Time in Glucose Hospital Target: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Time in Glucose Hospital Target (NCT05135676), a randomized na study. The primary outcome is Co-primary Outcome: CGM-measured Mean Glucose (Superiority).

2. Clinical Question

In practical terms, this is a comparison of two inpatient management strategies among adults with type 2 diabetes. The questions are related but distinct: whether the intensive strategy yields a different mean glucose during the CGM observation period, and whether lower glucose can be pursued without an undesirable safety trade-off. Mean glucose is the reported continuous co-primary endpoint; serious adverse event counts provide a separate, limited view of safety. The posted record does not report enough detail to characterize all eligibility criteria, the full safety assessment, or any clinical outcome comparison.

3. Trial Design

The registry describes a randomized, single-masked, parallel-group clinical trial, with 166 participants recorded as enrolled. Its title is Time in Glucose Hospital Target (TIGHT). The sponsor is Jaeb Center for Health Research. The record lists the study as completed, with a start date of 2022-05-10 and primary completion date of 2024-03-01. The phase is listed as NA.

Participants were assigned to a standard target strategy or an intensive target strategy. The standard target group used a glucose target of 140-180 mg/dL and masked CGM. The intensive target group used a glucose target of 90-130 mg/dL and real-time CGM. Because the monitoring approach differs alongside the glucose target, the contrast represents the combined management strategies described in the registry, not an isolated test of target range independent of monitoring.

The posted primary outcome data list 50 participants in the standard target group and 60 in the intensive target group for the mean-glucose analysis. These are the group counts attached to that outcome; the registry enrollment entry and outcome-specific counts are distinct record items and should not be treated as interchangeable. The record does not explain the difference between the enrollment entry and the posted group counts.

The primary measurement was CGM mean glucose, in mg/dL, over 4-10 days after randomization, and the registry labels this co-primary outcome as a superiority comparison. The record also lists daytime CGM metrics from 06:00 AM to 00:00 AM and nighttime metrics from 00:00 AM to 06:00 AM as secondary outcomes. It lists eight outcome measures and one analysis. Details such as the randomization procedure, masking roles, treatment protocol, and analysis population 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.

Overall Study

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.

6. Primary Endpoint

The registry lists the primary outcome as Co-primary Outcome: CGM-measured Mean Glucose (Superiority), measured over 4-10 days.

The posted co-primary endpoint is CGM-measured mean glucose, expressed in mg/dL, during 4-10 days following randomization. It summarizes glucose readings across the monitored interval as a single continuous measure for each group. The comparison is framed as a superiority question: whether mean glucose differs between the intensive and standard target groups.

Mean glucose is useful for describing overall glucose exposure during the observed period. It gives a direct summary of the central level of the CGM measurements and allows the groups to be compared on a continuous scale. A difference in means is interpreted in the original unit, mg/dL, rather than as a count or a proportion. It does not describe the full distribution of readings, the time spent in particular glucose ranges, or the frequency and duration of low-glucose episodes.

The record's broader hypothesis refers to achieving a mean glucose target of 90-130 mg/dL without increasing hypoglycemia. That objective involves both glucose levels and safety. The posted mean-glucose estimate addresses the continuous endpoint, while adverse-event counts offer only a separate registry summary of serious and other events. The record does not give a complete posted analysis of the other co-primary endpoint or detailed CGM hypoglycemia measures within the primary outcome description. As a result, mean glucose should be read as one component of the clinical question, not as a complete account of target attainment or safety.

7. Results Posted in the Registry

• Intensive Target Group: mean 170.00 versus 175.00 on control; difference -5.00 (95% CI -17.22 to 7.22), standardized effect d = -0.15; z-test p = 0.422.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary reported comparison concerns a continuous outcome: mean glucose measured by CGM. A continuous outcome can take many values on a scale, here mg/dL. Comparing group means is a way to summarize how the central level of that measure differs between randomized groups. The reported contrast is intensive target group minus standard target group, so a negative mean difference corresponds to a lower mean in the intensive group.

What does the mean difference represent?

The registry reports a mean of 175.0 mg/dL for the standard target group and 170.0 mg/dL for the intensive target group. The posted estimated difference is -5.0 mg/dL. This estimate describes the contrast in group averages, not the change experienced by every participant. Some individuals may have had values above or below their group average, and the group summaries do not show the distribution of individual responses.

The standard deviations, 33.0 and 32.0, describe the spread of individual outcome values around the group means. Standard deviation is not the same as the uncertainty in a group mean or in the difference between means. It describes variability among observations; a confidence interval for the difference describes uncertainty about the estimated group contrast. Keeping these concepts separate matters when reading summaries with a continuous endpoint.

How should the confidence interval be read?

