Effect of Lactose on Blood Lipids: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Lactose on Blood Lipids (NCT04924530), a randomized na study. The primary outcome is Postprandial Changes in Plasma Triglyceride Concentrations.

2. Clinical Question

3. Trial Design

The registry identifies this as a randomized, triple-masked, crossover study. The interventions were maltodextrin, sucrose, and lactose. A crossover design means that the comparison is made across conditions in which participants receive different interventions, rather than assigning each participant permanently to only one condition. Randomization is intended to allocate treatment order according to a chance-based procedure. The precise sequence allocation, masking procedures, and practical conduct of the trial are not reported in the registry record.

The record lists an enrollment of 26 and a completed status. The recorded start date is 2021-06-07, and the primary completion date is 2022-07-13. The primary outcome summaries list 24 for each intervention. These figures describe the enrollment record and the posted outcome groups; the registry record does not explain why they differ or provide a detailed participant flow.

The study concerns cardiovascular risk factor, lipaemia, and metabolic disease. Its brief summary states that it aims to determine the extent to which ingestion of free sugars influences postprandial triglyceridaemia in men and women. No further demographic profile, eligibility criteria, setting, or recruitment details are reported in the registry record.

The primary outcome is postprandial changes in plasma triglyceride concentrations, assessed as incremental area under the curve over 6 hours. Secondary outcomes include de novo lipogenesis, dietary fat oxidation, insulin, glucose, galactose, fructose, and triglyceride concentrations in VLDL-rich and chylomicron-rich fractions. The registry record lists 12 secondary outcomes as posted. The reported information does not establish which secondary outcomes were analyzed or how they were handled statistically.

The record identifies the University of Bath as sponsor. The study phase is listed as NA. Statistical details such as the original sample-size rationale, period and sequence modeling, handling of carryover, and the prespecified 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.

Before assignment: 2 dropouts

Period 1

Period 2

Period 3

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Crossover design so n = 24 for all arms

6. Primary Endpoint

The registry lists the primary outcome as Postprandial Changes in Plasma Triglyceride Concentrations, measured over 6 hours.

The primary endpoint is the incremental area under the curve (iAUC) of plasma triglyceride concentrations from timepoints 0, 30, 60, 120, 180, 240, 300, and 360 minutes. It covers 6 hours and is expressed as mmol/L x 360 min. The reported quantity is a mean of a continuous measure, with standard deviation as the measure of spread.

An area-under-the-curve measure combines concentration measurements across a period into a single summary. The incremental form focuses on change relative to a reference level rather than simply summing all measured concentrations. It can therefore represent the overall postprandial response across the observation period, rather than the response at one selected timepoint. A larger iAUC indicates a greater accumulated concentration response as represented by this measure; interpretation still depends on how the outcome was calculated and analyzed.

Using a summary over time can be useful when the research question concerns the overall post-meal response and the outcome changes throughout the observation period. It also reduces a series of time-specific measurements to one value for each intervention condition. That makes a mean comparison possible, but it does not show when any difference arose, whether it persisted, or how the concentration pattern varied across timepoints. The registry record names the timepoints and endpoint but does not report the detailed calculation procedure beyond identifying the outcome as iAUC.

7. Results Posted in the Registry

• Sucrose: mean 90.00 versus 51.00 on control; difference 39.00 (95% CI -7.74 to 85.74), standardized effect d = 0.47; z-test p = 0.102.
• Lactose: mean 98.00 versus 51.00 on control; difference 47.00 (95% CI 2.51 to 91.49), standardized effect d = 0.60; z-test p = 0.038.

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help interpret the posted summaries. The registry record reports group means, standard deviations, mean differences, confidence intervals, and unadjusted p-values. It does not report a full statistical analysis plan or identify the model used to produce the comparisons. The following explanations describe the general meaning of these statistics; they do not establish which specific methods were used in the trial.

What does a mean summarize?

A mean is a measure of the central value of a set of observations. For this continuous endpoint, it gives a concise summary of the iAUC values in each intervention group. A mean can be useful for comparing average responses, but it does not show the shape of the outcome distribution or how responses vary between individuals. The reported standard deviation describes the spread of the observations around the mean. It is not the same as the uncertainty in the mean or the uncertainty in a difference between means.

The posted means are 51.0 for maltodextrin, 90.0 for sucrose, and 98.0 for lactose. The associated standard deviations are 68.0, 95.0, and 88.0, respectively. These summaries show that the observed values are variable relative to their group means. In general, greater variability makes an estimated mean difference less precise when other features of the data are unchanged. For a crossover study, however, the variation in within-participant differences may be especially important, and it cannot be recovered from separate group means and standard deviations alone.

