Effects of a Patient Portal Intervention to Address Diabetes Care Gaps: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of a Patient Portal Intervention to Address Diabetes Care Gaps (NCT04894903), a randomized na study. The primary outcome is Number of Participants With Diabetes Care Gaps at Baseline, 3 Months, 6 Months, and 12 Months.

2. Clinical Question

The care gaps were defined using four types of monitoring or preventive care: a diabetes eye exam, a hemoglobin A1C blood test, urine microalbumin testing, and pneumococcal vaccination. The registry describes the measure as the number of gaps per patient out of four possible, while the primary outcome title refers to the number of participants with diabetes care gaps. The registry record does not clarify further how those descriptions map to the reported analysis.

The practical research question is whether adding patient-facing notifications and the option to initiate orders changes these measured care gaps relative to usual portal access. The outcome is based on electronic health record abstraction, rather than solely on participants’ reports of their care. The study also lists secondary outcomes concerning orders, understanding, confidence, distress, satisfaction or usability, blood glucose control, treatment intensification, and diabetes self-efficacy; results for those outcomes are not reported in the registry record.

3. Trial Design

The trial is described as randomized, with a parallel-group design and no masking. Participants were assigned to the Diabetes Care Gaps Patient Portal Intervention or usual care. In the intervention arm, the existing My Health at Vanderbilt portal included the intervention; in the usual-care arm, participants had portal access without the intervention. The registry does not report how randomization was generated, whether allocation was concealed, or whether anyone involved in outcome assessment was masked.

The registry lists enrollment as 440 and the trial status as completed. The recorded study start was 2022-04-30 and the primary completion date was 2024-02-03. The study concerns diabetes mellitus, and its phase is listed as NA. The sponsor is Vanderbilt University Medical Center. The registry record does not report additional details about recruitment, the number assigned to each arm, or the number who completed each follow-up assessment.

The primary outcome is assessed at baseline and at 3-month, 6-month, and 12-month follow-up. The registry defines the care gaps using four measures of diabetes monitoring or preventive care, with their presence assessed by electronic health record abstraction. It also lists multiple secondary outcomes, including patient-initiated orders and changes in several patient-reported or clinical measures.

The record mentions a mixed-effects regression analysis and says that this analysis requires at least one measurement in the follow-up period. It does not identify the model specification, the effect measure, covariate adjustments, or the approach used to account for repeated measurements. The detailed statistical analysis plan, including any interim analyses or sample-size justification, is 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.

Before assignment: 440 patients were enrolled and sent the baseline survey. 433 patients completed the baseline survey. Completion of the baseline survey was required prior randomization and assignment to a study arm.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Two participants in the control group were excluded from analysis because they dropped out from the study prior to 3-month data collection. The mixed effects regression analysis requires at least one measurement in the follow-up period. Two individuals dropped out prior the any follow-up data collection and therefore are not included in the mixed effects regression analysis.

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: Two participants in the control group were excluded from analysis because they dropped out from the study prior to 3-month data collection. The mixed effects regression analysis requires at least one measurement in the follow-up period. Two individuals dropped out prior the any follow-up data collection and therefore are not included in the study analysis.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Diabetes Care Gaps at Baseline, 3 Months, 6 Months, and 12 Months, measured over Baseline, 3-month follow-up, 6-month follow-up, and 12-month follow-up.

The primary outcome concerns diabetes care gaps at baseline and at 3-month, 6-month, and 12-month follow-up. The registry describes four possible gaps: no diabetes eye exam in the last 12 months, no hemoglobin A1C blood test in the last 6 months, no urine microalbumin in the last 12 months, and no pneumococcal vaccination of any kind among the listed vaccine types. The presence of a gap is assessed by electronic health record abstraction.

The primary outcome title refers to the number of participants with diabetes care gaps, while its description says that the number of gaps per patient is counted out of four possible. The registry record does not explain whether the analysis uses a participant-level count, a count of gaps, or another summary of these measures. That distinction matters when interpreting a result: a count of participants with any gap is not the same quantity as the number of gaps per participant.

Tracking these measures over time can describe whether recorded monitoring and preventive care gaps persist or change during follow-up. Comparing the randomized groups can address whether their recorded patterns differ, subject to the analysis and missing-data approach. The endpoint does not by itself describe every aspect of diabetes care, and it does not establish why a particular gap was present. No primary endpoint estimates or results are reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Randomization is a design method for assigning participants to study groups by chance. In general, it is intended to make groups comparable at the start of a trial with respect to measured and unmeasured characteristics. It does not guarantee identical groups, especially in any one study. The registry describes this trial as randomized but does not report the randomization sequence method, allocation concealment, or any stratification factors.

What does a parallel-group design mean?

In a parallel-group trial, participants are assigned to one group and followed in that group. Here, the two groups are the intervention portal and usual portal access. The registry says masking was none. In general, lack of masking means participants or study personnel may know which intervention is being used. This can matter for behaviors, reporting, or other study procedures, but the record does not explain who knew assignments or how that could have affected outcome assessment. The primary outcome is described as being abstracted from electronic health records, but the record does not report whether abstractors were masked.