General: a confidence interval gives a range of values compatible with the observed data under the assumptions of the analysis procedure. It communicates uncertainty in the estimate, rather than the range containing most individual participants' glucose values. For this comparison, the interval runs from -17.22 to 7.216 mg/dL. Since it includes zero, the posted results do not tightly distinguish a negative mean difference from no difference or a positive difference. The endpoints are not guarantees about the true value, and the interval does not resolve limitations in outcome measurement or analysis.

The unadjusted p-value is 0.422. General: a p-value measures how incompatible the observed data are with a specified null model, assuming that model and the analysis assumptions hold. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The record calls this value unadjusted; a multiplicity-adjusted value is not reported in the registry record. The p-value should be considered alongside the effect estimate and interval rather than used as a standalone conclusion.

Why can variability affect precision?

General: when outcome values vary widely within groups, a given number of observations may provide less precision about the difference in group means than when values are tightly clustered. The posted standard deviations are much larger than the estimated difference in the original mg/dL scale. That context helps explain why the interval spans values on either side of zero. The data summaries alone do not identify which source of variation contributed most or whether particular subgroups had different patterns.

General: over-dispersion means that observed variability is greater than a model would expect under its assumed variance structure. If it occurs, standard errors and confidence intervals based on an overly restrictive variance assumption can be too small, making results appear more precise than they are. The posted record lists a mean and standard deviation and gives a comparison interval, but does not describe a model assessment for over-dispersion. This concept is therefore relevant as a general statistical consideration, not a claim that over-dispersion was found in this trial.

What does randomization contribute?

General: random assignment aims to make treatment groups comparable, on average, with respect to measured and unmeasured characteristics at the point of assignment. It supports a causal interpretation of a between-group contrast when allocation is implemented as intended and outcome follow-up and analysis are appropriate. The record identifies the study as randomized, but the randomization sequence, concealment, handling of post-randomization events, and analysis population are not reported in the registry record. The design label alone cannot answer those implementation questions.

The comparison is between two bundled strategies. One group had a target of 90-130 mg/dL with real-time CGM; the other had a target of 140-180 mg/dL with masked CGM. Accordingly, a group difference would describe the contrast between these strategies as implemented, not the independent effect of the glucose target or the device feature considered separately. The record does not provide enough protocol detail to further decompose the contrast.

The endpoint was observed over 4-10 days following randomization. That observation window defines the scope of the result. It does not imply that the same difference would persist beyond the monitored inpatient period. The registry does not report participant-level distributions, the amount of missing CGM data, or a detailed model specification. Such information would help assess how the summary was constructed and how robust it is to incomplete or uneven monitoring.

Finally, the trial labels mean glucose as a co-primary outcome and describes a superiority test. A superiority test evaluates whether the data support a difference in the specified direction or scale, under the prespecified testing framework. The registry record does not report the original statistical analysis plan, the full list and hierarchy of co-primary tests, or whether any adjustment across outcomes was used. The interpretation therefore stays with the posted estimate and unadjusted p-value, without assuming an undocumented testing procedure.

10. Confidence Intervals and Precision

The estimated intensive-minus-standard mean difference is -5.0 mg/dL, with a confidence interval from -17.22 to 7.216 mg/dL. This range is central to understanding the precision of the estimate. It covers a lower mean in the intensive group, values close to no difference, and a higher mean in that group. The posted result therefore does not confine the plausible group contrast to a single direction.

A confidence interval describes uncertainty in an estimated group-level contrast. It is not a range of individual glucose values, and it does not mean that every value inside it is equally likely. Its width reflects the information and variability captured by the analysis, along with the assumptions used to construct it. The record provides group standard deviations of 33.0 and 32.0 and outcome counts of 50 and 60; those summaries offer context, but they do not replace the interval.

The interval's inclusion of zero is consistent with the unadjusted p-value of 0.422. This does not demonstrate equivalence or prove that there is no meaningful difference. Establishing equivalence would require a defined margin and a suitable analysis, neither of which is reported in the registry record. Likewise, clinical importance cannot be inferred solely from whether an interval includes zero; it depends on the clinical context and a prespecified threshold for a meaningful difference, which is not reported in the registry record.

The result is also limited to mean glucose over the 4-10-day CGM period. It cannot convey the precision of other outcomes, the time spent in glucose ranges, or the frequency of hypoglycemia. Those measures require their own estimates and uncertainty summaries. The available interval is informative about the mean contrast, but it should not be treated as a general precision statement for every outcome in the trial.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed mean difference, not counts from the completed trial. For this continuous-outcome comparison, the figures are participants per arm, not total events. The planning assumptions listed in the registry are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row uses the stated fraction of the observed effect as the effect to detect. The original trial design assumptions are not reported in the registry record.