What does a mean difference represent?

A mean difference expresses the estimated contrast between the mean outcome in an intervention condition and the mean in its comparator. The posted mean differences are 39.0 for sucrose compared with maltodextrin and 47.0 for lactose compared with maltodextrin. Positive values indicate that the reported intervention mean is greater than the maltodextrin mean for these contrasts. The units remain mmol/L x 360 min because the contrast is in the same units as the endpoint.

The registry record identifies the design as crossover, in which observations from different conditions may be related because the same participants take part across conditions. In general, a statistical analysis of crossover data can use that within-participant relationship. It may also need to account for sequence or period effects, depending on the design and analysis plan. The registry record does not report whether these features were modeled, so the posted mean differences cannot be described as paired, adjusted, or based on any particular crossover model.

How should a confidence interval be read?

A confidence interval describes the uncertainty around an estimate under the statistical procedure used to construct it. In general, a 95% confidence interval is produced by a method that, over repeated applications under its assumptions, would cover the target quantity at the stated rate. It is not a guarantee that a particular interval contains the true value, nor does it assign a probability to the true value after the data have been observed.

For sucrose compared with maltodextrin, the interval is -7.74 to 85.74. It includes zero, the value corresponding to no mean difference. For lactose compared with maltodextrin, the interval is 2.507 to 91.49 and does not include zero. Both intervals are broad relative to their point estimates, so a range of difference sizes remains compatible with the posted data. The interval’s validity depends on the method and assumptions used, which are not described in the registry record.

What does a p-value tell the reader?

A p-value measures how unusual data at least as incompatible with a specified null model would be if that model and the test assumptions were true. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. The unadjusted p-values are 0.102 for sucrose and 0.038 for lactose. They should be interpreted with the effect estimates and confidence intervals, not in isolation. The registry record does not report a multiplicity adjustment, so these p-values should not be described as adjusted for the set of comparisons.

General statistical practice also distinguishes a prespecified analysis from an exploratory one. Prespecification can reduce the risk that results are selected or interpreted differently after they are known. The registry record does not describe the analysis plan, the timing of its finalization, or any interim analyses. Those aspects cannot be inferred from the posted results.

Why can over-dispersion matter?

Over-dispersion occurs when observed variation is greater than a statistical model expects. In general, a model that understates variation may produce confidence intervals that are too narrow and p-values that overstate the evidence against a null hypothesis. The posted endpoint is continuous, and the registry record does not state that an over-dispersion model was relevant or used. The broader lesson is to check whether the statistical model reflects the observed data and design, rather than to assume a particular correction was needed in this trial.

The primary outcome summaries and comparisons are useful for understanding the direction, magnitude, and uncertainty of the posted results. They do not provide the participant-level data or analytical details needed to verify the calculation independently or assess every modeling assumption.

10. Confidence Intervals and Precision

Precision concerns how narrowly the data estimate a quantity, rather than whether an estimate is important. The reported confidence intervals provide a range of values compatible with the data under the method’s assumptions. The sucrose comparison has a mean difference of 39.0 and a 95% confidence interval from -7.74 to 85.74. Because the interval crosses zero, the data do not distinguish the estimate from no difference at the level implied by that interval. It also permits positive differences of appreciable size, so the result is not evidence that the conditions are identical.

The lactose comparison has a mean difference of 47.0 and a 95% confidence interval from 2.507 to 91.49. This interval is above zero but remains wide, indicating uncertainty about the size of the difference. A confidence interval is more informative than a point estimate alone because it shows how much uncertainty accompanies the estimated contrast. It does not show the distribution of individual responses or the range of likely outcomes for a future participant.

The group standard deviations are 68.0 for maltodextrin, 95.0 for sucrose, and 88.0 for lactose, while the outcome summaries list 24 observations for each intervention. In general, precision depends on sample size, variability, and the analysis design. In a crossover study, correlation between measurements from the same participant can also affect precision. The registry record does not report the within-participant correlation or the method used to account for it, so the group-level summaries are not enough to assess that contribution.

The intervals should also be considered in the context of two posted comparisons and the absence of reported multiplicity procedures. If several comparisons are assessed, the interpretation of nominal confidence intervals and p-values may differ from an analysis that controls the overall error rate. No such adjustment is reported in the registry record. Thus the intervals communicate uncertainty for the individual posted comparisons, not necessarily uncertainty accounting for all outcomes and comparisons in the study.

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 displays are planning illustrations based on the observed effects, not the trial’s original design calculations. The assumptions stated for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers detecting the stated fraction of the observed effect. The table and curve therefore describe hypothetical future planning scenarios rather than the number of participants who took part in this trial.