How should repeated follow-up measurements be understood?

The primary outcome is scheduled at baseline and at three follow-up times. Repeated measurements can describe how an outcome changes within participants and how patterns differ between groups over time. Measurements from the same participant are generally related rather than independent. A mixed-effects regression model is one general approach for analyzing repeated measurements while representing within-participant correlation. The registry mentions a mixed-effects regression analysis, but does not report its exact form, outcome distribution, link function, time coding, interaction terms, covariates, or method for estimating uncertainty. The model therefore cannot be reconstructed from the registry record.

The endpoint itself is a count-oriented measure: its description refers to the number of care gaps per patient, out of four possible. Count outcomes can have different distributions depending on their frequency and spread. General statistical methods may need to account for discreteness, clustering, or over-dispersion, which occurs when variability exceeds what a chosen model expects. The record does not report whether over-dispersion was assessed or which modeling assumptions were used. The title’s reference to the number of participants with gaps adds a separate ambiguity about the precise analyzed quantity.

What does the record say about the analysis population?

The registry notes that the mixed-effects regression analysis requires at least one measurement in the follow-up period. It says two control participants were excluded from analysis after dropping out before 3-month data collection, and also states that two individuals dropped out before any follow-up data collection. The record does not describe a full analysis population definition, such as whether all randomized participants were included whenever possible, or how participants with partial follow-up contributed to the model. It also does not report the amounts of missing data at each assessment or the assumptions used to handle missingness.

How should statistical significance and precision be assessed?

In general, a p-value quantifies how incompatible data are with a specified statistical model and null hypothesis; it does not measure the size or importance of an effect. A confidence interval communicates uncertainty around an estimate under the analysis assumptions. Neither is reported for the primary comparison in the registry record. Without an estimate, an interval, or a test result, statistical significance and precision cannot be evaluated.

The registry describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; the effect to detect is specified as a fraction of the observed effect. Such calculations are planning illustrations rather than observed trial outcomes, and they are not the trial’s own design justification. No numerical planning results or trial-specific sample-size justification are reported in the registry record. A planning calculation also cannot replace an analysis of the observed data or be interpreted as observed power.

Why do multiplicity and analysis details matter?

The record lists a primary outcome assessed at several times and multiple secondary outcomes. When many outcomes, time points, or comparisons are examined, the chance of at least one apparently notable result can increase. General approaches include prespecifying a primary contrast or adjusting inference across a family of tests. The registry record does not report a multiplicity strategy, identify a single primary time-point contrast, or state which analyses were prespecified. Those details are needed to understand how a set of results should be interpreted, but cannot be inferred from the list of outcomes alone.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measured at baseline and at 3-month, 6-month, and 12-month follow-up, along with multiple secondary outcomes. Each additional outcome, time point, or contrast can create another opportunity to observe a result that appears notable by chance. In general, multiplicity is addressed by prespecifying a primary comparison, controlling error rates across a defined family of tests, or clearly labeling additional analyses as exploratory. The registry record does not report which approach was used, whether a particular follow-up time was primary for inference, or how secondary outcomes were handled.

The registry states that two participants in the control group were excluded from analysis because they dropped out before 3-month data collection. It also states that the mixed-effects regression analysis requires at least one follow-up measurement and that two individuals dropped out before any follow-up data collection. It does not report missingness at each later assessment, reasons for missing data beyond these statements, or whether participants with partial follow-up were included. The record also does not report imputation, weighting, sensitivity analyses, or assumptions about the relationship between missingness and outcome values.

These details matter because analyses can be affected if participants with missing outcomes differ from those with observed outcomes. A mixed-effects model can use available repeated measurements under particular assumptions, but mentioning that model does not establish which assumptions were made or whether they are appropriate. The registry record does not provide enough information to assess the robustness of the primary analysis to missing data or multiplicity.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (0.9%), death (0.5%).

The registry reports serious adverse events for both groups. In usual care, 7 participants were listed as affected among 217 at risk; in the intervention group, 7 were listed as affected among 216 at risk. The record lists 0 other adverse events in each group. Deaths are listed as 1 in usual care and 0 in the intervention group. The serious-event terms with the largest listed percentages are hospitalization at 0.9% and death at 0.5%.

These are descriptive figures. The registry record does not provide a statistical comparison, confidence intervals, event-level details, or an assessment of whether events were related to the intervention. The record does not explain how the listed serious-event terms relate to the group counts, so they should not be treated as separate, additive categories or used to infer additional totals. The numbers at risk for these listings also should not be assumed to represent the number completing every primary outcome assessment.

Counts of serious events and deaths do not answer the primary question about diabetes care gaps. They also do not, by themselves, establish that an intervention caused or prevented an event. The figures describe what is recorded in the safety listing, with the limits of detail noted in the registry record.