The 80% power column describes the number of participants per arm that a planning calculation estimates would be needed to have an 80% chance of detecting the specified effect under the stated assumptions, if that effect were the true effect. The 90% power column gives the analogous planning target for a 90% chance. These are design calculations for hypothetical future studies, not evidence about the probability that the completed trial detected an effect.

The rows consider planning for 100%, 75%, and 50% of the observed effect. A smaller target effect is harder to distinguish from random variation, so the planning calculation calls for more participants per arm as the effect to detect becomes smaller. The posted figures are linked to this trial's observed estimate and variability and should not be taken as universally applicable design recommendations.

This is not observed power. Observed power would attempt to characterize power using the estimated result after the study; it is not a useful substitute for interpreting the estimate and confidence interval. The planning figures instead illustrate how sample-size requirements change under stated assumptions. They do not establish the adequacy of the completed trial's design, since its original design assumptions are not reported in the registry record.

12. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -5.000; pooled standard deviation σ = 32.458.
2. n = 2 × (1.96 + 0.842)² × 32.458² / -5.000² = 661.5, rounded up to 662 participants per arm for 80% power.

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

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -5                      # observed difference in means
sigma <- 32.4575220979782                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(5135676)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 662 (80% power), 886 (90% power)
 75% of observed effect: n per arm = 1,176 (80% power), 1,575 (90% power)
 50% of observed effect: n per arm = 2,646 (80% power), 3,543 (90% power)
power.t.test: 662.5 
Simulated power at n = 662 per arm: 0.800 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates multiple outcomes, time points, groups, or tests. Repeated testing can increase the chance of obtaining at least one apparently unusual result even when no underlying effect exists. Statistical plans may address this through a prespecified hierarchy, adjustment procedures, or other strategies. Which approach is suitable depends on the study's objectives and design.

The registry describes mean glucose as a co-primary outcome and lists daytime and nighttime CGM metrics as secondary outcomes. It also lists eight outcome measures and one analysis. The full outcome hierarchy, the exact set of planned statistical tests, and any multiplicity adjustment are not reported in the registry record. The posted p-value of 0.422 is explicitly unadjusted. It should not be presented as adjusted for multiple outcomes or as the definitive result of an undocumented testing hierarchy.

The record gives a mean difference and confidence interval for the mean-glucose comparison, but does not report the complete statistical analysis plan or identify all analysis populations. It does not state whether the analysis followed intention-to-treat, per-protocol, or another population definition. Details about handling withdrawals, protocol deviations, missing outcome values, and incomplete CGM records are also not reported in the registry record.

Missing data can affect interpretation if the availability of outcome measurements differs between groups or relates to glucose levels. General: methods for incomplete data rely on assumptions about why observations are missing and how missingness relates to observed and unobserved values. Without the number and pattern of missing CGM measurements or the analysis approach, readers cannot assess those assumptions from the registry summary. This does not show that missingness occurred; it means its extent and handling are not reported.

Similarly, the registry does not report whether interim analyses were planned or conducted, nor does it provide the original sample-size justification. These details should not be inferred from the posted planning illustration, which is based on the observed effect and is not the trial's original design rationale. The available results support description of the reported comparison, but not reconstruction of unreported analysis decisions.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: severe hypoglycemia event (2.0%).

The registry's adverse-event summary reports serious adverse events in 1 participant in the standard target group and 0 participants in the intensive target group. The serious-event groups at risk are listed as 50 and 60, respectively. The record lists 0 deaths in each group. For other adverse events, it reports 0 affected participants in both groups, with the same group counts at risk.

The registry also lists “Severe hypoglycemia event” as a serious adverse event term at 2.0%. That entry should be read as a reported serious-event term, not as a complete measure of all hypoglycemia or all low-glucose CGM readings. The adverse-event summary does not provide event timing, duration, recurrence, severity details beyond the listing, or a statistical comparison of the groups.

These counts describe events recorded in the registry, not proof that an event was caused by a particular strategy. Small event counts provide limited information about comparative safety, and the available summary does not show the full safety profile during or after the monitored period. The absence of reported other adverse events or deaths in these entries should not be interpreted as evidence that no such events occurred outside the registry's reported scope.

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

The posted primary comparison shows a mean CGM glucose of 175.0 mg/dL in the standard target group and 170.0 mg/dL in the intensive target group. The estimated mean difference is -5.0 mg/dL for intensive versus standard, with a confidence interval from -17.22 to 7.216 mg/dL. These results describe the observed group averages and the uncertainty around their contrast during 4-10 days after randomization.

The estimate points toward a lower mean in the intensive group, but the interval includes zero and extends in both directions. The result therefore does not establish a clear difference in mean glucose, nor does it establish that the two strategies are equivalent. The unadjusted p-value of 0.422 is one part of the statistical summary; it should be interpreted alongside the estimate, interval, outcome definition, and limitations of the registry detail.

