported uptake among 417 analyzed in the intervention group, and 123 among 459 analyzed in the control group. The endpoint is relevant to the stated study aim because the project sought to increase vaccine uptake. Self-report is the measurement used in the registry record; whether it was verified against another source is not reported in the registry record.

A participant count makes the event totals directly interpretable, but counts alone do not account for differing group sizes. The comparison therefore also presents an estimated risk difference and a confidence interval. Those statistics describe the contrast between the groups and its uncertainty, rather than merely comparing the raw event counts. The registry record does not report further details on outcome ascertainment, missing responses, or the exact analysis population definition.

7. Results Posted in the Registry

• Intervention: 28.1% (117/417) versus 26.8% (123/459) on control; risk difference 1.3 percentage points (95% CI -4.7 to 7.2), risk ratio 1.05 (0.84–1.30); unadjusted two-proportion z-test p = 0.676.

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.

Percent of Participants Not Vaccinated Who Reported HIGH Intention to Uptake COVID-19 Vaccine

• 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:
  • Intervention: 53.0% (159/300) versus 52.1% (175/336) on control; risk difference 0.9 percentage points (95% CI -6.9 to 8.7), risk ratio 1.02 (0.88–1.18); unadjusted two-proportion z-test p = 0.817.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Intervention and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary endpoint: high intention to take up COVID-19 vaccine among participants who were not vaccinated. The unadjusted comparison has a risk-difference confidence interval spanning zero and p = 0.817. As with the primary endpoint, this result does not precisely distinguish the groups under the posted unadjusted comparison. It concerns reported intention within a subset, whereas the primary endpoint concerns self-reported vaccine uptake in the primary analysis population.

The broad pattern is therefore that neither posted unadjusted comparison provides a precise distinction between groups. The secondary measure can add context about stated intention, but it is not a substitute for the primary uptake outcome. Intention and reported behavior are different outcomes, and participants eligible for the secondary outcome are not the same analysis group as those in the primary endpoint.

Secondary results generally carry less weight than a prespecified primary outcome because they may be one of several questions considered, may use a smaller subset, and may be more vulnerable to chance findings when multiple outcomes are tested. The registry lists two outcome measures in total, one primary and one secondary, but does not report a multiplicity adjustment or a hierarchy for interpreting the secondary result. The Chi-squared analysis entry is not clearly linked to an endpoint in the record and should not be treated as a confirmed secondary finding.

The result is unadjusted and should be considered alongside the endpoint definition, analysis population, and uncertainty. The registry record does not report whether the secondary outcome was prespecified in a detailed analysis plan, how missing responses were handled, or whether additional secondary analyses were conducted but not posted. Without those details, the posted secondary result is best treated as supportive descriptive evidence rather than a stand-alone basis for broad conclusions.

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

• Percent of Participants Not Vaccinated Who Reported HIGH Intention to Uptake COVID-19 Vaccine

9. Statistical Interpretation

10. Statistical Methods Explained

The central comparison is between two randomized groups: participants assigned to motivational interviewing-based SMS/MMS messages and participants assigned to a simple website. The design is listed as parallel, with no masking. The primary outcome is binary, so each participant's reported uptake contributes to one of two categories: event or no event. The posted event counts and analyzed group sizes are the basis for the arm-level comparison.

What does a risk difference describe?

A risk difference expresses the difference between the event proportions in two groups. A value of zero corresponds to equal event proportions; positive and negative values indicate opposite directions according to the comparison's group ordering. The posted primary risk difference is 0.013. Because the calculation is unadjusted, it compares the arm-level results without adjustment for other participant characteristics. The registry record does not report an adjusted primary estimate.

How should the confidence interval and p-value be read?

A confidence interval describes a range of values compatible with the observed data under the method's assumptions. It communicates uncertainty around an estimate; it is not a range containing a known true value with certainty. For the primary risk difference, the interval runs from -0.047 to 0.072. Since it crosses zero, the posted interval is compatible with both a difference favoring either group and little or no difference. The unadjusted p-value is 0.676. A p-value measures how unusual data at least as inconsistent with a specified null model would be if that model and the analysis assumptions held. It is not an effect size and is not the probability that the null hypothesis is true.

Why can counts and contrasts tell different things?

