Providing Financial Incentives to Improve Adherence to Referral Eye Care Visits: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Providing Financial Incentives to Improve Adherence to Referral Eye Care Visits (NCT04328207), a non-randomized na study. The primary outcome is Percentage of Participants That Adhere to Referral Appointment.

2. Clinical Question

The practical question is whether attendance differs between these two approaches to encouraging follow-up after screening. The endpoint is about completing a referral appointment, not about a diagnosis, change in vision, or management outcome. The comparison therefore addresses a step in the care pathway: whether a participant who requires referral attends the appointment within the stated period.

The registry record describes the study as using telemedicine to prevent blindness in an at-risk rural Alabama population and identifies glaucoma detection and management as a central motivation. It also lists other eye conditions. The primary outcome, however, is referral-appointment adherence across the study population; the registry record does not report condition-specific attendance results.

Because allocation is nonrandomized, the comparison does not by itself isolate the effect of the incentive. Differences between groups might reflect differences in who entered each group as well as the incentive and education. The registry record does not describe how participants were assigned to groups or report an adjustment that addresses potential baseline differences.

3. Trial Design

The registry record describes a parallel, nonrandomized study with double masking. It records enrollment of 900 and a phase of NA. The study status is completed. Its recorded start date is 2020-11-18, and its primary completion date is 2024-07-01. These dates describe the registry record; they do not supply additional details about recruitment, follow-up conduct, or the timing of analyses.

The parallel groups differ in the incentive offered for completing a referral appointment. The financial-incentive group receives eye health education and an incentive after a referral visit is completed, if a referral was required. The no-financial-incentive group receives standard-of-care eye health education alone. The registry record does not report the incentive amount, the form of payment, or further details about how education was delivered.

Although masking is recorded as double, the registry record does not identify which people were masked or explain how masking was maintained. It also does not report the method used to allocate participants. The nonrandomized allocation is important: unlike random assignment, it does not ensure that measured and unmeasured participant characteristics are balanced between groups. As a result, an observed difference in attendance would need to be interpreted as an association between group membership and attendance, unless a suitable analysis and its assumptions support a more specific interpretation.

The listed outcome is binary: a participant either attended the referral appointment within the stated timeframe or did not. The registry record gives the timeframe as screening through six months after screening. It does not report a detailed protocol for handling multiple referrals, appointments outside that period, or participants for whom no referral was required.

The sponsor is the University of Alabama at Birmingham. The registry record describes one outcome measure and no posted analyses. It does not report a statistical analysis plan, sample-size justification for the study itself, or details of any interim analyses. The planning illustration described for the posted calculations is not evidence about how the trial was originally designed.

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 Percentage of Participants That Adhere to Referral Appointment, measured over The time frame will be whether they attended the referral appointment between the screening date until 6 months after the screening date..

The primary endpoint is the percentage of participants who adhere to attending a referral appointment. Attendance is assessed from the screening date through six months after screening. The registry description compares participants receiving financial incentives and eye health education with participants receiving standard-of-care education alone.

This is a binary endpoint at the participant level: the outcome is whether the referral appointment was attended within the stated window. Such an endpoint can express completion of a care step in a form that is easy to interpret as a proportion of participants. It does not indicate how soon an appointment occurred, whether the referral led to a diagnosis, or whether subsequent care was completed.

The endpoint is relevant to the study's stated goal of improving access to detection and management, because attending a referral appointment is part of that pathway. It remains a process measure rather than a direct measure of vision, disease progression, or blindness prevention. The registry record does not report whether every participant required a referral or how participants without a required referral are treated in the endpoint denominator. That definition matters because the endpoint's percentage depends on who is counted as eligible for referral and how missing attendance information is handled.

No observed attendance percentage or group comparison is reported in the registry record. The endpoint definition alone therefore does not establish whether attendance differed between groups or how precisely any difference was estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas are relevant to interpreting a binary attendance outcome and a nonrandomized group comparison. The registry record does not report the analysis used for this endpoint. The explanations below are general and should not be read as a description of an unreported analysis.

What does a binary endpoint represent?

For a binary endpoint, each participant is classified according to whether the defined event occurred. In this study, the event is attending a referral appointment during the period from screening through six months after screening. A group percentage is formed from the number meeting the definition and the number included in its denominator. The denominator must be defined consistently: for example, the analysis needs to address people who did not require a referral and people whose attendance status is unknown. The registry record does not report the outcome counts, denominators, or detailed rules for those situations.

A percentage describes event frequency within a group, but a comparison requires a clearly stated contrast. Researchers may compare percentages directly or use a model that estimates a contrast while adjusting for recorded characteristics. Those approaches can answer related but not identical questions, and adjusted estimates depend on the variables and model assumptions. The registry record does not report which approach was used.

Why does nonrandomized allocation matter?

In a randomized study, allocation by chance is intended to make groups comparable on average, including for characteristics that are not measured. Nonrandomized allocation does not offer that protection. Participants receiving an incentive may differ from those receiving education alone for reasons related to access, referral need, motivation, or other factors. These are general examples of possible sources of confounding, not claims about what occurred in this study. The registry record does not describe the allocation mechanism or report baseline characteristics that would clarify comparability.

