This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Efficacy and Safety of a Glargine-based Hospital Discharge Algorithm in Coronary Artery Bypass Graft (CABG) Patients (NCT01792830), a non-randomized phase 3 study. The primary outcome is Efficacy, Measured by a Change in HbA1c Levels.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01792830 |
| Phase | Phase 3 |
| Condition | Poor Glycemic Control |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 175 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Emory University |
| Results posted | Yes |
2. Clinical Question
Population
Patients undergoing coronary artery bypass graft surgery with perioperative hyperglycemia, including patients with a history of diabetes and patients without a history of diabetes who had persistent hyperglycemia. The registry describes HbA1c-based subgroups, including admission HbA1c <7%, 7% to 9%, and >9%. The primary-outcome population is narrower: participants with blood drawn one month after hospital discharge.
Intervention
A discharge approach that assigns diabetes treatment according to glycemic status and prior treatment. Regimens described in the record include metformin, insulin glargine, and insulin glulisine, alone or in combinations. For patients with diabetes, the described approach varies with admission HbA1c and the insulin regimen used in hospital.
Comparator
The treatment groups are not randomized. They consist of different clinical and HbA1c categories receiving different discharge regimens, rather than exchangeable groups assigned to competing strategies. The registry explicitly cautions that the arms are not a like-for-like control comparison.
Primary question
The primary outcome is the change in HbA1c after discharge, described for measurement one month after discharge. The brief summary also mentions assessment at 4 and 12 weeks. The clinical question is whether HbA1c changes after discharge under the HbA1c-guided discharge approach; the registry record does not report the primary outcome estimates needed to describe the size or precision of any change.
The study is concerned with the transition from hospital-based glucose management to diabetes treatment after discharge. Its design describes a treatment algorithm rather than a randomized test of one regimen against another. The available record therefore supports describing which groups received which regimens and what outcomes were planned or listed, but not a causal comparison of those regimens.
3. Trial Design
This is a Phase 3, non-randomized, parallel study sponsored by Emory University. The registry lists enrollment of 175 and status as completed. The study start is recorded as 2012-10 and primary completion as 2014-02. Masking is listed as none. These design characteristics describe the trial framework, but do not establish that the treatment groups are comparable at baseline.
The study concerns patients undergoing cardiac surgery with perioperative hyperglycemia. The discharge approach is organized around diabetes history, admission HbA1c, and the treatment used in hospital. The record describes continuing a pre-hospital diabetes regimen for some patients, metformin for some patients, and glargine-based regimens, sometimes with glulisine, for other HbA1c groups. It also lists non-diabetic groups receiving no treatment, metformin, or insulin. The descriptions do not establish a randomized assignment among these regimens.
The primary outcome is described as change in HbA1c one month after discharge. The brief summary additionally refers to outcome assessment at 4 and 12 weeks, so the registry’s descriptions of timing are not fully aligned. The primary analysis population is identified as participants who had blood drawn one month after discharge. The record notes that most patients who completed the discharge part did so over the phone, and that one diabetic group with HbA1c <7% receiving metformin and insulin glargine is not included in the primary outcome table because no participants in that group had blood drawn.
Secondary outcomes listed include hospital readmission, hypoglycemic events, severe hypoglycemic events, and hyperglycemic events. The registry record does not report a statistical analysis plan, a sample-size justification, or the primary outcome estimates. With no random allocation and treatment groups defined by clinical characteristics, observed differences between groups would be descriptive and could reflect differences in the patients as well as differences in treatment.
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.
Only the first 6 of 11 groups are shown.
