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 Early Insulin and Development of ARDS (NCT00605696), a randomized phase 2 study. The primary outcome is Plasma Levels of Free Fatty Acids, Tumor Necrosis Factor-α, Interleukin-6, and Von Willebrand Factor Antigen.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00605696 |
| Phase | Phase 2 |
| Condition | Respiratory Distress Syndrome, Adult, Sepsis, Hyperglycemia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 18 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Albert Einstein College of Medicine |
| Results posted | Yes |
2. Clinical Question
Population
Hospitalized patients with high blood sugar and severe infections; further eligibility details are not reported in the registry record.
Intervention
Insulin targeting glucose 80-110 mg/dl within 6-12 hours after presentation to the ED.
Comparator
Insulin targeting 150-180 mg/dl for 48 hours after ICU admission, followed by usual clinical care.
Primary question
The registry lists plasma levels of free fatty acids, tumor necrosis factor-α, interleukin-6, and von Willebrand factor antigen, measured at Day 1, 3 and 7.
The study asks whether an early insulin strategy aimed at a lower glucose target is associated with differences in the listed laboratory measures compared with the control strategy. The registry describes the broader motivation as examining whether early insulin infusions can help prevent acute lung injury or acute respiratory distress syndrome in hospitalized patients with high blood sugar and severe infections. It also lists Murray Lung Injury Score as a secondary outcome.
The registry does not report results for the primary endpoint: it states that blood samples were collected, but no bioassays were performed and no data were collected from participants. Therefore, the record cannot establish whether the groups differed in the listed biomarkers, or whether either strategy changed the occurrence of acute lung injury or acute respiratory distress syndrome. Eligibility criteria beyond the brief population description, and the operational definition of high blood sugar or severe infection, are not reported in the registry record.
3. Trial Design
This was a randomized, parallel-group phase 2 study with no masking. The registry lists enrollment of 18 and status as completed. It names Albert Einstein College of Medicine as sponsor. The recorded study start is 2008-04 and primary completion is 2011-12. These design details describe how the study was organized; they do not supply an estimate of treatment effect.
The two arms differ in their insulin targets and timing. Participants in the early insulin group were to receive insulin targeting glucose 80-110 mg/dl within 6-12 hours after presenting to the ED. Participants in the control group were to receive insulin targeting 150-180 mg/dl for 48 hours after ICU admission, followed by usual clinical care. The registry identifies insulin as the intervention. It does not report details of insulin administration, glucose monitoring, adherence, or what usual clinical care entailed.
Random allocation is a design feature intended, in general, to reduce systematic differences between groups at study entry. It does not ensure that groups will be alike in every characteristic, especially when enrollment is limited. The record does not report the allocation method, whether allocation was concealed, or how many participants were assigned to each arm in the main outcome analysis. The adverse-event record lists 7 participants at risk in the early insulin group and 11 in the control group.
The primary endpoint names four plasma measures and specifies assessment at Day 1, 3 and 7. The record also lists Murray Lung Injury Score as a secondary outcome. It does not report the protocol, a sample-size justification, planned interim analyses, or a detailed statistical analysis plan. Although study status is completed, the absence of primary endpoint data means completion should not be read as evidence that the intended laboratory comparison was carried out.
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
| Stage | Early Insulin Group | Control Group |
|---|---|---|
| STARTED | 7 | 11 |
| COMPLETED | 7 | 11 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: Although blood samples were collected, there were no bioassays perfomed and therefore no data were collected from any study participants.
