A Research Study to Compare a New Weekly Insulin, Insulin Icodec Used With DoseGuide App, and Daily Insulins in People With Type 2 Diabetes Who Have Not Used Insulin Before: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Research Study to Compare a New Weekly Insulin, Insulin Icodec Used With DoseGuide App, and Daily Insulins in People With Type 2 Diabetes Who Have Not Used Insulin Before (NCT04760626), a randomized phase 3 study. The primary outcome is Change in Glycated Haemoglobin (HbA1c).

2. Clinical Question

The registry describes a randomized comparison in an insulin-naïve type 2 diabetes population and frames the study as taking place in a clinical practice setting. Random assignment was used to allocate participants to the weekly icodec strategy or a daily basal insulin analogue. The weekly strategy included DoseGuide-assisted dose recommendations, so the intervention being compared is not simply a different injection schedule: the app-supported titration approach is part of the intervention as described.

The primary outcome is a continuous measure: change in HbA1c from baseline to week 52, expressed as percentage of HbA1c. The reported group summaries are least-squares means, and the between-group result is also reported in an ANCOVA analysis labelled non-inferiority. The registry does not report the non-inferiority margin, so the margin-based decision cannot be independently reconstructed from the posted record alone. The question addressed by the outcomes is therefore about the measured HbA1c change and the reported analysis, not about every aspect of diabetes care or long-term outcomes.

3. Trial Design

This is a phase 3, randomized, parallel-group, open-label trial. The registry lists enrollment as 1085 and status as completed. Its title and summary identify the population as people with type 2 diabetes who had not used insulin before. Participants were assigned at random to insulin icodec with DoseGuide or to once-daily basal insulin analogue injections. Masking is listed as none, so the registry does not describe the trial as blinded.

The intervention group used once-weekly icodec with DoseGuide to guide titration. The comparator group received basal insulin analogue injections once daily. The listed comparator drugs are insulin glargine 100U/mL, insulin degludec, and insulin glargine 300U/mL. The record describes daily blood sugar measurement and electronic recording of safety data. Participants in the daily-insulin group recorded insulin doses in the electronic diary; those assigned weekly icodec recorded doses in DoseGuide.

The planned study duration is described as about 1 year and 2 months, with 8 planned clinic visits and additional visits as needed. Blood samples were planned at 6 clinic visits. The recorded study start is 2021-03-01, and primary completion is recorded as 2022-08-12. The registry does not report further detail about how the daily-insulin choices were allocated within the comparator group, the precise titration rules, or the handling of treatment changes beyond the summary description.

The trial’s randomized design supports a comparison of assigned treatment strategies, while the open-label design means participants and investigators were not masked according to the registry. The primary outcome is change in HbA1c between baseline and week 52. The registry also lists treatment discontinuation or intensification, treatment satisfaction, treatment-related impact, and hypoglycaemia outcomes among the secondary outcomes. The degree to which each of those outcomes was analyzed or adjusted for multiplicity 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set included all randomised participants. Overall number of participants analyzed = participants with available data for this outcome measure.

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: Full analysis set included all randomised participants.

6. Primary Endpoint

The registry lists the primary outcome as Change in Glycated Haemoglobin (HbA1c), measured over Baseline (week 0), week 52.

The primary endpoint is change in glycated haemoglobin (HbA1c) from baseline (week 0) to week 52, reported as percentage of HbA1c. It measures change over the study period in the blood sugar control outcome selected by the registry. The study summary describes blood sugar control as the central comparison between weekly insulin icodec and daily insulin treatment, making HbA1c the principal measure for that question.

A change score focuses on how the measure moved from each participant’s baseline rather than only comparing values at the end of follow-up. The registry reports group least-squares means for this change and standard errors around those means. The least-squares mean is an adjusted mean from a statistical model; the registry’s principal between-group analysis is identified as ANCOVA. The adjustment variables used in that ANCOVA are not reported in the registry record.

