A "Real World" Trial to Determine Efficacy and Health Outcomes of Toujeo (ACHIEVE CONTROL REAL LIFE STUDY PROGRAM): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A "Real World" Trial to Determine Efficacy and Health Outcomes of Toujeo (ACHIEVE CONTROL REAL LIFE STUDY PROGRAM) (NCT02451137), a randomized phase 4 study. The primary outcome is Percentage of Participants With Individualized Glycated Hemoglobin Target Attainment Per Healthcare Effectiveness Data and Information Set (HEDIS) Criteria Without Documented Symptomatic(Blood Glucose <=70 mg/dL [<=3.9 mmol/L]) and/or Severe Hypoglycemia.

2. Clinical Question

This question evaluates a combined clinical outcome rather than HbA1c change alone. A participant had to meet the applicable HbA1c target and avoid the specified hypoglycemia events during the assessment period to count as reaching the endpoint. The trial therefore asks whether the two treatment strategies differ in the proportion achieving both parts of that outcome. The registry describes a pragmatic, real-world comparison among people starting basal insulin, rather than a comparison limited to a laboratory measure or a single component of treatment experience.

3. Trial Design

This was a Phase 4, randomized, open-label, parallel-group trial with 3304 participants. The trial was completed. Its listed start date was 2015-06-16, and its primary completion date was 2018-03-02. Sanofi was the sponsor. The registry describes participants with type 2 diabetes who were insulin-naive, had uncontrolled disease, and were initiating basal insulin therapy.

Participants were allocated to Toujeo or Standard of Care. Toujeo was insulin glargine at 300 U/mL, given once daily. Standard of Care allowed either insulin glargine at 100 U/mL once daily or insulin detemir once or twice daily. The assigned therapies were described for use up to Month 12, with or without an available participant support program. The record does not report the allocation ratio or the process used to generate or conceal the randomization sequence.

The trial was open-label, meaning that masking was not used. This can matter when outcomes involve participant or provider reports and treatment decisions: knowledge of assignment may influence reporting or care. The primary endpoint, however, also included HbA1c target attainment and documented hypoglycemia, which are defined through clinical measurements and event records. Open-label status does not by itself establish whether a particular result was biased; it is a design feature to consider when interpreting the results.

The primary outcome was assessed from baseline to Month 6. The registry says the analysis used post-baseline data available during the six-month randomized period, defined as time from randomization up to Day 180 or the discontinuation date, whichever came earlier. Several secondary objectives concerned outcomes through Month 6 and Month 12, including treatment persistence, glycemic measures, body weight, patient- and provider-reported outcomes, and healthcare resource use. Details of the trial’s original sample-size justification and the statistical plan beyond the analyses described in the registry record are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Out of 3305 participants, 1 participant was not randomized, but received treatment; hence not included in any analysis (baseline, efficacy and safety). Assignment in arms was done centrally by interactive response technology (IRT) in 1:1 ratio.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on Intent-to-Treat (ITT) population which comprised of all randomized participants, irrespective of the treatment actually received, and analyzed according to the treatment group allocated by randomization.

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: Analysis was performed on all randomized participants. Participants belonging to more than one race, have been reported under each applicable category.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants With Individualized Glycated Hemoglobin Target Attainment Per Healthcare Effectiveness Data and Information Set (HEDIS) Criteria Without Documented Symptomatic(Blood Glucose <=70 mg/dL [<=3.9 mmol/L]) and/or Severe Hypoglycemia, measured over Baseline to Month 6.

The primary endpoint is the percentage of participants who reached an individualized HbA1c target without documented symptomatic or severe hypoglycemia from baseline to Month 6. The HEDIS target was below 8% for participants aged 65 years or older or with defined medical comorbidities, and below 7% otherwise. Documented symptomatic hypoglycemia required typical symptoms together with a measured plasma glucose concentration at or below 70 mg/dL (at or below 3.9 mmol/L). Severe hypoglycemia was defined as an event in which the participant needed another person’s assistance to actively administer carbohydrate, glucagon, or another resuscitative action.

This is a composite, or joint-success, endpoint: reaching the HbA1c target alone is not enough if a qualifying hypoglycemia event is also documented. It is used to capture two aspects of treatment outcome together—glycemic target attainment and avoidance of the specified hypoglycemia events. Because it is reported as a percentage, the result describes the proportion of participants meeting the full definition, rather than the average HbA1c value or the number of events per participant.