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.
| Characteristic | Unit | Early Insulin Group | Control Group | Total |
|---|---|---|---|---|
| Participants analysed | 7 | 11 | 18 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 2 | 4 | 6 |
| Age, Categorical: >=65 years | Participants | 5 | 7 | 12 |
| Age, Continuous | years | 74 (SD 21) | 70 (SD 18) | 72 (SD 19) |
| Sex: Female, Male: Female | Participants | 7 | 5 | 12 |
| Sex: Female, Male: Male | Participants | 0 | 6 | 6 |
| Region of Enrollment: United States | participants | 7 | 11 | 18 |
6. Primary Endpoint
The registry lists the primary outcome as Plasma Levels of Free Fatty Acids, Tumor Necrosis Factor-α, Interleukin-6, and Von Willebrand Factor Antigen, measured over Measured at Day 1, 3 and 7.
| Element | Registry entry |
|---|---|
| Result type | N/A |
| Measure of spread | not applicable |
| Unit | not stated |
| Endpoint classification | not reduced to a single comparison |
The registry's primary endpoint is titled “Plasma Levels of Free Fatty Acids, Tumor Necrosis Factor-α, Interleukin-6, and Von Willebrand Factor Antigen.” It specifies measurements at Day 1, 3 and 7. These are laboratory measurements rather than a direct count of a clinical event such as development of acute respiratory distress syndrome. In general, a laboratory endpoint records a measured characteristic in biological samples; its interpretation depends on how the samples are collected, tested, and analyzed.
The endpoint could have described levels of the named substances at the listed assessment times. However, the registry states that blood samples were collected, no bioassays were performed, and no data were collected from any study participants. Thus, there are no reported values, group summaries, estimates of between-group differences, or uncertainty intervals for this endpoint. The registry does not explain the specific rationale for choosing these measures, how they were assayed, or how results across the substances and assessment times would have been combined or interpreted. Those details are not reported in the registry record.
The registry separately lists Murray Lung Injury Score as a secondary outcome, but does not provide a result for it in the posted record. No inference about the primary endpoint or the secondary outcome can be made from the endpoint titles alone.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The central interpretive fact is that the registry reports no data for the primary endpoint. Although blood samples were collected, the planned bioassays were not performed. Consequently, the record contains no measured levels of free fatty acids, tumor necrosis factor-α, interleukin-6, or von Willebrand factor antigen for comparison between the randomized groups. There is no estimate of the size or direction of a group difference and no measure of its precision.
This absence is distinct from a result showing similar groups. Similarity would require observed data and an analysis capable of estimating how compatible the results are with different possible effects. Here, there are no primary endpoint values to compare. The record therefore does not demonstrate either a difference or a lack of difference between insulin strategies on the listed laboratory measures.
The registry describes a prevention question concerning acute lung injury or acute respiratory distress syndrome, but the primary endpoint as titled consists of plasma measurements. Murray Lung Injury Score is listed as a secondary outcome. No findings for that score or for development of acute respiratory distress syndrome are reported in the record. It is not possible to substitute the adverse-event counts for these missing efficacy outcomes: adverse-event summaries answer a different question and do not supply the missing biomarker or lung-injury results.
The safety listing reports 0 serious adverse events among 7 participants at risk in the early insulin group and 0 among 11 in the control group. It lists other adverse events affecting 2 in the early insulin group and 0 in the control group, and deaths of 0 in both groups. These are descriptive counts in a small study, not evidence that the strategies have equivalent safety. The registry does not provide a statistical comparison, details of event types for the other adverse events, or the ascertainment period.
Randomization is relevant to how a comparative study is designed, but it cannot compensate for the lack of endpoint observations. The record does not provide enough information to evaluate whether the intended comparison was analyzed as planned, whether outcome data were available for the secondary endpoint, or whether any adjusted analysis was conducted. The statistical interpretation should therefore remain limited to the reported design and adverse-event counts, alongside the explicit statement that primary endpoint data were not collected.
9. Statistical Methods Explained
Statistical methods determine how collected observations are converted into estimates and how uncertainty around those estimates is described. In this study, the key limitation precedes the choice of a test or model: the registry says no bioassays were performed and no primary endpoint data were collected. Without measured outcome values, an analysis cannot estimate a group contrast for the named plasma measures. The posted record reports no such estimate or inferential result.
