Overcoming Barriers and Obstacles to Adopting Diabetes Devices (ONBOARD) Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Overcoming Barriers and Obstacles to Adopting Diabetes Devices (ONBOARD) Trial (NCT04672655), a randomized na study. The primary outcome is Change in Hemoglobin A1c (HbA1c) Over Time.

2. Clinical Question

3. Trial Design

ONBOARD was a randomized, parallel-group trial with no masking reported. The registry lists the study as completed and gives enrollment of 150. The listed phase is NA. The study began on 2021-02-17, and the primary completion date is 2024-09-16. Those dates describe the registry record; they do not establish when every participant entered or finished follow-up.

The trial compared a behavioral intervention package with a CGM Only condition. Participants assigned to ONBOARD received the intervention and 12 weeks of continuous glucose monitoring supplies. The CGM Only group received supplies without the ONBOARD intervention during their 12-month participation. The descriptions say that participants in both groups provided HbA1c values, data downloads, and survey responses at various time points. The intervention description also identifies four 60-minute sessions with a study interventionist, but the available text ends before giving further detail about the intervention.

The primary measure was change in HbA1c over time, assessed at baseline, month 3, month 6, and month 12. The primary outcome population is described as participants with data at each respective time point. Secondary outcomes listed in the record were change in time in glucose target range, percent CGM wear time, diabetes distress measured with the T1-DDS, and satisfaction measured with the GMSS-T1D.

The registry record does not report the randomization method, concealment procedures, site details, or the statistical analysis plan. It also does not report whether the groups were balanced at baseline or how many participants completed each assessment. Those omissions limit reconstruction of trial conduct and analysis from the registry record alone.

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: 150 patients signed informed consent; of these, 138 completed the baseline survey, and 135 were randomized.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants with data at each respective time point

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 who met eligibility criteria

6. Primary Endpoint

The registry lists the primary outcome as Change in Hemoglobin A1c (HbA1c) Over Time, measured over baseline, month 3, month 6, month 12.

Hemoglobin A1c is a measure of glycated hemoglobin. It reflects glucose exposure over time and is used in clinical research as a continuous indicator of glycemic status. The primary outcome is defined as change in HbA1c over time, expressed as a percentage of glycated hemoglobin, with assessments at baseline, month 3, month 6, and month 12.

A continuous measure can retain information across its range rather than dividing participants into categories. Repeated assessments can describe the outcome's course during follow-up, rather than relying only on a single measurement. Comparing patterns over time can therefore address whether observed changes differ between randomized groups. That is a general explanation of the rationale for longitudinal outcome measurement, not a claim about the findings of ONBOARD.

The registry specifies least-squares means and 95% confidence intervals. In general, a least-squares mean is a model-based estimate; the exact model and any adjustments used for this trial are not reported in the registry record. A confidence interval communicates uncertainty around an estimate under the analysis assumptions. The record does not give the primary outcome estimates, confidence interval bounds, or a numerical contrast between groups. It therefore does not allow the size or precision of any observed change to be assessed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, parallel-group comparison with no masking reported. Randomization is intended, in general, to reduce systematic differences between groups at the point of assignment. It does not guarantee that every measured characteristic will be balanced, nor does it by itself resolve missing outcome data or establish that an analysis followed the intended assignment. Baseline characteristics and the allocation procedure are not reported in the registry record.

What does a least-squares mean represent?

For a continuous outcome, a least-squares mean is generally an estimate derived from a statistical model. Depending on the model, it may account for factors or repeated measurements and can differ from a simple arithmetic average. The registry labels the primary outcome statistic as a least-squares mean but does not describe the model, covariates, or estimation details. It is therefore not possible to tell from the registry record exactly how the estimates were obtained or which comparisons they represent.

The outcome is measured repeatedly at baseline, month 3, month 6, and month 12. In general, observations from the same participant are likely to be related, so treating every observation as independent can misstate uncertainty. Longitudinal models can represent within-participant association and use the available repeated measurements under their assumptions. The particular method used in ONBOARD, its assumptions, and any sensitivity analyses are not reported in the registry record.

How should a 95% confidence interval be read?

In general, a confidence interval gives a range of values compatible with the observed estimate under a specified statistical procedure and its assumptions. A narrower interval indicates greater precision than a wider interval, while an interval alone does not establish whether an effect is important to participants or clinically meaningful. The registry specifies 95% confidence intervals for the primary outcome, but no interval bounds are reported. Its precision therefore cannot be evaluated from the available record.

A confidence interval for a group mean is not automatically a confidence interval for the difference between groups. To assess the comparative question, readers need an estimate of the relevant between-group contrast and its uncertainty, with the time point or longitudinal contrast clearly defined. Those details are not reported for the primary outcome. General interpretation of confidence intervals cannot substitute for the missing trial estimates.