The endpoint definition combines individualized targets with a common hypoglycemia threshold. That makes the HbA1c criterion dependent on age or comorbidity status, while the symptomatic-event criterion depends on symptoms and a measured glucose value. The registry specifies that the primary analysis used all available post-baseline data within the randomized period through Day 180 or discontinuation, whichever came earlier. Further details about how incomplete endpoint information was handled are not reported in the registry record.

7. Results Posted in the Registry

• Toujeo: 31.3% (517/1651) versus 27.9% (461/1653) on control; risk difference 3.4 percentage points (95% CI 0.3 to 6.5), risk ratio 1.12 (1.01–1.25); unadjusted two-proportion z-test p = 0.031.

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.

Percentage of Participants Reaching Individualized HbA1c Target Without Documented Symptomatic <3.0 mmol/L (<54 mg/dL) and/or Severe Hypoglycemia During the 6-Month Randomized Period

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Toujeo: 37.3% (616/1651) versus 34.3% (567/1653) on control; risk difference 3.0 percentage points (95% CI -0.3 to 6.3), risk ratio 1.09 (0.99–1.19); unadjusted two-proportion z-test p = 0.071.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Toujeo and the control. The interval gives the range of effects that remain compatible with the data.

The trial’s secondary objectives cover several kinds of outcome: changes in HbA1c, fasting plasma glucose, body weight, and basal insulin dose; persistence with assigned basal insulin; hypoglycemia; participant- and provider-reported treatment experience; global effectiveness ratings; and healthcare use and costs. These outcomes broaden the questions beyond the primary combined endpoint, but they do not all measure the same aspect of treatment or use the same statistical scale.

The registry lists 12 secondary outcomes. The detailed secondary result available in the registry record concerns reaching the individualized HbA1c target without documented symptomatic hypoglycemia below 3.0 mmol/L (below 54 mg/dL) or severe hypoglycemia during the six-month randomized period. For this result, the unadjusted recalculation reports a risk difference of 0.030, with a 95% confidence interval from -0.003 to 0.063 and p = 0.071. The confidence interval includes zero. The registry does not report a sponsor analysis for this secondary comparison. Detailed results for the other named secondary outcomes are not reported in the registry record.

The primary outcome uses a different hypoglycemia definition, including documented symptomatic events at or below 70 mg/dL, and the secondary outcome uses a more restrictive glucose threshold. These endpoints are related, but they are not interchangeable. The primary result therefore cannot be used to infer the result for every secondary definition or outcome domain.

Unadjusted secondary tests carry less weight when many outcomes are examined. General statistical explanation: as the number of separate tests increases, the chance of at least one apparently unusual result can increase even when no corresponding effects exist. A prespecified hierarchy or multiplicity adjustment can address that concern, but such procedures are not reported in the registry record. Secondary findings are best viewed as supportive or exploratory unless the analysis plan establishes otherwise. Because the detailed results are incomplete in the registry record, no overall pattern across all secondary endpoints can be established.

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

• Change From Baseline in HbA1c at Month 6 and Month 12
• Treatment Persistence Measured by Medication Possession Ratio (MPR)
• Percentage of Participants Reaching Individualized HbA1c Target Without Documented Symptomatic <3.0 mmol/L (<54 mg/dL) and/or Severe Hypoglycemia During the 6-Month Randomized Period
• Percentage of Participants Reaching Individualized HbA1c Target Without Documented Symptomatic <=3.9 mmol/L (<= 70 mg/dL) and <3.0 mmol/L (< 54 mg/dL) and/or Severe Hypoglycemia During the 12-Month Randomized Period
• Change From Baseline in Fasting Plasma Glucose (FPG) at Month 6 and Month 12
• Change From Baseline in Body Weight at Month 6 and Month 12
• Change From Baseline in Basal Insulin Dose at Month 6 and Month 12
• Percentage of Responders (Participants and Provider) Who Reported "Excellent" or "Good" Responses to Global Effectiveness Scale (GES) Question at Month 6, and Month 12

9. Statistical Interpretation

10. Statistical Methods Explained

The statistical results include a sponsor-reported logistic regression analysis and simple unadjusted recalculations from the posted group summaries. They address the same primary endpoint but do not have the same statistical form. Understanding the distinction helps prevent an odds ratio from being mistaken for a risk difference or for a direct statement about percentage-point change.