Event counts alone can mislead when group sizes differ. The intervention group has 417 participants in the primary analysis, while the control group has 459. A risk-based contrast accounts for each group's size by comparing proportions rather than raw event totals. The posted comparison also includes risk-ratio and odds-ratio estimates. These effect measures use different scales: a risk difference describes an absolute contrast, while ratios describe relative contrasts. Each needs to be read with its own confidence interval and interpretation. The record does not report confidence intervals for every separately listed analysis.

What is the role of the Chi-squared analysis?

The registry record lists a Chi-squared analysis with an odds-ratio value of 0.05 and p < .05, without confidence limits. A Chi-squared test is generally used to assess whether categorical outcomes and group assignment are associated. The record does not explain the relationship between this analysis entry and the posted primary counts and risk-difference comparison. It is not appropriate to assume that the reported entry is an adjusted analysis or to infer its coding and direction from the value alone. The statistical analysis plan and a reconciliation are not reported in the registry record.

Why does over-dispersion matter in general?

General statistical context: over-dispersion occurs when observed variation is greater than a model expects. If it is ignored in a setting where it is present, uncertainty estimates can be too narrow and tests can appear more decisive than warranted. The posted primary outcome is a participant-level binary measure, and the registry record does not report a model that addresses over-dispersion or a need for such an adjustment. This general concept should not be mistaken for a finding about this trial.

The registry lists the primary outcome as a count of participants with an event, but the accompanying comparison uses binary-outcome measures. The exact statistical procedures used for the posted unadjusted estimates, including any continuity corrections or covariate handling, are not reported in the registry record. The distinction matters because different methods can produce different estimates of uncertainty, particularly when the event is uncommon or when group sizes differ.

Random allocation is intended to support a fair comparison by distributing prognostic factors between groups on average, but it does not guarantee that the analyzed groups remain comparable when outcomes are missing or participants are excluded after allocation. The record does not describe allocation concealment, adherence, crossover, or the analysis population principle. These omissions limit assessment of how closely the posted analysis follows the randomized assignment.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.013, with a confidence interval from -0.047 to 0.072. The interval covers zero and extends to values on both sides of it. This means the observed data are compatible with more than one direction of difference, as well as with little difference. The result is therefore not precise enough to establish a clear direction or narrow range for the contrast.

Precision is distinct from whether a point estimate is close to zero. A point estimate is the value calculated from the observed groups; its confidence interval expresses uncertainty. The primary p-value of 0.676 is consistent with the interval's inclusion of zero, but the interval is more informative about the range of contrasts compatible with the data. Neither statistic demonstrates that the groups are equivalent.

The secondary endpoint's unadjusted risk-difference interval also crosses zero, from -0.069 to 0.087. It is a separate outcome in a subset of participants, so its uncertainty should be interpreted on its own rather than pooled with the primary endpoint. The registry record does not give confidence limits for the listed Chi-squared odds-ratio entry.

These intervals reflect sampling uncertainty under the analysis methods, but they do not account for every possible source of uncertainty. The record does not report how missing data were handled, whether the analysis was adjusted, or whether any sensitivity analyses were conducted. Those details could affect interpretation, but their impact cannot be quantified from the registry record.

12. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm, not total participants, because the outcome is not time-to-event. The planning illustration assumes two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. It uses the observed effect as its starting point and considers detecting 100%, 75%, or 50% of that observed effect. The original trial design assumptions and original sample-size justification are not reported in the registry record.

For 100% of the observed primary effect, the illustration gives 19,675 participants per arm for 80% power and 26,339 per arm for 90% power. For 75% of the observed effect, the corresponding figures are 34,854 and 46,658 participants per arm. For 50% of the observed effect, they are 78,137 and 104,603 participants per arm. The corresponding secondary-endpoint planning figures are based on that endpoint's observed contrast, rather than the primary result.

In plain terms, 80% power is the probability of detecting an effect of the specified size under the planning assumptions if that effect is truly present; 90% power uses a higher probability of detection and therefore requires a larger planned sample in these illustrations. The figures are not the sample sizes used to establish what happened in this study. Smaller effects are harder to distinguish from random variation and require more participants under the stated assumptions. These calculations are planning illustrations, not observed power.

The illustration depends on an observed estimate that is itself uncertain. It should not be read as a recommendation for a future study without considering its research question, population, outcome definition, analysis plan, and other design considerations. The registry record does not report the assumptions used in the study's original planning.