The displayed figures are participants per arm, not total participants across the arms and not total events. For each comparison, the columns for 80% and 90% power indicate the number of participants per arm in the illustration for a design with the corresponding probability of detecting the specified effect under the stated assumptions. Higher planned power generally requires more participants when the other assumptions are held constant. The record’s own original sample-size assumptions and justification are not reported in the registry record.

The rows represent planning for 100%, 75%, and 50% of the observed effect. Smaller effects are harder to distinguish from random variation, so they require more participants for the same target power, all else being equal. The values depend on the observed effect and the assumptions used in the calculation; they should not be read as universal requirements for a study of this topic. The calculation is not observed power. Observed power is not a substitute for interpreting the estimated effect and its confidence interval.

These planning figures also do not demonstrate that the posted trial was adequately or inadequately sized. The record identifies a crossover design, while the illustration assumptions describe equal allocation between two arms and a normal approximation. The registry record does not explain whether the illustration reflects features of the original crossover design. A future protocol would need a planning method aligned with its intended design, analysis, outcome variability, and target effect.

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 = 39.000; pooled standard deviation σ = 82.611.
2. n = 2 × (1.96 + 0.842)² × 82.611² / 39.000² = 70.4, rounded up to 71 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 <- 39                      # observed difference in means
sigma <- 82.6105320162024                     # 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(4924530)
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 = 71 (80% power), 95 (90% power)
 75% of observed effect: n per arm = 126 (80% power), 168 (90% power)
 50% of observed effect: n per arm = 282 (80% power), 378 (90% power)
power.t.test: 71.4 
Simulated power at n = 71 per arm: 0.804 (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

The primary outcome comparisons reported are sucrose versus maltodextrin and lactose versus maltodextrin. The record also lists 12 secondary outcomes as posted, covering metabolic measures and triglyceride fractions. It does not report results for those secondary outcomes in the information available in the registry record. The registry lists 13 outcome measures and zero analysis records, so the detail needed to understand the full set of planned or completed analyses is limited.

Multiplicity arises when several statistical questions are examined. In general, the more comparisons or outcomes assessed, the greater the chance that at least one result appears notable by chance if each is interpreted on its own. Methods to address this can adjust the significance threshold or use other procedures, but the appropriate approach depends on the analysis plan and goals. Whether any multiplicity adjustment was planned or applied is not reported in the registry record. The posted p-values are explicitly unadjusted and should be interpreted as individual comparison results.

The registry record does not report how missing outcome data were handled, whether any participants withdrew from particular conditions, or whether all randomized participants were included in each analysis. It gives an enrollment of 26 and primary outcome summaries of 24 for each intervention, but does not provide a participant flow or explain the relationship between those figures. No imputation method or missing-data sensitivity analysis is reported.

The record also does not identify the analysis population, such as whether the summaries represent all randomized participants or a subset with complete measurements. That distinction matters because exclusions and missing data can affect how representative an estimate is of the randomized group. The record does not report a prespecified protocol analysis or the handling of protocol deviations.

Because this is a crossover design, the analysis population and treatment of within-participant observations are especially relevant. The registry record does not describe how sequence, period, or carryover were handled. These omissions do not establish that the analysis was incorrect; they limit what can be concluded from the registry summaries alone.

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.

The registry reports serious adverse events for maltodextrin, sucrose, and lactose. For each intervention, 24 were listed as at risk and 0 as affected by a serious adverse event. The record also reports 24 at risk and 0 affected by other adverse events for each intervention. Deaths are recorded as 0 for each intervention.

These figures describe the adverse-event entries in the registry record. A count of 0 means no events were recorded in that category for the listed group; it does not prove that an event could never occur or rule out events that were not captured or reported. The record does not provide event descriptions, severity details beyond the serious/non-serious categories, or information about the method and duration of safety monitoring.

The study is recorded as a crossover trial, but the registry record does not explain how adverse events were assigned when participants received more than one intervention. The group figures should therefore be read as posted, without assuming a particular attribution or comparison method. These safety entries do not replace the outcome estimates for postprandial triglyceride concentrations, and they do not establish a general safety profile beyond the trial record.

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

The posted results estimate differences in the mean postprandial triglyceride iAUC between each sugar condition and maltodextrin. The sucrose comparison has a positive estimate, but its confidence interval crosses zero. The lactose comparison also has a positive estimate, and its confidence interval lies above zero. Both intervals are broad, leaving uncertainty about the size of the corresponding differences. These are statistical descriptions of the posted endpoint, not conclusions about clinical benefit or harm.

