Low Indexes of Metabolism - Information to Teams (LIMIT): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Low Indexes of Metabolism - Information to Teams (LIMIT) (NCT02476578), a randomized na study. The primary outcome is Death From Any Cause.

2. Clinical Question

3. Trial Design

LIMIT was a randomized, parallel-group trial with no masking. The registry lists enrollment as 8584 and identifies Clalit Health Services as the sponsor. The study started in 2015-10, reached primary completion in 2016-11 and is marked completed. Its phase is recorded as NA. These registry entries describe the design and administrative record, not the details of how randomization was generated or implemented.

One group was assigned to an email alert for primary care providers about low BMI, HbA1c% or cholesterol, with advice to consider dietary and medical revision. The control group was assigned to receive no email. The registry does not describe masking, which is consistent with its explicit NONE entry; it does not report whether outcome ascertainment was conducted by assessors unaware of group assignment.

The primary outcome was death from any cause during 1 year, reported as participants with an event. The registry presents 285 deaths among 4269 participants in the Intervention Email group and 245 among 4218 in Control. These are the group-level outcome counts used in the posted comparison. The enrollment entry and the arm counts are separate registry entries; details explaining their relationship are not reported in the registry record.

The study purpose was to assess whether alerting primary care providers about low metabolic measures would affect treatment and improve overall survival and other health indexes. That statement describes the study objective, not a finding. The record does not report the original sample-size justification, the randomization procedure, allocation concealment, follow-up procedures, or the trial's prespecified statistical analysis plan. It also does not provide further participant characteristics or eligibility criteria beyond the population description in the title and summary.

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: Some of the participants may fit more than one criterion, as explained in the Eligibility section. Hence 7 subgroups are created: A) Fit only alert 1. B) Fit only alert 2. C) Fit only alert 3. D) Fit both alert 1. and alert 2. E) Fit both alert 1. and alert 3. F) Fit both alert 2. and alert 3. G) Fit all alerts: 1. , 2. and 3.

Overall Study

Reasons for not completing this period:

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 Death From Any Cause, measured over 1 year.

The primary endpoint was death from any cause, recorded as the number of participants who died during 1 year. An all-cause mortality outcome counts deaths without restricting the endpoint to a particular cause. This makes it a broad measure of survival and avoids relying on assigning individual deaths to a specific cause. The registry describes the endpoint as an impact on overall survival.

The reported event counts are 285 in the Intervention Email group and 245 in Control. The corresponding group sizes are 4269 and 4218. The posted risk estimates and intervals compare the probability of this recorded event between the randomized groups. They do not establish why a death occurred or whether the email changed any particular aspect of care.

As a general statistical point, an outcome needs a defined observation period and a clear event definition to support an interpretable comparison. The registry supplies a 1 year timeframe and defines the primary event as death from any cause. It does not describe how deaths were identified, how incomplete follow-up was handled, or the analysis population used. Those omissions limit assessment of outcome ascertainment and the extent to which the posted comparison includes all randomized participants.

7. Results Posted in the Registry

• Intervention Email: 6.7% (285/4269) versus 5.8% (245/4218) on control; risk difference 0.9 percentage points (95% CI -0.2 to 1.9), risk ratio 1.15 (0.97–1.36); unadjusted two-proportion z-test p = 0.099.

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.

Impact on Evaluation Rate

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Intervention Email: 79.3% (3387/4269) versus 79.1% (3337/4218) on control; risk difference 0.2 percentage points (95% CI -1.5 to 2.0), risk ratio 1.00 (0.98–1.02); unadjusted two-proportion z-test p = 0.797.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Intervention Email and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes broaden the study's stated objectives beyond mortality. The registry names evaluation rate, medical costs and a composite measure of medical treatment. Among these, a binary comparison is posted for evaluation rate; results for medical costs and the composite measure of medical treatment are not reported in the registry record. The record also does not give detailed definitions for these outcomes or their measurement procedures.