Overall Study
| Stage | Control, Non-diabetic, no Treatment | Non-diabetic, Metformin | Non-diabetic, Insulin | Diabetic, HbA1C <7%, Metformin | Diabetic, HbA1C <7%, Metformin and Insulin Glargine | Diabetic, HbA1C <7%, Insulin Glargine |
|---|---|---|---|---|---|---|
| STARTED | 84 | 3 | 4 | 21 | 2 | 4 |
| COMPLETED | 59 | 3 | 4 | 18 | 1 | 3 |
| NOT COMPLETED | 25 | 0 | 0 | 3 | 1 | 1 |
Reasons for not completing this period:
| Reason | Control, Non-diabetic, no Treatment | Non-diabetic, Metformin | Non-diabetic, Insulin | Diabetic, HbA1C <7%, Metformin | Diabetic, HbA1C <7%, Metformin and Insulin Glargine | Diabetic, HbA1C <7%, Insulin Glargine |
|---|---|---|---|---|---|---|
| Withdrawal by Subject | 22 | 0 | 0 | 2 | 1 | 1 |
| Death | 3 | 0 | 0 | 1 | 0 | 0 |
| Adverse Event | 0 | 0 | 0 | 0 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: This analysis includes participants who had blood drawn one month after hospital discharge. Most patients that completed the discharge part did so over the phone. The "diabetic, HbA1C \<7%, metformin and insulin glargine" is not included in this table as no participants in this group had blood drawn for this analysis.
5. Baseline Characteristics
The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.
Only the first 6 of 12 groups are shown.
| Characteristic | Unit | Control, Non-diabetic, no Treatment | Non-diabetic, Metformin | Non-diabetic, Insulin | Diabetic, HbA1C <7%, Metformin | Diabetic, HbA1C <7%, Metformin and Insulin Glargine | Diabetic, HbA1C <7%, Insulin Glargine |
|---|---|---|---|---|---|---|---|
| Participants analysed | 84 | 3 | 4 | 21 | 2 | 4 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 41 | 0 | 1 | 11 | 0 | 1 |
| Age, Categorical: >=65 years | Participants | 43 | 3 | 3 | 10 | 2 | 3 |
| Sex: Female, Male: Female | Participants | 23 | 0 | 1 | 6 | 1 | 1 |
| Sex: Female, Male: Male | Participants | 61 | 3 | 3 | 15 | 1 | 3 |
| Region of Enrollment: United States | Participants | 84 | 3 | 4 | 21 | 2 | 4 |
6. Primary Endpoint
The registry lists the primary outcome as Efficacy, Measured by a Change in HbA1c Levels, measured over One month after hospital discharge.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | percent of glycosylated hemoglobin |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is change in HbA1c, reported as a percentage of glycosylated hemoglobin. HbA1c is a laboratory measure associated with average blood glucose over the preceding months; the registry description says higher percentages indicate higher blood glucose in the previous three months. The record gives a normal HbA1c level as below 5.7 percent. This endpoint is used to assess glycemic status after discharge and to describe change over time following the discharge regimen.
The stated primary measurement timeframe is one month after hospital discharge. The brief summary also says HbA1c change will be assessed at 4 and 12 weeks. Because those descriptions are not identical, the registry record does not establish a single consistent schedule across all its text. The primary outcome population is defined as those with a blood draw at one month. The record specifically excludes the diabetic group with HbA1c <7% receiving metformin and insulin glargine from the primary outcome table because no participants in that group had blood drawn.
HbA1c is a continuous measure, and the registry identifies the reported summary type as a mean with standard deviation. A mean describes the group’s average value; the standard deviation describes variation among measurements. Neither measure, on its own, gives the uncertainty around a treatment effect. The registry record does not report the actual change estimates, their standard deviations, confidence intervals, or statistical tests. Thus the endpoint’s definition and intended use are available, but the magnitude and precision of the observed change are not reported in the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive constraint is the non-randomized design. Participants were placed in groups distinguished by diabetes history, HbA1c, and discharge regimen; the registry itself cautions that the arms are not a like-for-like control comparison. A difference in HbA1c across these groups would not isolate the effect of a particular drug or regimen. The groups may differ in the clinical characteristics that determined their treatment, and the record does not report an adjusted comparison that accounts for such differences.
The registry identifies a primary continuous outcome and gives a summary format, but does not report primary outcome values or analyses. It is therefore not possible from the registry record to state the average HbA1c change, how variable that change was, whether groups differed, or how precise an estimated difference would be. The record lists no analyses. The planned outcome and its measurement description should not be confused with a result.
The primary outcome population also depends on blood collection after discharge. The record says that the analysis includes participants with a one-month blood draw and identifies a group that had no such blood draws and was therefore excluded. This limits what the primary outcome could describe: it is a result among those with the relevant measurement, not necessarily among everyone assigned to or treated according to the discharge approach. The record does not report the reasons for missing blood draws or a method for accounting for missing outcomes.