The primary outcome is continuous, so the central comparison is expressed as a mean difference rather than a count of events. A mean summarizes the average change in each group, but it does not describe every participant’s individual result or the full spread of individual changes. The registry states that the full analysis set included all randomized participants, while the number analyzed for the outcome measure was the number with available data for that outcome. The approach to missing week 52 values is not reported in the registry record.

HbA1c is useful for evaluating the study’s stated blood sugar-control question, but this endpoint alone does not capture all aspects of treatment experience or safety. The listed secondary outcomes address some other dimensions, including treatment satisfaction, treatment-related impact, treatment changes, and hypoglycaemia. Their presence does not make them equivalent to the designated primary endpoint.

7. Results Posted in the Registry

• Once Daily Basal Insulin Analogue: mean -1.31 versus -1.68 on control; difference 0.37 (95% CI 0.08 to 0.66), standardized effect d = 0.15; z-test p = 0.014.

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.

Change in Diabetes Treatment Satisfaction Questionnaire (DTSQs) in Total Treatment Satisfaction

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Once Daily Basal Insulin Analogue: mean 3.90 versus 4.68 on control; difference -0.78 (95% CI -1.47 to -0.09), standardized effect d = -0.14; z-test p = 0.027.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Once Daily Basal Insulin Analogue than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Treatment Related Impact Measure for Diabetes (TRIM-D) Compliance Domain

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Once Daily Basal Insulin Analogue: mean 87.37 versus 90.42 on control; difference -3.05 (95% CI -4.82 to -1.28), standardized effect d = -0.20; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Once Daily Basal Insulin Analogue than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The numerical secondary summaries cover treatment satisfaction and the TRIM-D compliance domain. Their recalculated comparisons point in the direction of higher average scores in the icodec group on the displayed contrast, but they are not additional primary outcomes. They provide information about participant-reported treatment experience alongside the primary HbA1c measure, rather than a substitute for it.

The registry lists other secondary outcomes, including treatment discontinuation or intensification and hypoglycaemia outcomes, but numerical results for those outcomes are not reported in the registry record. Thus the overall secondary-outcome picture is incomplete in the posted information. It is not possible to use the listed outcomes alone to make a comprehensive comparison across treatment experience, treatment changes, and hypoglycaemia.

Secondary results generally carry less weight than the designated primary result when several outcomes are examined. If many outcomes are tested, some may appear statistically notable by chance, even when there is no underlying effect. The registry does not report a multiplicity adjustment or a prespecified hierarchy for interpreting these outcomes. The unadjusted secondary p-values therefore should not be read as confirmatory evidence on their own.

These outcomes also differ in measurement and interpretation. A treatment satisfaction score, a compliance-domain score, and a hypoglycaemia episode count do not measure the same construct as HbA1c change. The secondary measures can enrich interpretation of the primary comparison, but their results require attention to their own analysis populations, missing data, and measurement scales. Those analysis details are not reported in the registry record for the secondary outcomes.

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

• Time From Baseline to Treatment Discontinuation or Intensification
• Change in Diabetes Treatment Satisfaction Questionnaire (DTSQs) in Total Treatment Satisfaction
• Treatment Related Impact Measure for Diabetes (TRIM-D) Compliance Domain
• Number of Severe Hypoglycaemic Episodes (Level 3)
• Number of Clinically Significant Hypoglycaemic Episodes (Level 2) (Below 3.0 Millimoles Per Liter [mmol/L] (54 Milligrams Per Deciliter [mg/dL]), Confirmed by Blood Glucose (BG) Meter)
• Number of Clinically Significant Hypoglycaemic Episodes (Level 2) (Below 3.0 mmol/L (54 mg/dL), Confirmed by BG Meter) or Severe Hypoglycaemic Episodes (Level 3)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous, so the registry reports a group mean change and a between-group mean difference. A mean is a summary of the group’s average change; it is not a statement that every participant experienced that change. The least-squares means are model-based adjusted means. Their reported standard errors describe the precision of the estimated means, rather than the range of individual participant values.

What does ANCOVA contribute?