Why do time points and multiple outcomes matter?

When a study measures outcomes at several times, the analysis may estimate separate group means at each time, compare groups at a selected time, or test a difference in trajectories. These answer related but distinct questions. Testing several contrasts can also increase the chance of a false-positive finding if multiplicity is not addressed. General approaches include defining a primary contrast in advance or adjusting inference across comparisons. The registry record does not report which approach, if any, was used.

The record lists five outcome measures and five analyses, with HbA1c as the primary outcome and several secondary outcomes. These counts do not reveal the number of statistical contrasts, whether outcomes were tested in a hierarchy, or whether multiplicity adjustments were applied. The analysis plan and any interim analyses are not reported in the registry record.

What can a planning illustration tell readers?

In general, a sample-size calculation describes the design conditions under which a future study is planned to detect an effect. The registry's stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is specified as a fraction of the observed effect. The record characterizes this calculation as a planning illustration computed from the observed effect, not the trial's own design. Such an illustration is not observed power and does not demonstrate that the trial was adequately powered. No numerical planning estimates are reported in the available information for this analysis.

For continuous outcomes, the amount of information depends on factors such as variability, the size of the contrast, and the analysis design. For repeated measures, within-participant association and the pattern of missing observations can also matter. Those are general statistical considerations; the relevant trial-specific values and assumptions beyond those stated above are not reported in the registry record.

Recalculation from posted summary statistics would require those statistics and a clearly defined comparison. The registry record does not report the primary estimates needed for such a recalculation. No independent numerical effect estimate or recalculated test can therefore be stated.

10. Multiplicity, Missing Data and Analysis Populations

The primary HbA1c outcome is measured at baseline, month 3, month 6, and month 12. The record also lists secondary outcomes for time in glucose target range, percent CGM wear time, diabetes distress, and system satisfaction. Several outcomes and time points can create multiple opportunities for statistical comparisons. In general, multiplicity procedures or a prespecified hierarchy can help control how those comparisons are interpreted. The registry record does not report whether such procedures were planned or used, or how the primary comparison was defined.

The record lists five outcome measures and five analyses, but those counts do not establish how many hypotheses were tested. It does not report whether there were additional contrasts, subgroup analyses, interim analyses, or adjustments for multiplicity. These analysis details are not reported in the registry record.

The primary outcome population is described as participants with data at each respective time point. This indicates that the reported outcome concerns participants contributing data to the relevant assessment, but the record does not specify the number contributing at each time point. It also does not explain how missing measurements were handled, whether participants with incomplete follow-up contributed to a model, or whether imputation or sensitivity analyses were used. The reason data may be missing and any assumptions about missingness are not reported in the registry record.

These omissions matter because missing observations can affect precision and, depending on the pattern and analysis assumptions, can affect estimates. The registry information is not enough to determine whether the missing-data approach was appropriate or how robust the results were to alternative assumptions. Without the analysis plan and numerical outcomes, neither multiplicity nor missing-data methods can be assessed in detail.

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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: diabetic ketoacidosis (1.5%), vomiting (1.5%), appendicitis (1.5%), hypoglycemic episode (1.5%).

The registry reports serious adverse event figures for the ONBOARD Intervention Group and the CGM Only Group. The serious-event populations at risk are 68 and 67, respectively; the affected counts are 4 and 1. The record lists zero deaths in each group. For other adverse events, the populations at risk are 68 and 67, and the affected counts are zero in both groups. These are registry counts; they do not by themselves describe the timing, duration, severity beyond the serious classification, or relationship of events to the intervention.

The listed serious terms are Diabetic Ketoacidosis, Vomiting, Appendicitis, and Hypoglycemic Episode, each reported as 1.5%. The registry entry does not identify the group associated with each term in the information available for this analysis. It also does not clarify whether the listed terms are exhaustive or whether a participant could contribute more than one event term. These percentages should therefore be read as reported, without inferring a group-specific rate or a total number of events.

In general, adverse-event counts require careful interpretation alongside definitions, follow-up, exposure, and ascertainment methods. Those details are not reported in the registry record. The figures alone do not establish causation or support a comparison of safety risks.

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

The registry record establishes that ONBOARD was a randomized, parallel-group study of a behavioral intervention package compared with CGM supplies alone in adults with type 1 diabetes. It defines HbA1c change over time as the primary outcome and lists several secondary measures. It also reports serious and other adverse-event counts. These registry entries describe the study question, design, and selected safety information.