The adverse-event information is reported by treatment group as counts of participants at risk and affected, with deaths also listed. These counts are useful as a descriptive account of events recorded in the registry, but they do not support an unadjusted safety ranking of the regimens. Group sizes differ, and the treatment groups are not randomized. The record does not provide an adjusted safety comparison, a confidence interval, or a consistent observation window for these counts. The serious-event terms listed with percentages do not have a stated denominator in the record, so those percentages should not be interpreted as group-specific event risks.
Readmission, hypoglycemic events, severe hypoglycemic events, and hyperglycemic events are named as secondary outcomes, but outcome estimates for these measures are not reported in the registry record. The completed status indicates that the study is marked completed; it does not fill in the missing endpoint estimates or clarify how the results were analyzed. Overall, the record provides design and outcome definitions, selected safety counts, and the reason one group is absent from the primary outcome population. It does not provide enough primary outcome information to estimate the size, direction, or precision of HbA1c change under the discharge strategy.
9. Statistical Methods Explained
The registry classifies the primary measure as continuous and its summary as a mean with standard deviation. A continuous outcome can take values along a measurement scale rather than belonging only to categories. HbA1c, expressed as a percentage of glycosylated hemoglobin, is treated this way. A mean provides a concise description of the group’s central value, while a standard deviation indicates how spread out individual measurements are around that mean. These summaries describe observed measurements; they do not by themselves demonstrate a treatment effect.
What does a change score represent?
A change score compares a later measurement with an earlier measurement for the same measure. In this study, the endpoint is described as change in HbA1c after hospital discharge. The registry does not report the numerical change estimates, so the direction and magnitude cannot be characterized. It also does not specify in the reported information how the change was calculated, such as the exact baseline value used for each participant. That detail is not reported in the registry record.
How should uncertainty be understood?
In general, a confidence interval expresses a range of values compatible with the data and the statistical model used, and indicates the precision of an estimate. A narrow interval generally indicates greater precision than a wide interval. The registry record does not report confidence intervals for the primary endpoint. It also does not report p-values or any other primary-outcome test results. Without an estimate and its uncertainty, the evidence cannot be characterized as precise or imprecise from the registry record alone.
Why does the non-randomized design matter?
Random allocation is one way to make comparison groups more similar, on average, with respect to both measured and unmeasured characteristics. Here, allocation is listed as non-randomized, and the treatment groups are defined by factors such as diabetes history, HbA1c category, and treatment regimen. These factors are also relevant to the clinical context of glucose management. Consequently, any observed group difference could be associated with the reason participants received a particular regimen, not solely with that regimen. General statistical adjustment can address some measured differences when appropriate data and a prespecified analysis are available, but the registry record does not report an adjusted primary analysis or its covariates.
What do the missing blood draws imply?
The primary outcome population is described as participants with blood drawn one month after discharge. The record specifically notes that no participants in the diabetic HbA1c <7% metformin-and-glargine group had blood drawn, so that group is not included in the primary outcome table. In general, an analysis restricted to participants with observed measurements can be affected if measurement availability is related to health status, treatment, or the outcome itself. The registry does not report why blood draws were unavailable, whether missingness was assessed, or whether any method was used to address it. It is therefore not possible to determine from the record whether the measured participants represent all participants eligible for that endpoint.
How do the safety summaries differ from the primary endpoint?
The primary endpoint is a continuous laboratory measure summarized by a mean and standard deviation. Safety information is presented as counts of participants at risk and affected, and deaths are listed separately. These are descriptive summaries of event occurrence, not estimates from a reported comparative model. In general, a count needs its relevant group denominator and observation period to be interpreted as a risk; the registry includes counts at risk for each group but does not state the observation window for the safety figures. It also does not report comparative uncertainty or adjustment for differences among groups.
What can be said about planning calculations?