The endpoint does not answer whether participants maintained a specific target throughout monitoring. A group mean can be within or outside a target even when individual readings vary, and the posted summaries do not provide individual distributions or time-in-range values for the primary comparison. Nor does mean glucose alone show whether hypoglycemia increased. The registry's serious adverse event counts are limited summaries and are not a substitute for detailed CGM safety outcomes.

The comparison also cannot isolate the effect of the target range from the effect of real-time versus masked CGM, because those features differ together between groups. The posted result does not establish effects on morbidity, mortality, or other clinical outcomes. The record's rationale identifies such clinical benefit as a question that a larger study could evaluate; it is not demonstrated by this mean-glucose result.

In short, the data provide an estimate of a short-term mean-glucose contrast between two inpatient management strategies. They do not, on their own, settle target attainment, hypoglycemia risk, clinical benefit, or the consequences of using these strategies beyond the recorded observation window.

16. Why This Trial Matters Statistically

This trial addresses a statistical question that is important in inpatient diabetes research: how to compare glucose-management strategies when both the average glucose level and hypoglycemia matter. A lower mean is not the only relevant outcome if a strategy could also alter the distribution of readings or the occurrence of low-glucose events. The registry's co-primary framing reflects that the mean-glucose question sits alongside a safety concern, although the posted mean comparison cannot resolve the full balance of those outcomes.

CGM makes it possible to observe glucose repeatedly during the inpatient period, rather than relying only on a single summary measurement. The posted endpoint condenses those measurements into a mean, which supports a straightforward comparison but necessarily discards information about timing and variation. Daytime and nighttime CGM metrics are listed as secondary outcomes, highlighting that glucose patterns can be considered across different portions of the day, although detailed results for them are not included in the registry record for the primary comparison.

The randomized parallel-group design is also statistically relevant. Randomization aims to support a fair comparison between strategies, while the confidence interval shows the remaining uncertainty in the mean difference. Here, the interval spans values on both sides of zero, making precision as important to the interpretation as the point estimate. Reporting the estimate and interval helps avoid reducing the result to a simple positive-or-negative label.

Finally, the trial's question is not identical to whether a particular glucose level improves patient-centered outcomes. A biomarker or intermediate measure can describe a physiological target, but a clinical benefit requires evidence on clinical outcomes. The registry rationale itself distinguishes feasibility of intensive glucose management from the need for a larger study to evaluate morbidity and mortality. That distinction is statistically important: evidence about a short-term mean is not a substitute for evidence about longer-term or clinical effects.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry summary leaves important details unavailable for interpreting the mean-glucose comparison. It does not report the eligibility criteria in full, baseline characteristics, the randomization procedure, or the roles covered by single masking. It also does not explain how the enrollment entry of 166 relates to the outcome-specific group counts of 50 and 60. Those records should be treated as distinct reported quantities rather than reconciled by assumption.

The intervention contrast combines different glucose targets with different CGM approaches: real-time CGM in the intensive group and masked CGM in the standard group. The result therefore cannot attribute any observed difference solely to target range. Details of insulin regimens, co-interventions, adherence, and treatment adjustments are not reported in the registry record.

The primary observation window is 4-10 days after randomization. The posted mean and standard deviation do not show the distribution of participant-level results, CGM completeness, or time spent in glucose ranges. The registry does not report the analysis population, the handling of missing or incomplete measurements, or sensitivity analyses. It also does not report the full analysis plan or multiplicity procedures, and the p-value is unadjusted.

The safety summary contains few reported serious events and does not provide a detailed comparative safety analysis. The listed serious adverse event term does not capture every form of hypoglycemia that could be measured by CGM. The posted primary result is a glucose measure, not a clinical outcome, and the registry does not report a clinical endpoint comparison in the registry record for this result. These limitations restrict how broadly the result can be interpreted beyond the recorded short-term mean-glucose comparison.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Time in Glucose Hospital Target — NCT05135676.
• Registry record last updated 2026-09-14; status last verified 2026-09.
• Results first posted 2026-09-14.
• 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.

21. Record Summary

The registry identifies NCT05135676 as the randomized, single-masked, parallel-group TIGHT trial, sponsored by Jaeb Center for Health Research. It lists 166 participants enrolled and the study as completed. The comparison is between an intensive target of 90-130 mg/dL with real-time CGM and a standard target of 140-180 mg/dL with masked CGM.

The posted co-primary mean-glucose outcome covers 4-10 days after randomization. The estimated intensive-minus-standard difference is -5.0 mg/dL, with a confidence interval from -17.22 to 7.216 mg/dL and an unadjusted p-value of 0.422. The interval includes zero, so the estimate remains uncertain in direction and magnitude.

The result describes a short-term group-average contrast. The registry record does not provide enough detail to resolve the full analysis plan, missing-data handling, or the complete safety and clinical-outcome questions raised by the trial.