ANCOVA is a regression-based method for comparing group outcomes while accounting for selected covariates. In a randomized trial, adjustment can improve precision when the included covariates are prognostic, and it can account for baseline outcome values if those are included in the model. The registry identifies the sponsor’s primary analysis as ANCOVA, but the covariates, model details, and missing-data procedures are not reported in the registry record. Consequently, the adjustment cannot be independently reproduced from the posted summary information.

The sponsor’s analysis is labelled non-inferiority. A non-inferiority assessment compares the estimated treatment difference with a prespecified margin that defines how much difference could be accepted under the study’s criterion. The registry provides the treatment-difference estimate, its confidence interval, and a p-value, but does not report the margin in this record. Without that margin and its orientation, the label can be reported but the decision rule cannot be independently evaluated from these details alone.

What is the recalculated comparison?

The recalculated result is an unadjusted re-analysis using posted group summary statistics. For the primary measure, it compares the displayed group mean changes and gives a mean difference with a confidence interval and unadjusted p-value. The calculation uses the reported standard errors and an approximate pooled standard deviation, as indicated in the statistical details. It is not the sponsor’s ANCOVA, does not reproduce its adjustment, and should not replace the registry’s reported analysis.

Standard errors and standard deviations answer different questions. A standard deviation describes spread among observations; a standard error describes uncertainty in an estimated mean. When only means and standard errors are posted, an underlying standard deviation may be approximated using assumptions about the data and analysis. That approximation makes the recalculation less direct than an analysis based on participant-level data. The approximation is explicitly indicated for the comparisons in the statistical details.

How should the p-values be read?

A p-value summarizes how compatible the observed data are with a specified null hypothesis under the analysis assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of a difference. The recalculated p-value is unadjusted. The sponsor’s p-value belongs to its ANCOVA non-inferiority analysis and should be interpreted in the context of that analysis and its decision rule. The registry does not report a multiplicity strategy for the family of secondary outcomes.

A confidence interval gives a range of estimates compatible with the data and model at the stated confidence level. It shows the precision of an estimate more directly than a p-value alone, but it does not account for every potential source of bias or model error. In particular, a narrow interval cannot correct for missing data, outcome measurement limitations, or incomplete reporting of the analysis plan.

Why can over-dispersion matter?

For count outcomes, over-dispersion means that variation is greater than expected under a simple model such as a Poisson model. If it is ignored, standard errors may be too small and confidence intervals too narrow. This is a general statistical consideration for event counts, including hypoglycaemia episode counts. The primary endpoint in this trial is continuous rather than a count, and the registry information summarized for it does not indicate that an over-dispersion model was used. The posted numerical details do not provide enough information to describe how any count-based secondary outcomes were modeled.

Finally, the registry states that the full analysis set included all randomized participants, but outcome-specific analysis numbers were based on available data. The distinction matters: randomization defines the assigned groups, whereas an outcome analysis may include only participants with usable data for that outcome. The precise handling of unavailable values, treatment discontinuation, and deviations from assigned treatment is not reported in the registry record. These details limit independent assessment of the analysis population and robustness of the estimates.

11. Confidence Intervals and Precision

The primary recalculated mean difference is 0.37, with a 95% confidence interval from 0.076 to 0.664, on the daily-analogue-minus-icodec contrast. The interval describes uncertainty in the estimated average difference under the assumptions of the unadjusted calculation. It does not mean that individual participant changes are confined to that range; the interval concerns the group-level mean contrast.

The sponsor-reported ANCOVA estimate is -0.38 with a 95% confidence interval from -0.66 to -0.09. That interval is reported on the sponsor’s treatment-difference scale and accompanies the analysis labelled non-inferiority. Because the subtraction order and non-inferiority margin are not fully specified in the registry record, interpretation should retain the sponsor’s stated scale and analysis rather than treating its interval as directly interchangeable with the recalculated interval.