What randomization contributes
In general, random allocation is used to create comparison groups without deliberately assigning participants according to characteristics that may be related to the outcome. This can reduce selection bias and, under suitable conduct, supports a causal interpretation of a between-group contrast. Randomization does not guarantee that baseline characteristics will be balanced in a particular study. Nor does it prevent bias from differences in care, outcome measurement, follow-up, or analysis after assignment. The registry states that allocation was randomized, but the method of sequence generation and allocation concealment is not reported in the registry record.
The study was unmasked. In general, masking can reduce the chance that knowledge of assigned treatment affects care or outcome assessment. Whether masking is practical or important depends on the intervention and the endpoint. The record does not describe outcome-assessor masking or explain how the absence of masking was handled. For a laboratory endpoint, measurement procedures may be standardized, but the registry does not report assay procedures or whether laboratory personnel were unaware of treatment assignments.
What an effect estimate and confidence interval would mean
In general, an effect estimate expresses the observed contrast between groups using a chosen scale, such as a difference in average measurements or a difference in the proportion experiencing an event. The scale should match the endpoint and analysis plan. A confidence interval describes uncertainty around an estimate under the assumptions of the analysis. It does not repair missing measurements, and it cannot be calculated for the primary endpoint from the information in the registry because no participant assay results were collected. No primary endpoint effect estimate, confidence interval, or p-value is reported.
The listed endpoint spans multiple substances and assessment times. A statistical plan would ordinarily specify which contrasts are primary, how repeated measurements are handled, and whether results are adjusted or otherwise combined. Those decisions affect what an estimate means. The registry does not report those planned methods. It would be inappropriate to infer a particular model, test, or summary method from the endpoint title alone.
Why sample size and precision are different questions
In general, the amount of information in a study affects the precision of estimates. A relatively small study may produce estimates with substantial uncertainty, even when assignment is randomized. The enrollment listed for this study is 18, and the adverse-event record gives group counts of 7 and 11. These counts do not establish the precision of the unreported biomarker outcomes, because there are no assay values from which to estimate outcome variability or a treatment contrast.
Planning calculations are not observed results. A sample-size illustration describes how many participants might be needed under specified assumptions; it does not state how many participants were observed to benefit, and it is not “observed power.” The registry does not report the trial's own sample-size justification. It also does not report assumptions for a planned primary endpoint calculation or a prospective power analysis. In general, a post hoc calculation based on an observed effect is not a substitute for an effect estimate with its uncertainty, and it cannot create evidence where outcome measurements are absent.
Why over-dispersion can matter in general
For outcomes analyzed as counts, statistical models may assume a particular relationship between the average count and its variability. Over-dispersion means the observed variability is greater than the model assumes. If ignored, standard errors can be too small and confidence intervals too narrow. This is a general issue for count outcomes; the registry does not report a count model, an over-dispersion assessment, or a method that was used for the primary endpoint. It would not be justified to state that over-dispersion was present in this study.
What is and is not documented
The registry lists 2 outcome measures and 0 analyses. It gives a primary endpoint title and a secondary outcome title, but it does not report the statistical methods used to analyze either outcome. Details such as analysis populations, handling of repeated assessments, covariate adjustment, missing-data procedures, and treatment of protocol deviations are not reported in the registry record. These omissions limit reproducibility and prevent readers from evaluating whether the intended comparisons were prespecified and conducted as planned.
Good interpretation separates study design, observed data, and analysis. Randomization describes assignment; it is not itself a result. An endpoint definition describes what investigators intended to measure; it is not an observed measurement. A test or confidence interval can characterize collected data; neither can be reconstructed when the relevant outcome data were not obtained. For this record, the clearest statistical conclusion is that the listed primary biomarker comparison cannot be evaluated from the reported results.