What does the primary percentage represent?

The endpoint is binary at the participant level: each participant either met the full definition or did not. The group percentages, 31.3% for Toujeo and 27.9% for Standard of Care, are the observed proportions classified as meeting the endpoint. They are not average HbA1c changes, and they do not describe the frequency of hypoglycemia events among participants who had such events. Since the endpoint requires both target attainment and no qualifying event, a participant who meets the HbA1c criterion but has a documented qualifying event does not count as a success for this combined outcome.

What does the odds ratio describe?

Logistic regression models the odds of a binary outcome. The sponsor-reported odds ratio of 1.19 compares the modeled odds of reaching the combined endpoint between the randomized groups under that analysis. An odds ratio above 1 indicates higher estimated odds in the group named in the comparison, but it is not the same as saying that the probability is 1.19 times as large. Odds and probabilities are related but have different scales. The registry reports a 95.4% confidence interval from 1.01 to 1.39 and p = 0.031 for this regression analysis. The adjustment variables, if any, are not reported in the registry record.

What does the unadjusted recalculation add?

The unadjusted recalculation uses the posted event counts and group sizes to estimate contrasts without a regression model. It reports a risk difference of 0.034, with a confidence interval from 0.003 to 0.065, and p = 0.031. A risk difference is on the probability scale, so its estimate describes the contrast in observed proportions. The recalculated risk ratio of 1.123 compares the proportions as a ratio, and the recalculated odds ratio of 1.179 compares the corresponding odds. Their confidence intervals are also reported from the unadjusted summaries. These recalculations are not substitutes for the sponsor’s model; they offer a transparent, simple comparison using the posted counts.

For a binary outcome, uncertainty in a proportion or difference between proportions depends partly on the number of participants and on how common the outcome is. A confidence interval gives a range of values compatible with the estimate under the method’s assumptions. A p-value measures how incompatible the observed result is with a specified null model, if that model and its assumptions hold. It does not give the probability that the null hypothesis is true, and it does not measure clinical importance.

Why can analysis choices matter?

Randomization supports a fair comparison by allocating participants to groups through chance. An intent-to-treat analysis keeps participants in their assigned group for analysis, whether or not the treatment actually received matched assignment. The registry defines its primary intent-to-treat population as all randomized participants and says they were analyzed according to allocated group. This approach retains the randomized comparison, though it does not eliminate the need to understand missing outcome information or deviations from treatment.

The registry states that the primary endpoint analysis used all post-baseline data available during the six-month randomized period, ending at Day 180 or discontinuation, whichever came earlier. This defines the observation window, but it does not explain how missing outcome components were handled, whether imputation was used, or whether sensitivity analyses were conducted. Those procedures are not reported in the registry record.

Open-label assignment may be relevant to outcomes shaped by participant or provider reporting. General statistical consideration: awareness of treatment can affect measurement or reporting, so objective outcome definitions and consistent ascertainment are important. The endpoint includes measured glucose, HbA1c target attainment, symptoms, and severe events requiring assistance. The registry record does not provide enough detail to assess the consistency of event ascertainment across groups.

Why does over-dispersion matter in other outcome types?

General statistical explanation: over-dispersion occurs when the observed variability in count data exceeds what a simple model assumes. It can make standard errors too small if the model does not allow for the extra variability. That issue is relevant to some event-rate analyses, but the primary endpoint is a participant-level percentage rather than a count rate. The registry describes secondary outcomes involving hypoglycemia incidence and rate, but further model details for those analyses are not reported in the registry record.

The trial’s broader statistical plan—including how the primary analysis handled covariates, whether interim analyses were performed, and how multiple outcomes were prioritized—is not reported in the registry record. The results should therefore be interpreted as the estimates and tests explicitly posted, rather than as evidence about undocumented analysis decisions.

11. Confidence Intervals and Precision

The primary result is estimated rather than known without uncertainty. The sponsor-reported odds ratio is 1.19, with a 95.4% confidence interval from 1.01 to 1.39. The interval lies above the null odds ratio of 1, but its lower end is close to that value. The reported p-value is 0.031. The unadjusted recalculation gives a risk difference of 0.034, with a 95% confidence interval from 0.003 to 0.065, and a p-value of 0.031. The lower end of that interval is close to zero, the null value for a risk difference.