For either interval, precision depends on the amount and variability of observed data and on the analysis model. A confidence interval reflects sampling uncertainty under those assumptions; it does not incorporate uncertainty from unreported modeling choices or possible biases. The standard errors reported for the group means are not measures of participant-to-participant spread. The recalculated intervals also rely on approximate standard deviations derived from posted summary statistics, rather than a re-analysis of participant-level data.

The registry reports the sponsor’s interval at 95% confidence and provides intervals for the recalculated comparisons. These intervals convey more information about plausible effect sizes than a p-value alone. They do not establish clinical importance by themselves, and the registry record does not provide a threshold for a clinically meaningful HbA1c difference. The intervals should therefore be read as statistical uncertainty around the estimates, alongside the analysis design and the limitations of the posted details.

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 sample-size figures are planning illustrations based on the observed effect, not the trial’s original design. For these continuous outcomes, the figures are participants per arm, not total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the number of participants per arm estimated to yield an 80% chance of detecting the specified effect under those planning assumptions, if that effect were the true effect and the assumptions held. The 90% power column gives the corresponding planning estimate for a 90% chance. These are design calculations, not a guarantee that a future study will reach a particular result.

Rows for planning at 100%, 75%, and 50% of the observed effect show how the assumed effect size changes the required enrollment. Smaller effects require more participants to distinguish them from no difference with the same planned power and significance level. The sample-size curve visualizes the same relationship across effect assumptions; it should be interpreted as a planning aid, not as a report of what happened in the trial.

This is not observed power. It is a prospective planning illustration derived from the observed estimate, and it does not measure the probability that the completed trial’s result is correct. Because the observed effect itself is uncertain, a planning exercise based on it should not be treated as a definitive sample-size recommendation. The calculations also do not establish that the original trial was designed around these assumptions.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.370; pooled standard deviation σ = 2.469.
2. n = 2 × (1.96 + 0.842)² × 2.469² / 0.370² = 699.2, rounded up to 700 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
delta <- 0.37                      # observed difference in means
sigma <- 2.46931164497315                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(4760626)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 700 (80% power), 936 (90% power)
 75% of observed effect: n per arm = 1,243 (80% power), 1,664 (90% power)
 50% of observed effect: n per arm = 2,797 (80% power), 3,744 (90% power)
power.t.test: 700.1 
Simulated power at n = 700 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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. Analyses Reported by the Sponsor

The sponsor-reported primary analysis is an ANCOVA treatment-difference analysis labelled non-inferiority. It reports -0.38, with a 95% confidence interval from -0.66 to -0.09 and p < 0.001. This is the registry’s own adjusted analysis and the relevant result to cite when describing the sponsor’s formal non-inferiority finding.

The recalculated comparison uses the posted group means and standard errors without the sponsor’s model adjustment. It gives 0.37, with a 95% confidence interval from 0.076 to 0.664 and an unadjusted p-value of 0.014, using daily analogue minus icodec as the contrast. The opposite signs reflect that the recalculation uses the reverse contrast direction from the reported treatment difference; the registry’s summary does not fully specify the sponsor’s subtraction order. The estimates are close in magnitude but are not identical.

Differences between the results can arise because an adjusted ANCOVA and an unadjusted summary-statistic calculation are not the same analysis. They can use different variance estimates and account for covariates differently. The recalculation is useful as a check on the direction and approximate scale of the posted group summaries, but it does not supersede the sponsor’s analysis or reproduce the non-inferiority decision rule. The non-inferiority margin is not reported in the registry record, so the stated label cannot be independently checked against that margin.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies change in HbA1c from baseline to week 52 as the primary outcome and reports a sponsor ANCOVA analysis labelled non-inferiority. It also lists several secondary outcomes. Testing multiple outcomes increases the chance of observing at least one apparently notable result by chance if each is interpreted without accounting for the family of tests. The registry does not report whether the secondary analyses were adjusted for multiplicity, whether an outcome hierarchy was specified, or whether any secondary findings were intended to be confirmatory.