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 = 0.268; treatment proportion p₁ = 0.281; average p̄ = 0.274.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6309 = 1.2366.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6309 = 0.5310.
4. n = (1.2366 + 0.5310)² / (0.0126)² = 19,674.9, rounded up to 19,675 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 <- 0.26797385620915                          # control proportion
p1 <- 0.280575539568345                          # 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(5096260)
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 = 19,675 (80% power), 26,339 (90% power)
 75% of observed effect: n per arm = 34,854 (80% power), 46,658 (90% power)
 50% of observed effect: n per arm = 78,137 (80% power), 104,603 (90% power)
power.prop.test: 19674.9 
Simulated power at n = 19,675 per arm: 0.800 (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

Multiplicity arises when several outcomes, time points, subgroups, or analyses are examined. As the number of tests increases, the chance of at least one apparently notable result can rise even when there is no underlying difference. The registry lists two outcome measures: the primary vaccine-uptake endpoint and one secondary endpoint concerning high intention among participants who were not vaccinated. A multiplicity adjustment, testing hierarchy, or other strategy for controlling interpretation across outcomes is not reported in the registry record.

The record gives one Chi-squared analysis entry with an odds-ratio value of 0.05 and p < .05, but it does not clearly identify how that entry relates to the posted primary and secondary contrasts. No confidence interval accompanies that entry. Because its relationship to the outcome comparisons is unclear, it should be interpreted cautiously and should not override the separately posted unadjusted comparisons without clarification.

The primary outcome analysis includes 417 intervention participants and 459 control participants, compared with total enrollment of 1180. The record does not describe participant flow, reasons for non-inclusion, or whether the posted analyses follow an intention-to-treat approach. It also does not report how missing outcome data were handled, whether any imputation was used, or whether sensitivity analyses assessed the consequences of missingness. The numbers excluded from the primary analysis cannot be inferred from the enrollment total and posted arm counts alone.

Analysis population details are also limited. The registry identifies participants analyzed but does not provide a full definition of inclusion in each analysis or describe adherence and crossover. Randomization supports interpretation of a group comparison when participants are analyzed appropriately, but exclusions or missing outcomes can weaken that protection. These missing reporting details do not establish that the analysis was biased; they limit the ability to evaluate that possibility.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports serious adverse events and other adverse events separately for the intervention and control groups. For serious adverse events, the at-risk counts are 696 in the intervention group and 484 in the control group, with zero affected participants reported in each. For other adverse events, the same at-risk counts are listed, again with zero affected participants in both groups. The registry also reports zero deaths in both groups, and lists no serious-event terms.

These figures describe what was posted in the registry; they do not establish that no adverse events occurred outside the recorded reporting process. The safety at-risk counts differ from the primary outcome analysis counts, and the record does not explain the relationship between those populations. The safety observation period, event definitions, collection procedures, and completeness of ascertainment are not reported in the registry record.

Zero reported events in both groups means that the registry lists none for these categories. It does not by itself show that the groups have identical safety profiles or rule out uncommon events. The safety figures should be read as a summary of reported counts, not as a measure of comparative risk precise enough to exclude differences.

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

The posted primary results describe a comparison of self-reported COVID-19 vaccine uptake 6 months after the intervention began. Uptake was reported by 117 of 417 analyzed intervention participants and 123 of 459 analyzed control participants. The unadjusted risk-difference estimate is 0.013, with a confidence interval from -0.047 to 0.072 and p = 0.676. The interval spans zero, so the data do not provide a precise estimate of the direction of the group contrast.

This does not prove that the groups had the same uptake, and it does not establish that the intervention had no effect. A non-significant result is not a test of equivalence. Equivalence requires a prespecified threshold for what counts as a sufficiently small difference and an analysis designed to evaluate that question; such details are not reported in the registry record.

The endpoint is self-reported, and the registry record does not report whether uptake was independently verified. Nor does it provide enough detail to assess missing outcome data, adherence, participant flow, or the analysis population principle. The posted results therefore answer a narrower question than whether the intervention would have the same effect under all implementation conditions or in other populations.