Statistical adjustment can account for measured characteristics when the data and model support it, but it cannot guarantee removal of bias from unmeasured factors. An adjusted association also does not automatically become causal. Interpretation depends on the assignment process, the quality of measurement, the selected adjustment variables, and the assumptions of the analysis. No adjusted effect estimate or corresponding methods are reported in the registry record.

How should uncertainty be understood?

In general, a confidence interval conveys uncertainty around an estimated contrast. A narrow interval indicates greater precision under the analysis assumptions; a wide interval indicates less precision. Its interpretation depends on the statistical model and the data. A confidence interval is not a guarantee that the true effect lies within its limits, and it does not resolve confounding from nonrandomized allocation. No confidence interval for referral attendance is reported in the registry record.

A significance test evaluates data against a stated null hypothesis under a statistical model. A p-value does not tell readers the probability that the null hypothesis is true, nor does it show whether an effect is large enough to matter. Interpretation should consider the estimate and its uncertainty, the study design, and the relevance of the endpoint. The registry record reports no test statistic or p-value for the primary outcome.

What do the planning assumptions mean?

The planning illustration is described as using a two-sided significance level of 0.05, equal allocation between groups, and a normal approximation. It expresses the effect to detect as a stated fraction of the observed effect. In general, a sample-size calculation asks how much information a proposed study would need under selected assumptions to detect a chosen effect with a selected operating characteristic. Its output is a planning quantity, not a result observed in participants.

The record's planning description explicitly distinguishes these calculations from the trial's own design. The registry record does not report the original sample-size justification or numerical planning outputs. Nor does the description establish that the study's actual group sizes were equal. Planning estimates based on an observed effect are sensitive to the estimate used; if an effect estimate is unavailable or imprecise, the resulting illustration should not be treated as a stable forecast of future performance.

Power is a design concept describing the probability of detecting a specified effect under assumed conditions. It is not appropriately inferred after the fact from whether a result is statistically significant. The sample-size illustrations are not observed power. Since no primary result is reported in the registry record, the attendance effect, its uncertainty, and its relation to the planning assumptions cannot be evaluated from this record.

What analytical details are absent?

The registry record does not report the model, adjustment variables, confidence intervals, hypothesis tests, or a missing-data approach for the primary endpoint. It also does not report how the analysis population was defined, whether the analysis followed group assignment or treatment received, or how participants without a required referral were handled. These choices affect both the meaning and precision of an estimated group comparison. Without those details and the outcome data, a reader cannot reproduce or fully assess a primary analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("No Financial Incentive", "Financial Incentive")
n      <- c(499, 401)
events <- c(80, 92)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  No Financial Incentive  499  80  16.0  12.9  19.6  13.1  19.5  13.1  19.5
ARM  Financial Incentive  401  92  22.9  18.9  27.4  19.1  27.3  19.1  27.3

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists one outcome measure and no analyses. It does not report a statistical analysis plan, the number of planned comparisons, or a strategy for controlling multiplicity. Multiplicity arises when several outcomes, time points, subgroup comparisons, or statistical tests are examined; repeated testing can increase the chance of finding an apparently notable result by chance. That is a general concern, not evidence that multiple tests were conducted in this study. With no analyses posted in the registry record, the approach to multiplicity cannot be assessed.

Missing data are also important for an attendance endpoint. A participant's attendance status may be unknown, and some participants may not require a referral. The registry record does not state how either circumstance was handled, whether missingness occurred, or whether any sensitivity analyses were performed. In general, excluding participants with unknown outcomes can bias a comparison if the reasons for missingness differ between groups or are related to attendance. Treating unknown status as nonattendance is a different assumption and should be stated explicitly.

The analysis population is not described. The registry record does not say which participants were included in the primary analysis or how group membership was used in that analysis. This omission matters in a nonrandomized study because the assignment process and the analysis population affect what contrast is being estimated. No participant flow, outcome denominator, or primary analysis result is reported in the registry record. As a result, the extent of missingness, the handling of eligibility, and the robustness of the comparison remain unknown.

12. 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 safety figures are reported separately for the no-financial-incentive and financial-incentive groups. For the no-financial-incentive group, the record lists 499 at risk for serious adverse events and 0 affected; it also lists 499 at risk for other adverse events and 0 affected, with deaths recorded as 0. For the financial-incentive group, the corresponding figures are 401 at risk for serious adverse events and 0 affected; 401 at risk for other adverse events and 0 affected, with deaths recorded as 0.

These are registry counts, not evidence that adverse events were impossible or that the interventions have no risk. The record does not report event terms, because no serious adverse terms are listed, and it does not explain the collection, monitoring, attribution, or reporting procedures. The at-risk counts describe the safety reporting entries and should not be mistaken for the primary attendance denominators.

Zero recorded events means that none were listed as affecting participants in these categories in the posted safety data. It does not establish that all possible harms were assessed equally or that the study could rule out uncommon events. Safety interpretation depends on what was actively monitored, the duration and completeness of observation, and how events were classified. Those details are not reported in the registry record.