The evaluation-rate comparison does not change what the primary endpoint measures. It addresses a different aspect of the intervention strategy, while the primary result concerns death from any cause during 1 year. Secondary outcomes can help describe possible processes or consequences associated with an intervention, but they cannot be treated as substitutes for the primary endpoint. Nor can the unreported secondary results be assumed to support or contradict the mortality comparison.

Unadjusted secondary results generally carry less weight when several outcomes or comparisons may be examined. As a general statistical point, considering more comparisons creates more opportunities for an apparently notable result to arise by chance. That concern is assessed in relation to the prespecified outcome hierarchy, planned analyses and any multiplicity adjustment. Those details are not reported in the registry record, and no reported analyses are listed. The evaluation result should therefore be read cautiously and in the context of incomplete reporting, rather than as independent confirmation of the primary finding.

The overall secondary-outcome pattern is consequently limited: the record identifies a set of objectives, but does not provide a complete set of outcome results or a documented approach to interpreting them together. A full account of secondary findings, their definitions and their planned role in inference is not reported in the registry record.

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

• Impact on Evaluation Rate
• Impact on Medical Costs
• Impact on a Composite Measure of Medical Treatment

9. Statistical Interpretation

10. Statistical Methods Explained

The posted results use counts of participants with an event and group sizes to describe a binary outcome. For the primary endpoint, the event is death from any cause during 1 year. The Intervention Email group has 285 events among 4269 participants, while Control has 245 among 4218. A binary endpoint records whether the defined event occurred for each participant; the group risk is the proportion with that event. The registry gives the risks as 0.067 and 0.058.

What does the risk difference express?

The risk difference subtracts the control risk from the intervention risk. It expresses the contrast on the absolute risk scale. The posted risk difference is 0.009, with an interval from -0.002 to 0.019. A value of zero represents equal risks. Because the reported interval includes zero, it does not rule out no absolute difference. It also includes values below and above zero, so the data remain compatible with effects in either direction over the interval's range. The registry does not identify the confidence level used for these intervals, so no particular confidence percentage should be inferred from this record alone.

How do relative measures differ?

The relative risk compares the intervention risk with the control risk as a ratio. A relative risk of 1 represents equal risks, values above 1 represent a higher risk in the intervention group, and values below 1 represent a lower risk. The posted estimate is 1.149, with an interval from 0.974 to 1.356. The odds ratio compares the odds of the event rather than the risks. Its estimate is 1.16, with an interval from 0.973 to 1.384; an odds ratio of 1 represents equal odds. Both posted relative-measure intervals contain their respective null value. Risk differences and relative measures describe the same group comparison on different scales, so their numerical values are not directly interchangeable.

What does the p-value tell the reader?

The unadjusted p-value for the primary comparison is 0.099. As a general statistical explanation, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions held. It is not the probability that the null hypothesis is true, and it does not measure the probability that chance alone caused the observed result. The word unadjusted indicates that this reported value is not described as adjusted for other comparisons. The registry does not report the trial's own testing plan or the exact test used, so the p-value should be interpreted as the posted unadjusted value rather than as a reconstruction of the sponsor's full analysis.

Why are intervals useful?

A confidence interval gives a range of effect estimates compatible with the data under the procedure used to calculate it. As a general principle, it communicates precision more directly than a point estimate alone. An interval that includes a null value indicates that the data do not exclude that value at the interval's stated level, while its width indicates how much uncertainty remains. Here, the risk-difference interval contains zero, and the relative-effect intervals contain 1. The registry does not report the interval level or calculation method, limiting a fuller assessment of their construction.

What can randomization establish?

Randomization assigns participants to groups by chance and is intended to balance measured and unmeasured characteristics on average, before considering what happens after assignment. This can support a causal comparison of assigned strategies, provided that follow-up, outcome measurement and analysis preserve the intended comparison. The record identifies the allocation as randomized and the model as parallel. It does not describe the random sequence, concealment, adherence, crossovers, outcome ascertainment or analysis population. These details are important for evaluating whether the posted estimate retains the benefits of random assignment.