The study’s stated aim concerns the extent to which ingestion of free sugars influences postprandial triglyceridaemia in men and women. The primary endpoint captures an accumulated response over 6 hours. It does not establish what happens outside that observation period, whether the outcome predicts a clinical event, or whether any difference persists over time. The registry record does not report clinical outcomes or longer-term follow-up results.

The trial lists secondary metabolic outcomes, including de novo lipogenesis, dietary fat oxidation, plasma insulin, glucose, galactose, fructose, and triglyceride measures in lipoprotein fractions. Their results are not reported in the registry record. It is therefore not possible to use these entries to explain a mechanism or to evaluate consistency across the listed metabolic measures.

The crossover design may support comparisons across conditions within participants, but the registry record does not describe the analysis model or the handling of within-participant dependence, period, sequence, or carryover. The posted group means and standard deviations do not provide those details. The result should be understood as the comparison reported in the registry, with its stated uncertainty, rather than as a complete account of the trial’s analysis.

Nothing in these data alone establishes whether a particular food or dietary choice is appropriate for an individual. The record addresses a specific experimental outcome under the study conditions, while the intervention dose and other session details are not reported. Interpretation should remain limited to the recorded endpoint and comparison.

16. Why This Trial Matters Statistically

This trial illustrates why an estimated difference should be considered together with its uncertainty. The point estimates for both sugar comparisons are positive, but the precision differs: the sucrose interval crosses zero, while the lactose interval does not. Neither estimate should be reduced to a yes-or-no conclusion based only on a p-value. The size and range of plausible values matter for understanding what the data can support.

The study also highlights the importance of matching statistical interpretation to design. A crossover trial can make use of comparisons within the same participant, but that structure requires analysis choices that reflect the relationship between repeated conditions. The registry record identifies randomization, triple masking, and a crossover model, yet does not describe how the posted comparisons account for that structure. This limits independent assessment of the analysis and its precision.

Outcome selection is another statistical issue. The primary endpoint condenses repeated concentrations across 6 hours into an iAUC, which provides a single summary of the postprandial response. That makes a primary comparison more focused, but it cannot convey the full time course. The listed secondary outcomes broaden the scientific questions, while also raising the need to understand which analyses were prespecified and how multiple comparisons were handled. Those details are not reported in the registry record.

Finally, planning estimates based on observed effects are not a substitute for prospective design assumptions. The sample-size illustrations show how the required participants per arm changes with the target effect and desired power under stated assumptions. They do not reveal the original rationale for enrollment or demonstrate the trial’s observed power. Keeping these distinctions clear helps readers separate the data’s uncertainty from planning scenarios and from claims about clinical meaning.

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 record leaves several design and analysis details unavailable. It does not report the intervention doses, administration procedures, or the interval and conditions between crossover periods. The sequence allocation and methods used to maintain triple masking are also not described. These details can affect how readers assess the conduct and interpretation of a crossover comparison.

The record does not report the eligible population in detail, participant characteristics, or the setting. Although the brief summary refers to men and women, no further demographic information is reported. The primary outcome summaries list 24 for each intervention, while enrollment is recorded as 26; a participant flow explaining these figures is not reported in the registry record.

Important analytical information is also absent. The registry record does not identify the model used for the comparisons, whether the within-participant structure was incorporated, how period or sequence effects were handled, or whether carryover was assessed. It does not report a statistical analysis plan, analysis population, missing-data method, or multiplicity procedure. The posted unadjusted p-values therefore cannot be assumed to account for all comparisons or design features.

The primary endpoint is a summary across specified timepoints, so it does not show the detailed time pattern. Secondary outcomes are listed, but their results are not reported in the registry record. Safety entries report no affected participants in the listed categories, but event descriptions and monitoring details are not included. Finally, the registry record does not report the original sample-size assumptions or justification. These limitations restrict interpretation of the posted summaries without establishing that any particular design or analysis decision was inappropriate.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Effect of Lactose on Blood Lipids — NCT04924530.
• Registry record last updated 2024-09-23; status last verified 2024-05.
• Results first posted 2024-09-23.
• 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 record describes a completed, randomized, triple-masked crossover study sponsored by the University of Bath. It lists an enrollment of 26 and compares maltodextrin, sucrose, and lactose. The primary endpoint is the mean incremental area under the curve for plasma triglyceride concentrations over 6 hours, in mmol/L x 360 min.

The posted mean differences versus maltodextrin are 39.0 for sucrose and 47.0 for lactose. Their respective 95% confidence intervals are -7.74 to 85.74 and 2.507 to 91.49; the unadjusted p-values are 0.102 and 0.038. The registry record does not report enough statistical detail to establish how the analysis accounted for the crossover design, missing data, or multiple comparisons. The results describe the recorded endpoint and do not establish longer-term or clinical effects.