10. Multiplicity, Missing Data and Analysis Populations
The primary endpoint names four plasma measures and three assessment times, and the registry also lists Murray Lung Injury Score as a secondary outcome. Considering multiple measures or times can raise multiplicity concerns: when many hypotheses are tested, the chance of at least one apparently unusual result can increase. General approaches include prespecifying a primary contrast, adjusting for multiple testing, or treating analyses as exploratory. The registry does not report whether any such approach was planned or used, so no correction method should be assumed.
The lack of primary endpoint measurements is more fundamental than multiplicity. The registry states that no bioassays were performed and no data were collected from participants for the primary endpoint. It does not report results for the secondary outcome. It also does not specify whether other outcome data were collected or analyzed. The number and structure of statistical tests actually performed are not reported in the registry record.
Missing-data methods are relevant when some outcome observations are absent among participants who otherwise contribute data. Here, the primary endpoint statement says that no data were collected, rather than describing a subset of missing assay results. The record does not report a missing-data strategy, imputation method, sensitivity analysis, or reasons for the absence of assays. The analysis population—such as whether analyses were intended to follow randomized assignment—is also not reported in the registry record. As a result, readers cannot assess how participant inclusion, missing outcomes, or deviations from assigned strategies might have affected an efficacy analysis.
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 |
|---|---|---|---|
| Early Insulin Group | 0 | 7 | 0.0% |
| Control Group | 0 | 11 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Early Insulin Group | 0 of 7 |
| Control Group | 0 of 11 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Early Insulin Group | 2 of 7 |
| Control Group | 0 of 11 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Early Insulin Group | Control Group |
|---|---|---|---|
| Hypoglycemia | Endocrine disorders | 2 of 7 | 0 of 11 |
The registry lists 0 serious adverse events among 7 participants at risk in the early insulin group and 0 among 11 in the control group. It lists other adverse events affecting 2 participants at risk in the early insulin group and 0 in the control group. Deaths are listed as 0 in each group. These are counts, not estimates adjusted for follow-up time or other participant characteristics.
Zero recorded serious adverse events does not establish that serious events could not occur, nor does it demonstrate equivalent safety between groups. In general, a small number of observed events provides limited information about uncommon outcomes. The record does not report the observation period for these safety summaries, the definitions used for serious and other events, or details about the two other adverse events. It also does not report a statistical comparison of the groups or an assessment of whether events were considered related to insulin.
These safety figures should be read separately from the absent primary endpoint results. They describe the adverse-event information recorded in the registry, but they do not answer whether the biomarker measures differed or whether acute lung injury or acute respiratory distress syndrome developed.
12. What the Result Does, and Does Not, Mean
The registry documents a randomized phase 2 comparison of two insulin strategies, along with a planned set of plasma measures and a secondary lung-injury score. It also makes clear that the planned primary laboratory assays were not performed and that no data were collected from participants for that endpoint. This is the main limit on what can be concluded: there is no reported primary outcome result to interpret.
The record does not show that the early insulin strategy changed the listed plasma measures, and it does not show that the two strategies produced the same measurements. Neither conclusion follows without observed values and an analysis. The study's stated prevention motivation likewise cannot be resolved by the posted primary outcome information. No result for the secondary Murray Lung Injury Score is reported, and the registry does not present a comparison of acute lung injury or acute respiratory distress syndrome occurrence.
The adverse-event counts give a limited descriptive account: no serious adverse events and no deaths are listed in either group, while other adverse events are listed for 2 participants in the early insulin group and none in the control group. Because the groups are small and event details and observation periods are not reported, these counts do not establish a safety difference or equivalence.
In general, randomized design can strengthen a comparison when outcomes are measured and analyzed appropriately. It cannot answer an endpoint question in the absence of endpoint data. The defensible reading is therefore narrow: the registry records the design, intervention targets, and safety counts, but it does not report an efficacy estimate for the primary endpoint. Any conclusion about the biomarker effects or prevention of lung injury would go beyond the posted results.
