Effects of a Supplement Derived From Palm Oil on Cholesterol Levels in the Blood: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of a Supplement Derived From Palm Oil on Cholesterol Levels in the Blood (NCT02634606), a randomized phase 3 study. The primary outcome is Change in LDL Cholesterol Over Time Between the 3 Groups.

2. Clinical Question

LDL cholesterol was selected as the principal outcome because the study summary identifies it as the main cholesterol measure of interest. The record describes an investigation in people taking statins, with the aim of assessing the effect of the palm-oil-derived supplement on blood cholesterol. That stated aim is distinct from a demonstrated treatment effect: the trial ended before randomization, and no outcome analysis is posted. Although enrollment is listed as 13, the record does not report participant assignments, group-specific follow-up, or changes in LDL cholesterol. Consequently, the clinical question remains unanswered by the results recorded for this trial.

3. Trial Design

The registry classifies the study as Phase 3, with randomized allocation and double masking, and lists a single-group model. It also lists three interventions: low-dose gamma delta tocotrienols, high-dose gamma delta tocotrienols, and a sugar pill. These design labels and intervention descriptions must be read alongside the explicit statement that randomization was not performed before study termination. The record therefore does not establish that the planned allocation or masking occurred.

The arm descriptions each refer to 33 intended subjects and describe the low-dose preparation as TRF (63mg), the high-dose preparation as TRF (127mg), and the comparator as placebo. Those descriptions are not evidence that participants were actually assigned to those groups. The registry lists enrollment as 13, but does not report the distribution of those participants across groups or describe what study procedures they completed.

The primary outcome timeframe is baseline and 12 weeks, while the summary describes supplement consumption for 3 months. The record lists a start date of 2011-03-22, primary completion on 2015-11-13, and status as terminated. The reason for termination is not reported in the registry record. No further details of recruitment, treatment exposure, visit completion, adherence, or follow-up are 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: Thirteen subjects with cholesterol level (\>180 mg/dl), aged 37-64, taking statins drug at the time were to be assigned to the low dose arm (TRF 63 mg), high dose arm (RTF 127 mg) or control (sugar pill) for 12 weeks. Randomization was not performed due to limited subjects recruited.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Randomization was not performed prior to study termination

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: Enrolled subjects were not assigned to an intervention arm prior to study termination.

6. Primary Endpoint

The registry lists the primary outcome as Change in LDL Cholesterol Over Time Between the 3 Groups, measured over Baseline and 12 weeks.

The primary endpoint is change in LDL cholesterol over time between the three planned groups, with measurements identified at baseline and 12 weeks. LDL cholesterol is the specific outcome emphasized in the study summary and title, which describe an investigation of blood cholesterol in people taking statins. The endpoint is intended to capture whether LDL cholesterol changed differently across the low-dose supplement, high-dose supplement, and sugar-pill groups during follow-up.

Comparing change over time between groups can address a different question from simply describing each group’s measurements separately: it asks whether the pattern of change varies by assigned intervention. The registry names mixed-effects regression models as the planned method for evaluating this endpoint. In general, such models can account for measurements taken repeatedly from the same participant and can estimate differences in outcome patterns across groups. That general rationale does not establish that the model was fitted in this trial.

No LDL cholesterol values, estimated changes, uncertainty intervals, or model results are reported in the registry record. The endpoint is defined, but its observed result and the precision of any comparison cannot be assessed. The stated measurement timeframe also does not establish that participants completed both measurements.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint description names mixed-effects regression models for evaluating change in LDL cholesterol over time between three groups. This is the only specific statistical method identified for the primary outcome. The registry record does not report a fitted model, estimated coefficients, model diagnostics, or an analysis result, and it does not provide further details of the statistical plan.

What a mixed-effects model is intended to do

In general, a mixed-effects model can represent both average patterns across a study population and the correlation among repeated measurements from the same participant. Repeated LDL cholesterol measurements from one person are usually not independent observations: an individual’s measurements may be more alike than measurements from different people. A model that accounts for this structure can estimate average change over time while allowing for participant-level variation. This is a general explanation of the method, not confirmation that any such model was fitted for this trial.

For a comparison involving multiple groups, a model can represent whether the outcome’s time pattern differs across groups. The study’s stated endpoint—change over time between the three groups—calls for that kind of comparison. However, the registry does not report the model’s precise terms, coding of time, handling of baseline measurements, covariance structure, estimation method, or planned contrasts. Those details are not reported in the registry record and should not be inferred from the general name of the model.

What repeated measurements contribute

When measurements are collected at baseline and follow-up, analysis can use information about both the initial level and subsequent outcome. A comparison of changes is intended to distinguish a general change over time from a difference in change associated with the groups. In this trial, the stated time points are baseline and 12 weeks. The record does not say how many participants contributed measurements at either time point, whether measurements were repeated as planned, or how incomplete follow-up would have been handled.

In general, a confidence interval expresses a range of values compatible with an estimate under the model and its assumptions. Its width conveys precision: wider intervals indicate greater uncertainty. A p-value addresses the compatibility of data with a specified null hypothesis under a statistical model; it does not measure the size or clinical importance of an effect. Neither confidence intervals nor p-values for the primary comparison are reported in the registry record. No numerical effect estimate is available to interpret.