How should the secondary comparisons be viewed?

The registry names evaluation rate, medical costs and a composite measure of medical treatment as secondary outcomes. It posts a binary comparison for evaluation rate, but results for the other named outcomes are not reported in the registry record. The record lists no reported analyses. The posted risk comparisons are therefore not a substitute for a complete analysis plan or a report of all prespecified outcomes. When multiple outcomes are considered, unadjusted results can be more vulnerable to chance findings across the set; interpreting them requires knowing which comparisons were planned and how multiplicity was handled. That information is not reported in the registry record.

These methods describe associations in the posted group summaries. The record does not provide enough detail to independently reproduce the trial's analysis, assess adjustments, or determine how missing observations were handled. Interpretation should therefore retain the distinction between the displayed estimates and a fully documented analysis of all randomized participants.

11. Confidence Intervals and Precision

Precision concerns how much uncertainty surrounds an estimated effect. The primary risk difference is 0.009, and its reported interval ranges from -0.002 to 0.019. Since the interval crosses zero, the data are compatible with no absolute difference as well as with differences in either direction. The point estimate alone is not a precise description of all effects consistent with the data.

The relative risk is 1.149, with an interval from 0.974 to 1.356. The odds ratio is 1.16, with an interval from 0.973 to 1.384. Each interval crosses 1, the null value for a ratio. These results indicate uncertainty around the relative comparisons as well as the absolute comparison. The unadjusted p-value of 0.099 is another summary of the data under a null model, but it does not show the size or range of effects compatible with the observations.

As a general explanation, narrower intervals indicate greater precision for the same effect scale and method, while wider intervals indicate more uncertainty. Precision depends on factors such as event information and the analysis method. The registry does not report the interval confidence level or calculation method, so the intervals should be quoted as posted without assigning an unstated confidence percentage. The group counts and event counts help describe the data behind the comparison, but do not remove uncertainty in the estimated effect.

The interval should not be treated as a range guaranteed to contain the true effect, nor does its inclusion of a null value prove that the strategies have identical effects. The registry record supports a bounded but unresolved comparison: its point estimates favor neither a conclusive claim of a mortality difference nor a claim of equivalence. More detailed inference is limited by the lack of a reported analysis plan and information about missing data and analysis populations.

12. Planning a Similar Trial: Sample Size

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

The sample-size figures are planning illustrations based on the observed effect, not the trial's original design assumptions. They give participants per arm, rather than total events. The calculation assumptions listed in the registry are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for each illustration is the stated fraction of the observed effect. The trial's own original sample-size assumptions are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a future study to have an 80% chance of detecting the selected effect under the stated assumptions, if that effect is real and the model assumptions hold. The 90% power column uses a higher target probability of detecting that effect, which generally requires a larger planned sample. The rows that use 75% or 50% of the observed effect represent smaller target effects than the row using 100% of the observed effect; smaller effects need more participants to detect reliably.

For the primary outcome, the illustrations are 12,205 participants per arm for 80% power and 16,338 for 90% power when planning for 100% of the observed effect. The corresponding figures rise for the smaller-effect rows. The evaluation-rate illustrations are also available, but their very different values reflect that outcome's observed comparison, not a direct measure of trial quality. The plotted curve is a visual rendering of planning relationships and is not additional trial evidence. These figures are planning illustrations, not observed power.

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 = 0.058; treatment proportion p₁ = 0.067; average p̄ = 0.062.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3421 = 0.6706.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3421 = 0.2879.
4. n = (0.6706 + 0.2879)² / (0.0087)² = 12,204.2, rounded up to 12,205 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 <- 0.0580844001896633                          # control proportion
p1 <- 0.0667603654251581                          # 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(2476578)
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 = 12,205 (80% power), 16,338 (90% power)
 75% of observed effect: n per arm = 21,345 (80% power), 28,575 (90% power)
 50% of observed effect: n per arm = 47,233 (80% power), 63,231 (90% power)
power.prop.test: 12204.2 
Simulated power at n = 12,205 per arm: 0.802 (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. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary outcome and names secondary outcomes, but the analysis strategy for considering them together is not reported in the registry record. In particular, it does not state whether the trial planned a hierarchy of outcomes, a multiplicity adjustment, or a rule for interpreting secondary comparisons. The record lists no reported analyses. The posted unadjusted comparisons should not be mistaken for a documented multiplicity strategy.