The listed secondary outcomes include treatment discontinuation or intensification, treatment satisfaction, treatment-related impact, and hypoglycaemia measures. Numerical summaries are available for treatment satisfaction and the TRIM-D compliance domain, while numerical results for other listed secondary outcomes are not reported in the registry record. The reported secondary p-values are unadjusted recalculations, and they should be interpreted cautiously in the context of multiple outcomes.

The primary population is described as the full analysis set, including all randomized participants. The registry also states that the number analyzed for an outcome measure is the number of participants with available data for that measure. This means the outcome-specific analysis population can depend on data availability. The missing-data mechanism, the method used to handle missing week 52 measurements, and sensitivity analyses are not reported in the registry record. Without those details, it is not possible to assess whether missing outcomes could have shifted the estimated treatment contrast.

The registry does not report the covariates included in the ANCOVA, the non-inferiority margin, or detailed rules for handling treatment discontinuation and changes in assigned therapy. These omissions constrain independent evaluation of the primary analysis. They do not invalidate the posted result, but they limit what can be concluded about the robustness of the result under alternative reasonable analysis choices.

16. 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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: coronary artery disease (0.9%), acute myocardial infarction (0.7%), acute kidney injury (0.6%), covid-19 pneumonia (0.6%), covid-19 (0.6%).

The registry reports serious adverse events for the insulin icodec group among 542 participants at risk, with 45 recorded as affected. For the once-daily basal insulin analogue group, 538 participants were at risk and 57 were recorded as affected. The reported other adverse event figures are 43 among 542 at risk for icodec and 55 among 538 at risk for the daily analogue group. The registry also records 3 deaths in the icodec group and 6 in the daily analogue group.

These figures are counts, not adjusted comparisons. The at-risk numbers differ from one another in the safety record, and the registry does not provide a statistical comparison or an account of event timing in the information. The serious-event terms listed include coronary artery disease (0.9%), acute myocardial infarction (0.7%), acute kidney injury (0.6%), COVID-19 pneumonia (0.6%), COVID-19 (0.6%), and cardiac failure (0.6%). These are reported terms and percentages, not proof that the events were caused by study treatment.

Serious adverse events, other adverse events, deaths, and specific event terms are different reporting categories and should not be added together. A participant may have more than one event, and the registry summary does not establish whether listed categories overlap. The figures describe the posted safety record; they do not establish comparative safety or rule out uncommon or delayed events.

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

The primary data describe average HbA1c change through week 52 among participants with outcome data, with a sponsor-reported ANCOVA analysis and a separate unadjusted recalculation. The sponsor’s result is labelled non-inferiority, while the recalculation provides a simple comparison of posted summary statistics. These analyses answer related but not identical statistical questions, and the sponsor analysis should remain the reference for the trial’s formal reported conclusion.

The estimates do not imply that each participant had the group-average change. Individual responses can vary, and the group means do not show the distribution of outcomes for each person. The registry record does not report a threshold for a clinically meaningful difference or the non-inferiority margin, so the statistical estimates alone cannot establish how important a difference is in clinical terms.

The trial compares treatment strategies that differ in both injection schedule and use of DoseGuide for titration. The result therefore pertains to the assigned strategies as described, rather than isolating the effect of dosing frequency or the app separately. The open-label design is relevant context, particularly for participant-reported outcomes. The registry does not report whether treatment adherence, app use, or other implementation measures altered the observed results.

The findings also do not resolve every outcome relevant to treatment. Secondary numerical summaries are limited, and results for several listed outcomes are not reported in the registry record. The safety counts do not by themselves establish causation or provide a fully adjusted comparison. Results beyond week 52, and the durability of any observed differences, are not reported in the registry record.

Overall, the posted results support a statistical description of the measured outcomes under the reported analyses. They do not by themselves establish broad claims about all people with type 2 diabetes, all insulin regimens, or outcomes beyond the follow-up described. Interpretation should retain the trial population, open-label design, available-data analysis population, and incomplete reporting of certain methods and outcomes.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares an insulin strategy with once-weekly dosing and app-guided titration against once-daily basal insulin analogues in an insulin-naïve type 2 diabetes population. The primary measure is a continuous outcome, and the record includes both a sponsor-reported adjusted analysis and a recalculation based on posted summary statistics. That contrast illustrates why an estimate should be interpreted with its analysis method and contrast direction, not just its sign or p-value.