In general, a sample-size calculation is a planning exercise based on assumptions about the effect to be detected, variability, significance level, allocation, and model. A planning illustration is not observed power and does not describe the probability that the completed study detected an effect. The registry’s planning description refers to a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as fractions of an observed effect. It says that non-time-to-event calculations are expressed as participants per arm and time-to-event calculations as total events. These assumptions do not substitute for a reported trial design justification, and the registry record does not report such a justification or primary-outcome analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary HbA1c outcome and several secondary outcomes: hospital readmission, hypoglycemic events, severe hypoglycemic events, and hyperglycemic events. The brief summary mentions HbA1c assessments at 4 and 12 weeks, whereas the primary outcome description gives one month after discharge. The record does not explain whether those timepoints were treated as separate primary analyses, how they were prioritized, or whether adjustment for multiple testing was planned. Such multiplicity matters in general because testing several outcomes or timepoints can increase the chance of a statistically notable result arising by chance. No multiplicity strategy is reported in the registry record.
The registry record lists no analyses and does not report primary outcome estimates, p-values, confidence intervals, or results for the listed secondary outcomes. It therefore does not permit a review of the actual analysis choices or any adjustment for multiple comparisons. The record also does not report a statistical analysis plan, the approach to model selection, or whether subgroup analyses were planned.
For the primary outcome, the stated analysis population includes participants who had blood drawn one month after discharge. A diabetic HbA1c <7% group receiving metformin and insulin glargine is omitted because no participants in that group had a blood draw. The record does not report the reasons for absent measurements, the amount of missing data in other groups, or whether an imputation or sensitivity analysis was used. In general, restricting analysis to observed outcomes can create bias if the availability of a measurement is related to the outcome or treatment. The record does not provide enough information to assess that possibility.
The record also does not specify how participants were handled if they did not follow their discharge regimen or did not complete follow-up. The analysis population beyond the blood-draw description is not reported in the registry record. As a result, it is not possible to determine whether the outcome description reflects all enrolled participants, those who received treatment, or another analysis set.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Control, Non-diabetic, no Treatment | 1 | 84 | 1.2% |
| Non-diabetic, Metformin | 0 | 3 | 0.0% |
| Non-diabetic, Insulin | 0 | 4 | 0.0% |
| Diabetic, HbA1C <7%, Metformin | 0 | 21 | 0.0% |
| Diabetic, HbA1C <7%, Metformin and Insulin Glargine | 0 | 2 | 0.0% |
| Diabetic, HbA1C <7%, Insulin Glargine | 0 | 4 | 0.0% |
| Diabetic, HbA1C 7%- 9%, Metformin | 2 | 14 | 14.3% |
| Diabetic, HbA1C 7%-9%, Metformin and Insulin Glargine | 0 | 7 | 0.0% |
| Diabetic, HbA1C 7%-9%, Insulin Glargine | 0 | 20 | 0.0% |
| Diabetic, HbA1C >9%, Metformin and Insulin Glargine | 1 | 4 | 25.0% |
| Diabetic, HbA1C >9%, Insulin Glulisine | 0 | 12 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Control, Non-diabetic, no Treatment | 5 of 84 |
| Non-diabetic, Metformin | 0 of 3 |
| Non-diabetic, Insulin | 1 of 4 |
| Diabetic, HbA1C <7%, Metformin | 0 of 21 |
| Diabetic, HbA1C <7%, Metformin and Insulin Glargine | 0 of 2 |
| Diabetic, HbA1C <7%, Insulin Glargine | 0 of 4 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Control, Non-diabetic, no Treatment | 6 of 84 |
| Non-diabetic, Metformin | 1 of 3 |
| Non-diabetic, Insulin | 0 of 4 |
| Diabetic, HbA1C <7%, Metformin | 1 of 21 |
| Diabetic, HbA1C <7%, Metformin and Insulin Glargine | 0 of 2 |
| Diabetic, HbA1C <7%, Insulin Glargine | 1 of 4 |
Most frequent serious adverse event terms
| Event term | Organ system | Control, Non-diabetic, no Treatment | Non-diabetic, Metformin | Non-diabetic, Insulin | Diabetic, HbA1C <7%, Metformin | Diabetic, HbA1C <7%, Metformin and Insulin Glargine | Diabetic, HbA1C <7%, Insulin Glargine |
|---|---|---|---|---|---|---|---|
| Cardiac disorder resulting in inpatient death | Cardiac disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Cardiac disorder resulting in death during hospital readmission | Cardiac disorders | 0 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Control, Non-diabetic, no Treatment | Non-diabetic, Metformin | Non-diabetic, Insulin | Diabetic, HbA1C <7%, Metformin | Diabetic, HbA1C <7%, Metformin and Insulin Glargine | Diabetic, HbA1C <7%, Insulin Glargine |