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

The trial is designed to compare an intervention embedded in a patient portal with usual access to that portal. The intervention is intended to notify patients when selected diabetes monitoring and preventive care are due and permit them to initiate orders. Its primary outcome concerns care gaps defined from four measures and assessed using electronic health record abstraction at baseline and follow-up.

The registry record does not report primary outcome estimates, group results, confidence intervals, or p-values. It therefore does not establish the direction or magnitude of any difference between groups, or how precisely a difference may have been estimated. A completed trial status and an enrollment listing do not substitute for outcome results.

Even with reported results, a difference in the recorded care-gap measure would describe the outcome as defined and analyzed, rather than every dimension of diabetes care. The record’s wording leaves unclear whether the main quantity is the number of participants with gaps or the number of gaps per participant. Interpretation would depend on that operational definition, the analysis model, and how missing follow-up measurements were handled.

The trial’s secondary outcomes cover a range of measures, including patient-initiated orders, understanding, confidence, diabetes distress, satisfaction or usability, blood glucose control, treatment intensification, and self-efficacy. Their presence in the registry does not indicate that any particular result was observed. Results for these outcomes are not reported in the registry record.

Accordingly, the record supports a description of the research question, design, assessment schedule, and selected safety counts. It does not support a conclusion about the intervention’s effect on care gaps or other outcomes. The statistical interpretation remains incomplete without the posted outcome estimates and the analysis details needed to assess their precision and assumptions.

13. Why This Trial Matters Statistically

This trial asks a statistical question about a measurable care process: whether recorded gaps in selected diabetes monitoring and preventive care differ over time between randomized groups. Connecting the intervention to electronic health record abstraction gives the primary outcome an operational basis that is distinct from participant recollection alone. The repeated assessment schedule also raises an important analytic issue: measurements from the same participant over time are related, so an analysis must account for that structure.

The registry mentions mixed-effects regression, which is a broad class of methods often used for repeated measurements. But naming a model family is not enough to assess its implications. The exact outcome definition, model form, contrast, covariate adjustment, and missing-data assumptions all shape what an estimate means. The discrepancy between the primary outcome title and its description makes clarity about the analyzed quantity especially important.

The trial also lists several secondary outcomes and multiple follow-up times. That range can provide a broader picture of how a portal intervention is evaluated, while making prespecification and multiplicity handling important for interpreting statistical evidence. The registry record does not report those analysis details or the primary outcome results, so readers cannot assess the strength or precision of the central comparison.

The safety listings illustrate another statistical distinction: descriptive event counts are not equivalent to a comparative estimate, and they do not establish causation. Overall, this record highlights why trial interpretation depends not only on design labels but also on transparent outcome definitions, results, uncertainty measures, and analysis choices.

14. 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.

A central limitation of the available registry record is that it does not report primary outcome results or estimates for the secondary outcomes. No group-specific outcome values, between-group contrast, confidence interval, or p-value is reported. This prevents assessment of the observed effect, its precision, and the data’s compatibility with different possible effects.

The primary outcome is not fully clear from its wording. Its title refers to the number of participants with diabetes care gaps, while its description says the number of gaps per patient out of four possible. The record does not clarify which summary is analyzed or how the repeated measurements enter the model. The registry identifies electronic health record abstraction but does not report further details about measurement procedures or missing records.

Missing follow-up is acknowledged, including two control participants who dropped out before 3-month data collection and two individuals who dropped out before any follow-up data collection. The record does not provide follow-up completeness by time point, reasons for missingness beyond these statements, or sensitivity analyses. The mixed-effects regression analysis is mentioned, but its specification and missing-data assumptions are not reported.

The record also does not report the randomization method, allocation concealment, masking of outcome assessors, a multiplicity strategy, or a detailed statistical analysis plan. The number assigned to each arm and the number assessed at each time point are not given in the record. These omissions constrain evaluation of potential bias, model assumptions, and the reliability of any eventual comparison. They are limitations of the information available in the registry record, not proof that the study did not address these issues.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effects of a Patient Portal Intervention to Address Diabetes Care Gaps — NCT04894903.
• Registry record last updated 2025-04-09; status last verified 2025-03.
• Results first posted 2025-04-09.
• 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.

18. Record Summary

NCT04894903 is a completed, randomized, parallel-group trial sponsored by Vanderbilt University Medical Center. It compares usual access to the My Health at Vanderbilt portal with access to the portal containing the Diabetes Care Gaps Patient Portal Intervention. The registry lists enrollment as 440, no masking, a study start of 2022-04-30, and primary completion on 2024-02-03.

The primary outcome concerns diabetes care gaps measured at baseline and at 3-month, 6-month, and 12-month follow-up. The gaps are defined using four monitoring and preventive care measures, assessed through electronic health record abstraction. The record mentions mixed-effects regression and notes exclusions related to dropout before follow-up.

Primary and secondary outcome results, detailed model specifications, and a multiplicity strategy are not reported in the registry record. Safety listings report serious events, other events, and deaths by group, but do not provide a comparative analysis or establish causation.