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

The study addresses whether a financial incentive, alongside eye health education, is associated with referral appointment attendance compared with education alone. Its endpoint focuses on a practical step between screening and further eye care, with attendance assessed through six months after screening. That makes the question relevant to access and follow-up, but it is not a direct measure of vision outcomes or prevention of blindness.

The registry record does not report attendance results. It therefore does not show whether the groups had different attendance, which group had a higher percentage, or how precisely any difference was estimated. The completed status and recorded enrollment of 900 do not fill that gap. Without the event counts, denominators, effect estimate, and uncertainty, the magnitude and direction of the primary comparison cannot be characterized.

Even if a difference were available, the nonrandomized allocation would require caution. Group membership may be associated with characteristics that also affect the likelihood of attending a referral appointment. A raw comparison could therefore combine any incentive-associated difference with pre-existing group differences. The registry record does not report an adjustment strategy or evidence that these concerns were addressed.

The outcome also has a defined scope. Attendance within the stated interval is not the same as receiving a diagnosis, completing treatment, or preventing vision loss. The record does not report condition-specific results or subsequent clinical outcomes. Thus, the study's stated goal and the listed primary endpoint are connected, but they are not interchangeable.

The safety entries report zero affected participants for the listed serious and other adverse-event categories in both groups. Those entries are useful as recorded safety information, but they do not prove an absence of harm or replace details about monitoring. Overall, the registry record describes the intervention contrast and outcome question, while leaving the central attendance result and its statistical uncertainty unreported.

14. Why This Trial Matters Statistically

This trial raises a statistical question about a measurable step in eye care: whether a financial incentive is associated with attending a referral appointment after screening. That endpoint can help distinguish the goal of improving follow-up from broader goals such as disease detection, management, or prevention of vision loss. Keeping those outcomes separate is important when interpreting a study's claims.

The nonrandomized design makes the comparison especially instructive. A group difference is not automatically attributable to the intervention when people are not assigned by chance. Researchers and readers need to consider how groups were formed, whether relevant baseline characteristics were measured, and whether the analysis handled those characteristics appropriately. The registry record does not report these details or a primary analysis, so it illustrates why design information and results must be read together.

The record also highlights the distinction between a completed study and a fully interpretable statistical result. Enrollment and completion dates describe trial conduct, but they do not show what happened on the primary endpoint. For interpretation, event counts, denominators, effect estimates, uncertainty, and analysis details are needed. Without these, neither a difference nor equivalence can be concluded.

Finally, a process endpoint can be meaningful without being a clinical outcome. Referral attendance may form part of a pathway to care, but conclusions about that step should not be extended to diagnoses, vision, or disease outcomes that are not reported. The statistical value of the study depends on keeping the endpoint's scope clear and distinguishing observed evidence from the study's broader motivation.

15. 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 most important limitation for interpretation is that the registry record does not report the primary outcome results. Attendance counts, percentages, denominators, the group contrast, and its precision are absent. Consequently, the primary question cannot be answered from the record. The lack of a reported analysis is distinct from a finding of no difference.

The nonrandomized allocation limits causal inference. The record does not describe how participants entered the groups, what baseline characteristics were measured, or whether any statistical adjustment was performed. Differences in group composition could influence attendance independently of the incentive. This concern cannot be resolved from the information recorded.

The endpoint's denominator and missing-data rules are not reported. The registry describes attendance after screening if a referral appointment was required, but does not explain how participants without a referral or participants with unknown attendance status were included. Those decisions could affect the reported percentage and the comparison.

Other details needed for a full assessment are also not reported in the registry record: who was masked, how masking was maintained, the statistical model, prespecified adjustment variables, the analysis population, and any sensitivity analyses. The record does not report the original sample-size justification or the numerical planning outputs. The planning assumptions described for the illustration should not be taken as a substitute for that missing design rationale.

The safety counts lack accompanying information about event collection and follow-up. Zero recorded events cannot establish that every relevant event was detected or that uncommon risks were excluded. In addition, the primary endpoint measures attendance rather than a subsequent clinical outcome, so it cannot by itself establish effects on diagnosis, management, or vision.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Providing Financial Incentives to Improve Adherence to Referral Eye Care Visits — NCT04328207.
• Registry record last updated 2025-08-07; status last verified 2025-07.
• Results first posted 2025-07-25.
• 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.

19. Record Summary

The registry record describes a completed, parallel, nonrandomized study with double masking and recorded enrollment of 900. The study compares financial incentives plus eye health education with standard-of-care eye health education alone. The primary outcome is the percentage of participants attending a referral appointment from screening through six months after screening.

The record lists glaucoma, diabetic retinopathy, refractive errors, cataract, and behavior, health among the conditions. It identifies the University of Alabama at Birmingham as sponsor and records the study start date as 2020-11-18 and primary completion date as 2024-07-01.

No primary attendance results or analyses are reported in the registry record. The safety entries list zero serious adverse events, zero other adverse events, and zero deaths in each group. The allocation method, primary analysis details, missing-data handling, and original sample-size justification are not reported in the registry record.