As a general statistical issue, evaluating several outcomes or contrasts can increase the chance of obtaining a small p-value somewhere, even if none of the underlying effects is present. Methods such as prespecifying a primary comparison, controlling an overall error rate, or treating secondary findings as exploratory can address different aspects of this problem. Which approach, if any, was planned for LIMIT is not reported in the registry record. This makes the unadjusted secondary comparison less informative as confirmatory evidence.

Information about missing data is also not reported. The record does not explain whether all randomized participants had outcome status available, how loss to follow-up was handled, or whether any imputation or sensitivity analysis was used. The arm counts and event counts are posted, but those figures alone do not establish the completeness of follow-up or the handling of unavailable outcomes.

The analysis population is not specified. The record does not say whether the comparison followed an intention-to-treat approach, included only participants with available outcome data, or used another population definition. It also does not provide a participant flow account or information about protocol deviations. These omissions make it difficult to judge how closely the analyzed groups correspond to the randomized groups and how robust the findings are to missingness or exclusions.

These are reporting limitations, not evidence that a particular problem occurred. The registry record does not establish that data were missing, that multiplicity was mishandled, or that participants were excluded. It simply does not report the details needed to assess those possibilities.

15. 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.

The adverse-event entries list zero serious adverse events and zero other adverse events for both Intervention Email and Control. The at-risk figures are 4269 for Intervention Email and 4218 for Control. The registry also records 285 deaths in Intervention Email and 245 in Control; these are the primary outcome event counts, not a replacement for the adverse-event entries.

The absence of adverse-event reports in these fields should be read as what the registry lists, not as proof that no adverse events occurred. The record does not provide serious adverse event terms, and no serious terms are listed. It does not describe how adverse events were solicited, defined, assessed, or reported, or whether the zero values mean none occurred or none were entered. The email intervention is an alerting strategy, but that alone does not determine what adverse effects might be relevant or how they were monitored.

These entries therefore provide limited safety information. They do not establish comparative safety, and the mortality counts should be interpreted under the primary endpoint rather than as an adverse-event summary. Further methods and reporting details are not reported in the registry record.

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

The posted primary result describes a randomized comparison of an email alert strategy with no email for death from any cause during 1 year. The risk difference estimate is 0.009, and its interval includes zero. The relative risk and odds ratio intervals include 1, and the unadjusted p-value is 0.099. Together, these results show an uncertain estimate rather than a precise distinction between the groups.

The result does not establish that the alert strategy caused higher mortality, and it does not establish that the strategies have identical effects. A failure to exclude a null value is not proof of equivalence. The data remain compatible with a range of effects, as indicated by the posted intervals. The registry does not report an equivalence or non-inferiority design, and the original sample-size rationale is not reported.

The trial compared assignment to receive an email with assignment to no email. It did not directly compare particular diets, medications or clinical revisions. The intervention description advises providers to consider appropriate dietary and medical revision, but the record does not show whether providers received, read or acted on messages, or whether care changed. Accordingly, the posted estimates address the assigned alert strategy, not the effect of a specific clinical treatment.

The outcome is limited to all-cause death over 1 year. The record does not establish effects on longer-term survival, particular causes of death, symptoms, quality of life or other outcomes for which results are not posted. The secondary-outcome record is incomplete, and details of missing data, outcome ascertainment and analysis populations are not reported. Those limitations counsel against extending the primary comparison beyond the defined endpoint and timeframe.