The study's stated community focus and four-county setting are important context for interpreting its intended reach. The results should not automatically be generalized beyond the studied setting, especially because the registry record does not report detailed subgroup results or other setting-specific implementation outcomes. The secondary endpoint measures high intention among participants who were not vaccinated; it is related to the study's aims but is not the same as uptake.

The registry also includes a separate Chi-squared analysis entry with an odds-ratio value of 0.05 and p < .05, but its relationship to the posted arm-level comparisons is not clear. That entry and the primary unadjusted comparison should not be treated as interchangeable. A fuller explanation of the analysis and its outcome mapping is not reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison should be read through both its estimate and its uncertainty. The primary point estimate alone gives one summary of the observed contrast, while the confidence interval shows that a broad range of values remains compatible with the data. The interval crosses zero, so a single point estimate cannot establish a clear direction of difference.

It also shows why event counts must be interpreted in relation to group size. The raw counts are 117 and 123, but the analyzed groups have different numbers of participants. The risk difference and other effect measures provide contrasts that account for the group denominators. Each measure answers a slightly different descriptive question, and estimates should be paired with their uncertainty.

Trial reporting matters as much as the calculations. The registry identifies the randomized parallel design, intervention, comparator, endpoint, and analyzed counts, but does not report important details about missing data, the analysis population, or the original sample-size justification. In addition, a Chi-squared analysis entry is not clearly connected to the arm-level contrasts. Such gaps make it harder for readers to understand how the posted results were generated and how they should be weighted.

Finally, planning illustrations based on an observed contrast can look demanding when the observed effect is small. They are useful for understanding how power, effect size, and sample size relate under stated assumptions, but they do not describe the trial's original design or its achieved probability of detecting an effect. Separating planning calculations from observed results avoids treating a retrospective illustration as evidence about trial power.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record offers limited detail for evaluating the conduct and interpretation of this trial. Although total enrollment is 1180, the primary analyzed groups are 417 and 459 participants. Participant flow and reasons for the difference are not reported in the registry record, so the extent and handling of missing primary outcomes cannot be assessed.

The primary endpoint is participant-reported vaccine uptake. Whether self-report was confirmed by another source, how responses were collected, and how incomplete or inconsistent responses were classified are not reported in the registry record. These measurement details matter because the endpoint relies on participant reporting.

The study is listed as randomized and unmasked, but allocation concealment and implementation details are not reported. The registry describes the intervention and control content at a broad level, yet message timing, message frequency, adherence, exposure to other information, and fidelity of delivery are not described in the posted record. It is not possible to determine from the registry record how closely participants received the intended intervention.

The displayed comparisons are unadjusted. The registry record does not report a full analysis plan, covariate adjustment, subgroup analyses, sensitivity analyses, or an approach to multiplicity. It also does not explain the relationship of the Chi-squared odds-ratio entry to the arm-level primary result. The original sample-size justification and design assumptions are not reported; the planning illustration is based on the observed effect and stated assumptions instead.

The secondary endpoint applies only to participants who were not vaccinated, and its analysis counts are smaller than those of the primary endpoint. The record does not give further details about selection into that subset or missing secondary responses. These limitations restrict interpretation of both the size and generalizability of the posted comparisons, without establishing that any particular bias occurred.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Getting to Yes, Michigan! (G2YMI) — NCT05096260.
• Registry record last updated 2026-05-01; status last verified 2026-04.
• Results first posted 2026-05-01.
• 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.

22. Record Summary

Getting to Yes, Michigan! (G2YMI) is a completed randomized, parallel-group study sponsored by the University of Michigan, with enrollment of 1180. It focused on community-centered approaches to COVID-19 vaccine uptake, including motivational interviewing-based SMS/MMS messages compared with a simple website and later access to the messages.

The primary endpoint was participant-reported COVID-19 vaccine uptake 6 months after the start of the intervention. The posted unadjusted risk difference is 0.013, with a 95% confidence interval from -0.047 to 0.072 and p = 0.676. The interval includes zero and does not precisely establish the direction of the contrast. One secondary endpoint concerns high intention to take up vaccine among participants who were not vaccinated; its unadjusted confidence interval also spans zero.

The record lists zero serious adverse events, zero other adverse events, and zero deaths in both groups. Details about missing data, analysis populations, multiplicity, the original sample-size justification, and the relationship of a separate Chi-squared analysis entry to the arm-level results are not reported in the registry record.