A confidence interval describes the uncertainty of an estimate under the statistical method used. In repeated use of the same method and assumptions, intervals at a stated confidence level would cover the underlying parameter at that rate. It is not a guarantee that the particular interval contains the true value, nor does it include every source of uncertainty, such as missing-data bias or measurement differences.

The odds-ratio interval and risk-difference interval are on different scales and should not be compared as if their endpoints represented the same quantity. The registry reports a 95.4% interval for the sponsor model and a 95% interval for the unadjusted risk-difference recalculation. The precise width and location of each interval communicate uncertainty for its own measure. Neither interval establishes whether the observed difference is important in clinical practice, and neither addresses outcomes beyond the defined six-month combined endpoint.

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 planning figures are illustrations based on the observed effect, not the trial’s original design calculations. For this binary endpoint, the figures are participants per arm. They are not total event counts. The registry’s planning 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 the stated fraction of the observed effect.

The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions, if that effect and the assumptions were true. The 90% power column gives the corresponding planning number for a 90% chance. Higher planned power generally requires a larger sample, all else equal. For this endpoint, planning for 100% of the observed effect gives 2,787 participants per arm for 80% power and 3,730 for 90% power. Planning for 75% of the observed effect gives 4,913 per arm for 80% power and 6,576 for 90% power. Planning for 50% of the observed effect gives 10,959 per arm for 80% power and 14,670 for 90% power. Smaller effects need more participants to detect under otherwise unchanged assumptions.

These are prospective planning illustrations, not a measure of the probability that the completed trial found a real effect. They are not observed power. The original sample-size justification and assumptions used to design the trial are not reported in the registry record. The planning illustration should therefore be read as a separate, hypothetical calculation using the listed assumptions, not as evidence about the trial’s original design choices.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.279; treatment proportion p₁ = 0.313; average p̄ = 0.296.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6456 = 1.2653.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6451 = 0.5430.
4. n = (1.2653 + 0.5430)² / (0.0343)² = 2,786.4, rounded up to 2,787 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.278886872353297                          # control proportion
p1 <- 0.313143549364022                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2451137)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 2,787 (80% power), 3,730 (90% power)
 75% of observed effect: n per arm = 4,913 (80% power), 6,576 (90% power)
 50% of observed effect: n per arm = 10,959 (80% power), 14,670 (90% power)
power.prop.test: 2786.4 
Simulated power at n = 2,787 per arm: 0.796 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s posted primary analysis is a logistic regression, reporting an odds ratio of 1.19 with a 95.4% confidence interval from 1.01 to 1.39 and p = 0.031. The unadjusted recalculation from the posted event counts and group sizes reports a risk difference of 0.034, with a 95% confidence interval from 0.003 to 0.065 and p = 0.031. The recalculation also reports an odds ratio of 1.179 and a risk ratio of 1.123, each with its own confidence interval.

The sponsor model and the recalculation use different methods and measures. Logistic regression estimates an odds ratio, while the simple recalculation derives contrasts directly from the group summaries without adjustment. Their estimates should not be expected to match exactly, and the odds ratio should not be interpreted as a risk difference. In this case, the reported estimates point in the same direction and the p-values are similar. The registry does not identify a discrepancy between the posted sponsor result and the recalculation.

The sponsor analysis is the trial’s reported analysis. The recalculation is useful as an independent, unadjusted check based on the posted summaries, but it does not replace the sponsor’s model or establish which covariates, if any, were considered. Details of the regression specification are not reported in the registry record. Interpretation should preserve that distinction and avoid treating the simple recalculation as an exact reproduction of the regression analysis.

15. Multiplicity, Missing Data and Analysis Populations

The trial has a primary endpoint and multiple secondary objectives. The registry lists 12 secondary outcomes, covering glycemic measures, persistence, hypoglycemia, treatment satisfaction, global effectiveness, and healthcare use. Testing several outcomes can create a multiplicity issue: when many hypotheses are assessed, the chance of at least one small p-value arising by chance can increase. General statistical approaches to this issue include adjusting significance thresholds, controlling an error rate, or specifying an ordered testing strategy. The registry record does not report whether any such approach was planned or used.

Without a documented multiplicity strategy, the primary result should be distinguished from secondary findings. The primary endpoint has a defined reported sponsor analysis. Secondary results may involve different outcome definitions and time points, and their interpretation depends on whether they were prespecified and how they were analyzed. The available registry record does not report enough information to determine the status or interpretation framework for every secondary analysis.