The trial also highlights the distinction between non-inferiority reporting and a straightforward test of whether a difference is zero. A non-inferiority conclusion depends on a margin and a defined analysis strategy. The registry labels the sponsor analysis as non-inferiority, but the margin is not reported in the registry record. Statistical readers should therefore distinguish the posted label from an independent verification of the decision rule.

Planning figures based on observed effects can help readers understand how enrollment needs change when the target effect is smaller or when planned power is higher. They are illustrations rather than a reconstruction of the trial’s original design, since the original design assumptions are not reported. This separation between retrospective planning illustration and prospective design rationale is important when reading power calculations.

Finally, the registry’s mixture of primary, secondary, and safety outcomes makes multiplicity and missing-data reporting important. The record describes the outcome population as participants with available data, but gives limited detail on missing-data handling and adjustment across secondary tests. These are not minor technicalities: they determine how confidently estimates can be compared and how much weight secondary findings should carry.

19. 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 does not report the non-inferiority margin or the full ANCOVA specification. Without those details, the sponsor’s non-inferiority decision rule cannot be independently reconstructed, and covariate adjustment cannot be assessed. The recalculated comparison is approximate and unadjusted, based on posted summary statistics rather than participant-level data; it should not be treated as a replacement analysis.

The open-label design is a further interpretive limitation. Participants and investigators were not masked according to the registry, and treatment satisfaction and treatment-related impact are participant-reported measures. The record does not describe methods intended to address possible effects of knowing assigned treatment on those measures. The weekly strategy also includes DoseGuide, so the effects of dosing schedule and app-guided titration are not separately identified by this comparison.

Outcome-specific analysis numbers are based on available data, but the registry does not report the missing-data approach, reasons for missing observations, or sensitivity analyses. This limits assessment of whether missing values could have affected the primary or secondary results. The record also does not report detailed adherence or app-use results, which would help characterize exposure to the assigned strategies.

Numerical secondary results are incomplete in the registry record. Several listed outcomes, including hypoglycaemia measures and treatment discontinuation or intensification, do not have numerical results in the information summarized. Multiplicity handling is not reported, so unadjusted secondary findings warrant caution. Safety counts are descriptive, and the registry summary does not provide an adjusted comparison, event timing, or enough detail to attribute events to treatment.

The results address the recorded study population and follow-up through week 52 for the primary endpoint. Longer-term outcomes and the generalizability of results beyond the described insulin-naïve population are not reported in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Research Study to Compare a New Weekly Insulin, Insulin Icodec Used With DoseGuide App, and Daily Insulins in People With Type 2 Diabetes Who Have Not Used Insulin Before — NCT04760626.
• Registry record last updated 2026-08-11; status last verified 2026-08.
• Results first posted 2025-08-26.
• 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.

23. Record Summary

The registry record describes a completed phase 3 randomized, parallel-group, open-label study of insulin icodec with DoseGuide versus once-daily basal insulin analogues in people with type 2 diabetes who had not used insulin before. Enrollment is recorded as 1085. The primary outcome is change in HbA1c from baseline (week 0) to week 52.

The posted group summaries show mean HbA1c changes of -1.68 for insulin icodec and -1.31 for once-daily basal insulin analogue. The sponsor-reported ANCOVA result is labelled non-inferiority and reports a treatment difference of -0.38, with a 95% confidence interval from -0.66 to -0.09 and p < 0.001. The registry does not report the non-inferiority margin.

Secondary outcomes include treatment satisfaction, treatment-related impact, treatment changes, and hypoglycaemia. Safety figures are descriptive, and several analysis details—including missing-data handling and multiplicity procedures—are not reported in the registry record.