|---|---|---|---|---|---|---|---|
| Trauma from car accident | Social circumstances | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Hematoma | Blood and lymphatic system disorders | 0 of 84 | 1 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Bladder cancer | Renal and urinary disorders | 0 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 1 of 4 |
| Cholecystectomy | Hepatobiliary disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Lower gastrointestinal bleeding | Gastrointestinal disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Hypotension | Vascular disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Seizure | General disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Nephrolithiasis | Renal and urinary disorders | 1 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
| Gout | Musculoskeletal and connective tissue disorders | 0 of 84 | 0 of 3 | 0 of 4 | 1 of 21 | 0 of 2 | 0 of 4 |
| Stomach virus | Gastrointestinal disorders | 0 of 84 | 0 of 3 | 0 of 4 | 0 of 21 | 0 of 2 | 0 of 4 |
Serious events with the highest arm-level rate: cardiac disorder resulting in inpatient death (25.0%), cardiac disorder resulting in death during hospital readmission (25.0%).
The registry reports serious and other adverse-event counts by treatment group, alongside the number at risk and the number of deaths. These are descriptive group-level figures. For example, the control non-diabetic no-treatment group lists 1 participant affected by a serious event among 84 at risk, and 6 affected by another event category among 84 at risk; 5 deaths are listed. The diabetic HbA1c 7%- 9% metformin group lists 2 participants affected by serious events among 14 at risk and 1 death. The diabetic HbA1c >9% metformin-and-glargine group lists 1 participant affected by serious events among 4 at risk and 2 affected by other events among 4 at risk, with no deaths listed.
Other groups have zero affected participants in one or both event categories, while some have nonzero counts. These figures should not be used to rank regimens: allocation was non-randomized, the groups differ in size and clinical definition, and the registry record does not report an adjusted comparison or uncertainty estimates. The observation period for the safety counts is not reported in the registry record. Death counts are listed separately and should not be added to event counts.
The serious-event terms include cardiac disorder resulting in inpatient death and cardiac disorder resulting in death during hospital readmission, each listed as 25.0%. The denominator for these percentages is not reported in the registry record. They should therefore be read as listed event-term percentages, not as rates for a treatment group or as participant-level risks.
12. What the Result Does, and Does Not, Mean
The record describes a clinically structured discharge approach: diabetes regimen after cardiac surgery is linked to diabetes history, admission HbA1c, and treatment needs in hospital. It also identifies HbA1c change as the primary outcome and specifies that the primary analysis population consists of participants with a one-month blood draw. This information helps define the intended clinical question and the way the registry organizes treatment groups.
It does not establish that one discharge regimen is better than another. Participants were not randomized, and the registry warns that the arms are not a like-for-like control comparison. The groups differ by clinical characteristics that help determine their treatment. Any contrast between them could therefore reflect those differences as well as treatment. A causal treatment conclusion would require a suitable comparison and analysis that the registry record does not report.
The record also does not provide the primary HbA1c change estimates or their uncertainty. Without those results, it is not possible to describe how much HbA1c changed, whether any change was statistically distinguishable from no change, or how precise an estimate was. The named secondary outcomes likewise have no reported estimates in the registry record, although adverse-event counts are listed separately.
The safety information should be interpreted as descriptive counts within the registry’s treatment groups, not as proof of comparative safety. Non-random allocation, group-size differences, the absence of a reported observation window, and lack of adjusted estimates limit cross-group interpretation. The listed deaths are separate counts, and the event-term percentages do not have a stated denominator.
Thus, the registry record is informative about study design, treatment categories, the intended endpoint, and some recorded safety counts. It is not sufficient to quantify the primary outcome, establish a treatment effect, assess its precision, or support a comparative safety conclusion. These are limits of what the registry record reports, rather than evidence that the outcome did or did not occur.