The record does not report numerical primary outcome estimates or a between-group contrast. As a result, it cannot show the direction or size of any HbA1c difference, how precisely such a difference was estimated, or whether the groups differed statistically. No conclusion about the effect of the intervention on HbA1c can be drawn from the outcome information. The same caution applies to the listed secondary outcomes, for which numerical findings are not reported.

Even if a numerical result were available, statistical evidence and practical importance would be separate considerations. In general, an estimate should be interpreted with its uncertainty, the outcome's clinical context, the analysis population, and limitations in study conduct and follow-up. The registry record does not provide enough outcome detail for that assessment.

The safety counts are descriptive. They do not show whether an event was caused by the intervention, and the listed event terms are not assigned to a group in the available record. Zero deaths and zero affected counts for other adverse events should be understood as the recorded figures, not as proof that no unrecorded event occurred or as a complete safety assessment.

Accordingly, the most defensible interpretation is limited: the trial tested a behavioral package alongside CGM supplies against supplies alone, but the registry information available for this analysis does not report the primary comparative results needed to characterize its outcome findings.

13. Why This Trial Matters Statistically

ONBOARD addresses a question about a behavioral intervention intended to help adults with type 1 diabetes manage barriers associated with continuous glucose monitoring and increase readiness for closed loop. The design compares that intervention plus CGM supplies with supplies alone. From a statistical perspective, this makes the distinction between the intervention package and the shared supply component central to interpreting the randomized comparison.

The primary outcome is continuous and measured at multiple time points. Such a design raises important questions about how change is defined, how repeated observations are modeled, and which time-specific or overall contrast answers the main question. Those are broadly relevant issues in longitudinal clinical research. The registry identifies the schedule and the use of least-squares means with 95% confidence intervals, but does not report enough analysis detail to resolve them for this trial.

The record also lists patient-relevant measures beyond HbA1c, including time in glucose target range, device wear time, diabetes distress, and system satisfaction. Including more than one type of outcome can give a broader picture of a behavioral intervention's intended effects. It also makes prespecification, multiplicity, missing-data handling, and clear reporting important. The registry's available information does not explain those analytic choices or provide numerical results for the listed outcomes.

For readers assessing evidence, a completed randomized trial is most informative when its estimates, uncertainty, analysis population, and methods are transparently reported. The absence of numerical primary results in the registry record limits what can be learned about comparative effects. That limitation is itself statistically important: design information can frame a question, but it cannot replace outcome estimates.

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.

The primary limitation for interpreting the trial from the registry record is that numerical HbA1c estimates and a between-group comparison are not reported. Although the outcome is defined as change over time and the record names least-squares means with 95% confidence intervals, there are no estimates or interval bounds to evaluate. Numerical results for the listed secondary outcomes are also not reported in the information available for this analysis.

The primary outcome population is described as participants with data at each respective time point, but the number contributing at each assessment is not reported. The registry record does not explain how missing measurements were handled, what assumptions were used, or whether sensitivity analyses were conducted. These details are important for understanding both the analysis population and the reliability of longitudinal estimates.

Important design and analysis details are also absent from the available record. The method of randomization, concealment, baseline group characteristics, statistical model, covariate adjustment, primary contrast, and multiplicity approach are not reported. The record does not report interim analyses or a statistical analysis plan. Without these details, readers cannot fully assess whether the analysis matches the trial question or how robust its conclusions may be.

The intervention description available for this analysis is incomplete, ending after the reference to sessions with a study interventionist. The safety terms are listed without a group assignment, and the record does not provide further context about event ascertainment or attribution. These limits make it difficult to evaluate intervention delivery and interpret safety information beyond the reported counts.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Overcoming Barriers and Obstacles to Adopting Diabetes Devices (ONBOARD) Trial — NCT04672655.
• Registry record last updated 2025-09-05; status last verified 2025-08.
• Results first posted 2025-09-05.
• 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

NCT04672655 is the completed, randomized, parallel-group ONBOARD trial in adults with type 1 diabetes. The registry lists enrollment of 150 and no masking. The intervention group received the ONBOARD behavioral package and 12 weeks of CGM supplies; the CGM Only group received supplies without the intervention during their 12-month participation.

The primary outcome was change in HbA1c at baseline, month 3, month 6, and month 12, described using least-squares means and 95% confidence intervals. The registry identifies the primary outcome population as participants with data at each respective time point, but does not report numerical primary results or a between-group estimate. Secondary outcomes include time in glucose target range, CGM wear time, diabetes distress, and system satisfaction.

Serious adverse events affected 4 participants in the intervention group and 1 in the CGM Only group; deaths were zero in both groups. Numerical outcome findings and important analysis details are not reported in the registry record.