The primary population is identified as intent-to-treat and includes all randomized participants analyzed according to their assigned group. The endpoint analysis used available post-baseline data during the randomized period through Day 180 or discontinuation, whichever came earlier. The registry does not report the number of participants with missing primary outcome data, reasons for missingness, the method used to handle missing values, or sensitivity analyses addressing missingness. It is therefore not possible to judge from this record alone whether missing data could have altered the estimated group contrast.

The record also does not report interim analyses, stopping rules, or the trial’s original sample-size justification. These are separate from the planning illustration based on the observed effect. It does not report whether regression covariates were prespecified or how the sponsor’s logistic model was specified. These absences do not demonstrate that the procedures were not conducted; they mean that their details are not reported in the registry record and cannot be assessed from the posted information.

Randomization and intent-to-treat analysis support the intended comparison, but they do not remove uncertainty about outcome ascertainment, missing information, multiplicity, or analysis choices. Those issues are relevant context when deciding how much confidence to place in a single reported p-value or in a pattern of secondary findings.

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

Serious events with the highest arm-level rate: pneumonia (0.9%), coronary artery disease (0.6%), non-cardiac chest pain (0.6%), acute kidney injury (0.6%), sepsis (0.5%).

The registry reports serious adverse events among 1637 participants at risk in the Toujeo group and 1621 at risk in the Standard of Care Basal Insulin group. Serious adverse events affected 161 participants in the Toujeo group and 165 in the Standard of Care group. The registry also reports 7 deaths in the Toujeo group and 4 in the Standard of Care group. These are descriptive counts; the record does not provide a statistical comparison of these figures in the information.

The listed serious event terms include pneumonia at 0.9%, coronary artery disease at 0.6%, non-cardiac chest pain at 0.6%, acute kidney injury at 0.6%, sepsis at 0.5%, and acute myocardial infarction at 0.5%. These percentages should be read as the registry’s reported frequencies for the listed terms, not as evidence that the assigned treatment caused those events. The record does not provide enough detail to interpret the timing, relatedness assessments, or group-specific statistical uncertainty for each term.

Adverse-event counts are not the same as the primary efficacy endpoint. Differences in counts alone do not establish a treatment-related safety effect, particularly without information about event definitions, follow-up, exposure, and comparative analyses. The record reports 0 for other adverse events in each group; further interpretation of that entry is not reported in the registry record.

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

The posted primary data show that 31.3% of participants in the Toujeo group and 27.9% in the Standard of Care group met the combined six-month endpoint. The sponsor-reported logistic regression estimated an odds ratio of 1.19, with a 95.4% confidence interval from 1.01 to 1.39. The unadjusted recalculation estimated a risk difference of 0.034, with a 95% confidence interval from 0.003 to 0.065. These results describe a difference in the proportion meeting the endpoint under the definitions and analysis methods reported.

The outcome is not HbA1c target attainment by itself. Participants also had to have no documented symptomatic hypoglycemia at or below 70 mg/dL and no severe hypoglycemia during the specified period. The endpoint therefore does not reveal separately how many participants reached the HbA1c target, how many had qualifying events, or how each component contributed to the combined result. Those component-specific primary results are not reported in the registry record.

The result also does not show that the same pattern applies to every secondary outcome, every subgroup, or treatment after the period assessed for the primary endpoint. The trial followed assigned treatment up to Month 12, and the listed secondary objectives include outcomes at Month 12, but results for most named secondary outcomes are not reported in the registry record. The detailed primary endpoint covers baseline to Month 6, not an unlimited duration.

Statistical significance is not a measure of clinical importance. The confidence intervals convey uncertainty in the estimated contrasts but do not resolve questions about patient priorities, healthcare costs, treatment burden, or the importance of each endpoint component. The open-label design, availability of support programs, incomplete reporting of missing-data procedures, and absence of reported multiplicity details are also relevant when assessing what conclusions can be drawn.

Accordingly, the posted result supports a comparison of the two randomized strategies on the defined combined outcome in this trial population. It does not establish a universal effect, explain the mechanism of any difference, or determine the balance of outcomes for an individual. The registry record does not report the detailed information needed to answer those broader questions.