13. Why This Trial Matters Statistically
This study illustrates why clinical treatment groups and statistical comparison groups are not always the same thing. The discharge regimens were tied to patient characteristics, including diabetes history and HbA1c category. That organization may be clinically coherent, but it also means that the groups are not interchangeable by design. A comparison of their outcomes cannot automatically be interpreted as the effect of the medication differences alone.
The primary endpoint also depends on post-discharge measurement. The registry identifies the participants with a blood draw as the analysis population and notes that one group had no blood draws. This makes outcome availability part of the interpretation, not merely a reporting detail. A reader needs to know who contributed measurements, why others did not, and how missing values were handled to understand the population to which an estimate applies. Those details are not fully reported in the registry record.
Another statistical issue is the distinction between an endpoint definition and an endpoint result. The record gives a clear type of measure—mean change in HbA1c with standard deviation—but does not report the numerical outcome estimates. A specified endpoint does not itself reveal the observed response or its precision. Likewise, a completed status and an enrollment figure do not replace an analysis of outcomes.
The adverse-event summaries show why denominators and study context matter. Counts across different treatment groups may be useful for describing what was recorded, but group sizes and clinical profiles vary. Without randomization, a stated observation period, or adjusted comparisons, apparent differences in counts cannot be attributed to treatment. The record’s event-term percentages also lack a reported denominator.
For biostatistical interpretation, the key lesson is to keep design, measurement, analysis population, and results separate. This record supplies important context about the first three, while primary outcome estimates and analysis details are not reported. That distinction prevents a planned question or descriptive safety count from being mistaken for a demonstrated treatment comparison.
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.
Interpretation is limited first by the non-randomized allocation. The registry explicitly states that the arms are not a like-for-like control comparison. Treatment groups differ by diabetes history, HbA1c category, and discharge regimen, so unadjusted differences would be vulnerable to confounding. The registry record does not report an adjusted primary comparison or the covariates that might have been considered.
The primary outcome reporting is incomplete. The endpoint is defined as HbA1c change, with a mean and standard deviation, but the registry record does not report the outcome values, uncertainty estimates, or statistical tests. The description of timing also varies: one month after discharge in the primary outcome entry and 4 and 12 weeks in the brief summary. The record does not explain how those timings relate to one another.
The primary analysis population is based on availability of a blood draw. One diabetic group with HbA1c <7% receiving metformin and insulin glargine had no participants with a blood draw and is excluded. The reasons for this absence, missingness in other groups, and any approach to missing data are not reported in the registry record. This may limit representativeness of the endpoint population.
The treatment descriptions are not fully uniform across the registry narrative and arm descriptions. Some arm descriptions specify particular dose proportions or alternative regimens, while the broader summary gives grouped descriptions. The registry record does not report a detailed implementation or adherence analysis. It also does not provide a statistical analysis plan, sample-size justification, or interim-analysis information.
Safety counts lack a stated observation window and adjusted comparisons. The serious-event terms are accompanied by percentages whose denominator is not reported in the registry record. These limits make the safety figures descriptive rather than comparative.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Efficacy and Safety of a Glargine-based Hospital Discharge Algorithm in Coronary Artery Bypass Graft (CABG) Patients — NCT01792830.
- Registry record last updated 2018-09-24; status last verified 2018-08.
- Results first posted 2015-10-29.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The record describes a completed, Phase 3, non-randomized, parallel study sponsored by Emory University, with enrollment of 175 and no masking. It evaluates an HbA1c-guided discharge approach for patients with perioperative hyperglycemia after cardiac surgery. Listed regimens include metformin, glargine, and glulisine, used in different combinations according to diabetes history and HbA1c category.
The primary endpoint is change in HbA1c, summarized as a mean with standard deviation. Its primary-outcome description specifies a blood draw one month after discharge, while the brief summary also refers to 4 and 12 weeks. A diabetic HbA1c <7% group receiving metformin and insulin glargine is excluded from the primary outcome population because no participants in that group had blood drawn.
The registry lists secondary outcomes concerning readmission and glucose-related events and provides adverse-event counts by group. It does not report primary outcome estimates, outcome analyses, a statistical analysis plan, or a sample-size justification. Because allocation was non-randomized and groups were not like-for-like, the record does not support a causal comparison of the discharge regimens.