As a general principle, statistical uncertainty and clinical importance are different questions. The registry does not report a threshold for a meaningful mortality difference or a clinical interpretation standard. The point estimate and interval can be used to understand the scale and uncertainty of the comparison, but they do not by themselves determine the practical importance of the alerting strategy.

17. Why This Trial Matters Statistically

This trial illustrates the difference between a point estimate and the uncertainty around it. The primary comparison has an estimated risk difference of 0.009, but the interval ranges from -0.002 to 0.019. Reporting the estimate together with its interval avoids treating the point estimate as a definitive answer. The relative estimates tell a similar statistical story: their intervals include the null ratio value.

It also shows why an unadjusted p-value should not be read in isolation. The posted value is 0.099, while the interval for the absolute comparison includes zero. Neither statistic supplies the missing design and analysis details needed for a complete assessment. The registry record does not report the original sample-size rationale, the testing plan, missing-data handling, or the analysis population.

The intervention is an information strategy rather than a direct treatment assignment. Statistical interpretation must therefore distinguish the effect of assignment to an alert from any effect of provider action or subsequent care, which is not documented in the posted record. This distinction is important when the trial objective mentions treatment and health outcomes but the actual intervention is an email.

Finally, the secondary outcomes demonstrate the importance of complete outcome reporting. Evaluation rate has a posted comparison, while medical costs and the composite measure of medical treatment lack results in the registry record. Without the full outcome set and multiplicity approach, secondary findings cannot be weighed as a coherent confirmatory package. The case underscores that transparent reporting of methods, uncertainty and all planned outcomes is central to judging what a trial can support.

18. Limitations

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

The registry record leaves several trial-specific questions unresolved. Although the study is randomized and parallel, it does not describe sequence generation or allocation concealment. It identifies no masking, and does not say whether outcome ascertainment involved any safeguards against knowledge of group assignment. The email intervention itself could be noticed by providers; how that awareness affected subsequent care is not reported.

The record does not explain how low BMI, HbA1c% or cholesterol values were defined, or how participants meeting the study criteria were identified. It also does not provide baseline characteristics, so the reader cannot assess the composition of the groups from the registry summary. The official title and brief summary concern people older than 75 years, but further eligibility and setting details are not reported in the registry record.

Follow-up and analysis details are also absent. The primary timeframe is 1 year, but the record does not describe outcome ascertainment, censoring, incomplete follow-up, missing data, or the analysis population. Although enrollment is listed as 8584, details explaining how that entry relates to the posted arm counts are not reported. The original sample-size justification and the trial's formal statistical analysis plan are also not reported.

Only limited secondary results are available. The registry names medical costs and a composite treatment measure but does not report their results; it also lists no reported analyses. Adverse-event fields record zero events but provide no reporting methods or serious-event terms. These limitations constrain reproducibility and interpretation. They do not, by themselves, show that the trial was poorly conducted or that unreported events or analyses occurred.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Low Indexes of Metabolism - Information to Teams (LIMIT) — NCT02476578.
• Registry record last updated 2020-04-09; status last verified 2020-03.
• Results first posted 2019-07-29.
• 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.

22. Record Summary

LIMIT (NCT02476578), sponsored by Clalit Health Services, is listed as a completed randomized, parallel-group study with no masking. The record lists enrollment as 8584 and a study period from 2015-10 to primary completion in 2016-11. The intervention was an email alert to primary care providers about low BMI, HbA1c% or cholesterol; Control received no email.

The primary endpoint was death from any cause during 1 year. The registry reports 285 events among 4269 participants assigned to Intervention Email and 245 among 4218 assigned to Control. The posted risk difference is 0.009, with an interval from -0.002 to 0.019; the unadjusted p-value is 0.099. The interval includes zero, leaving uncertainty about the direction and size of the comparison.

The record names evaluation rate, medical costs and a composite measure of medical treatment as secondary outcomes, but only the evaluation-rate comparison is posted. Details of the analysis plan, missing data, analysis populations and original sample-size assumptions are not reported in the registry record.