13. Why This Trial Matters Statistically
This study illustrates an important distinction between a registered research question and an answer supported by data. A record can identify a primary endpoint, planned measurement times, and randomized treatment groups, yet still fail to provide the observations needed to estimate the intended contrast. Here, the registry explicitly states that samples were collected but no bioassays were performed, leaving the primary biomarker endpoint without participant data.
The example also shows why endpoint status matters when interpreting a trial. The title and rationale refer to prevention of acute lung injury or acute respiratory distress syndrome, while the named primary outcome consists of plasma measures. A secondary outcome is listed, but no result for it appears in the registry record. Those distinctions make it important to avoid treating a study's objective, its chosen measurements, and its reported results as interchangeable.
Randomization is a meaningful design strength in general, but design quality and evidence availability are separate considerations. The study is identified as randomized and parallel, yet those features cannot generate an efficacy estimate without measured outcomes. Likewise, adverse-event counts can describe recorded events but cannot stand in for the missing biomarker or lung-injury outcomes.
For clinical researchers and statisticians, the record underscores the value of transparent reporting about whether planned measurements were completed, what analyses were conducted, and how missing outcomes were handled. Such information helps readers distinguish an inconclusive or unavailable result from evidence of no effect. In this case, the record supports an account of study design and limited safety reporting, not a conclusion about the primary 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.
The most consequential limitation is stated directly in the registry: although blood samples were collected, no bioassays were performed and no primary endpoint data were collected from participants. This precludes estimation of group differences in the named plasma measures. The registry does not explain why the assays were not conducted or whether any assay results exist outside the posted record; those details are not reported in the registry record.
The registry lists 18 participants enrolled, while the adverse-event record gives 7 at risk in the early insulin group and 11 in the control group. It does not provide a participant flow, reasons for nonparticipation or withdrawal, or the analysis population for efficacy outcomes. The record also does not report baseline characteristics, adherence to insulin targets, glucose measurements, or deviations from assigned strategies. Without those details, readers cannot assess how treatment exposure or group characteristics may have influenced outcomes.
Details of the primary endpoint are limited to its title and assessment at Day 1, 3 and 7. Assay procedures, definitions, planned summaries, handling of repeated measurements, and a statistical analysis plan are not reported. The registry also does not give results for the secondary Murray Lung Injury Score. Safety reporting lacks event descriptions, ascertainment duration, and a stated relationship to treatment.
The recorded absence of masking and the randomized parallel design are known, but details of sequence generation, allocation concealment, and outcome-assessor masking are not reported. The registry does not provide the trial's sample-size justification or planned interim analyses. Together, missing outcome results and limited methodological detail make the posted record insufficient for evaluating efficacy or drawing a broad safety comparison.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Early Insulin and Development of ARDS — NCT00605696.
- Registry record last updated 2018-05-07; status last verified 2018-04.
- Results first posted 2017-04-10.
- 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
NCT00605696, titled “Early Insulin and Development of ARDS,” is a completed randomized, unmasked, parallel-group phase 2 study sponsored by Albert Einstein College of Medicine. Enrollment is listed as 18; the recorded study start is 2008-04 and primary completion is 2011-12.
The study compared an early insulin target of 80-110 mg/dl within 6-12 hours after ED presentation with a control strategy targeting 150-180 mg/dl for 48 hours after ICU admission, followed by usual clinical care. The primary endpoint named plasma levels of free fatty acids, tumor necrosis factor-α, interleukin-6, and von Willebrand factor antigen, measured at Day 1, 3 and 7. Murray Lung Injury Score was listed as a secondary outcome.
The registry states that no bioassays were performed and no primary endpoint data were collected. Safety listings report no serious adverse events or deaths in either group, and other adverse events affecting 2 participants in the early insulin group and none in the control group. Efficacy results and further analysis details are not reported in the registry record.