18. Why This Trial Matters Statistically

This trial is statistically useful because it compares two basal-insulin strategies through randomization in a pragmatic, real-world setting and reports a patient-level outcome combining glycemic target attainment with avoidance of specified hypoglycemia. The endpoint is more clinically textured than HbA1c alone, while also requiring careful interpretation because success depends on both components being met.

The primary result illustrates why effect measures matter. A logistic-regression odds ratio and an unadjusted risk difference describe the same broad group comparison on different scales. Readers who focus on the odds ratio alone may mistake it for a direct ratio of probabilities. Reporting the group percentages and the risk-difference recalculation makes the absolute scale of the observed proportions easier to understand, while the sponsor model remains the trial’s reported analysis.

The confidence intervals are as important as the point estimates. They show that uncertainty remains around the estimated contrast, including values close to the null. A p-value below 0.05 does not remove that uncertainty or establish the practical importance of the result. This case demonstrates the value of interpreting the estimate, interval, outcome definition, and analysis population together rather than treating statistical significance as a complete conclusion.

Finally, the record illustrates the limits of registry reporting for a broad pragmatic trial. The objectives span persistence, glycemic measures, hypoglycemia, treatment experience, and healthcare use, but detailed results for most named secondary outcomes are not reported in the registry record. Missing-data procedures, multiplicity handling, and the original sample-size justification are also not reported. These omissions make clear why a focused primary result can be interpreted, while broader claims about the trial’s full outcome profile require information not present in the registry record.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The open-label design is a trial-specific limitation to consider, particularly for participant- and provider-reported outcomes and for treatment decisions that may be influenced by awareness of assignment. The primary endpoint includes laboratory and event-based components, but the registry does not provide details of outcome ascertainment or whether assessment procedures were standardized across groups.

The treatment comparison was not a single drug against a single comparator regimen. Standard of Care could be Lantus once daily or Levemir once or twice daily. The registry record does not report the distribution of comparator regimens or the factors used to choose between them. Participant support programs were also described as available with or without treatment; their use by group is not reported in the registry record.

The primary endpoint is a combined measure. Its result cannot show whether any group difference arose from HbA1c target attainment, symptomatic hypoglycemia, severe hypoglycemia, or a combination of components. Component-specific primary results are not reported in the registry record. The individualized HbA1c criterion also varies according to age or comorbidity status, and the record does not report the distribution of participants across those target categories.

Although the primary analysis is defined as intent-to-treat, the analysis uses available post-baseline data through Day 180 or discontinuation, whichever came earlier. The amount and pattern of missing outcome information, handling of missing data, and sensitivity analyses are not reported in the registry record. The registry also does not report the model specification for the sponsor’s regression, the multiplicity strategy, or the trial’s original sample-size assumptions.

Finally, results are available for only a limited part of the trial’s broad secondary agenda in the record described. This limits the ability to assess whether the primary finding is accompanied by consistent results in persistence, Month 12 outcomes, patient experience, or healthcare use. These limitations affect the scope and confidence of interpretation; they do not by themselves show that the primary estimate is invalid.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A "Real World" Trial to Determine Efficacy and Health Outcomes of Toujeo (ACHIEVE CONTROL REAL LIFE STUDY PROGRAM) — NCT02451137.
• Registry record last updated 2019-09-10; status last verified 2019-08.
• Results first posted 2019-05-01.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

NCT02451137 was a completed Phase 4, randomized, open-label, parallel-group trial sponsored by Sanofi. It enrolled 3304 participants with uncontrolled type 2 diabetes who were insulin-naive and initiating basal insulin. Participants were assigned to once-daily Toujeo, insulin glargine at 300 U/mL, or Standard of Care consisting of Lantus or Levemir.

The primary endpoint was the percentage reaching an individualized HEDIS HbA1c target without documented symptomatic hypoglycemia at or below 70 mg/dL or severe hypoglycemia from baseline to Month 6. The reported proportions were 31.3% for Toujeo and 27.9% for Standard of Care. The sponsor’s logistic regression reported an odds ratio of 1.19, with a 95.4% confidence interval from 1.01 to 1.39 and p = 0.031. An unadjusted recalculation reported a risk difference of 0.034, with a 95% confidence interval from 0.003 to 0.065 and p = 0.031.

The registry record does not report the trial’s original sample-size justification, detailed missing-data procedures, or multiplicity strategy. Results for most named secondary outcomes are not reported in the registry record.