Design labels and the analysis

The record labels allocation as randomized and masking as double, but it also states that randomization was not performed prior to termination. It labels the model as single-group while listing three planned interventions. These details do not demonstrate that a randomized three-group analysis was possible or performed. In general, random allocation can support comparisons by reducing systematic differences between groups, while masking can limit some sources of bias in treatment delivery or outcome assessment. Those benefits depend on the procedures actually being implemented; the registry does not establish that they were implemented in this study.

The record’s sample-size assumptions describe a planning illustration using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a fraction of an observed effect. This is a general planning framework, not evidence of the trial’s original design or its own sample-size justification. No observed effect or resulting illustration is reported in the registry record, and the specified two-arm assumptions do not by themselves describe the stated three-group endpoint. The trial’s sample-size justification and any calculation in its original protocol are not reported in the registry record.

More broadly, the registry does not report interim analyses, adjustment for multiple comparisons, model assumptions, or sensitivity analyses. Since no primary analysis is posted, there is no basis for assessing whether model assumptions were checked or whether alternative analyses were conducted. These omissions limit statistical interpretation independently of the fact that randomization was not performed.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint concerns change in LDL cholesterol across three planned groups. The record does not report whether the analysis plan specified a single overall test, particular pairwise comparisons, or a procedure for controlling multiplicity. This matters in general because conducting several statistical tests can increase the chance of a false-positive finding if the testing strategy is not accounted for. No such adjustment or testing strategy is reported in the registry record, and no primary outcome analysis is available to evaluate.

Missing-data methods are also not described. The record does not report how many participants had LDL cholesterol measurements at baseline or 12 weeks, whether any participants withdrew, or how missing measurements would have been handled. Mixed-effects models can use available repeated measurements under particular assumptions about missingness, but naming that model alone does not specify the assumptions or establish an appropriate missing-data analysis. No imputation procedure, sensitivity analysis, or missingness assessment is reported in the registry record.

The analysis populations are not defined in the posted results. In particular, the record does not report an intention-to-treat population, a treated population, or a per-protocol population, and it does not provide participant counts for such groups. Because randomization was not performed before termination, the existence of randomized analysis groups cannot be assumed. The record also does not establish which participants received any study intervention or completed follow-up. These gaps prevent assessment of how participants were selected for analysis and whether the analysis population matched the intended trial question.

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

The registry record establishes that a trial was planned to compare two doses of a palm-oil-derived tocotrienol supplement with a sugar-pill comparator, using change in LDL cholesterol over time as the primary endpoint. It also establishes that the trial was terminated and that randomization was not performed before termination. Those details describe the status and intended question of the study; they do not establish an effect on LDL cholesterol.

No primary outcome results are reported. There are no posted group-specific LDL cholesterol measurements, estimates of change, uncertainty intervals, or statistical tests. Thus, the record cannot show whether LDL cholesterol increased, decreased, or remained similar in any group, nor whether changes differed across groups. Missing results should not be read as a null finding or as evidence of equivalence.

Although enrollment is listed as 13, enrollment alone does not show which interventions, if any, participants received, how many completed follow-up, or how their outcomes compared. The planned arm descriptions and the registry’s design labels do not establish that the intended allocation and masking occurred. Likewise, the zero counts in the adverse-event fields are accompanied by zero participants at risk and cannot be interpreted as a safety finding among treated participants.

In general, a trial’s result must be interpreted in light of its observed data, analysis, and precision. Here, those elements are not reported for the primary comparison. The record therefore documents an intended clinical question and an incomplete trial, rather than evidence for or against a difference between the listed interventions.

12. Why This Trial Matters Statistically

This trial is statistically informative as an example of why planned methods and analyzed outcomes need to be distinguished. The registry names a primary endpoint and a mixed-effects regression approach, but also states that randomization was not performed before termination and reports no primary analysis. A method listed in an outcome description is not the same as a completed analysis, and a planned comparison is not an observed treatment effect.

The record also illustrates how design information should be interpreted when entries do not align cleanly. It labels the model as single-group while listing randomized allocation, double masking, and three interventions. The explicit statement that randomization did not occur is particularly important: it prevents readers from assuming that the listed intervention groups formed a randomized comparison. Planned group descriptions should not be mistaken for actual assignments.

Finally, the record demonstrates that a numerical enrollment total and empty outcome reporting do not provide an estimate of effectiveness or precision. Without outcome measurements and an analysis, there is no basis for confidence intervals, p-values, or interpretation of the magnitude of change. This distinction is useful when reading registry records: statistical methods describe how a question might be answered, while results require data from an analysis that was actually performed.

13. 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 primary limitation is that the registry states randomization was not performed before study termination and provides no primary outcome analysis. The status is listed as terminated, but the reason for termination is not reported in the registry record. The record therefore does not establish whether participants received the planned supplements or comparator, or whether the intended baseline and 12-week measurements were collected.

The registry reports enrollment as 13, while each intervention description refers to 33 subjects. The latter descriptions are not actual group counts, and the record does not reconcile the planned arm descriptions with the listed enrollment. Participant flow, group assignment, exposure, follow-up, and the analysis population are not reported in the registry record.

There are no posted LDL cholesterol values, estimates, confidence intervals, or p-values. The statistical model is named, but its specification and any diagnostics are not reported. The record also does not describe missing-data handling, multiplicity control, interim analyses, or the trial’s sample-size justification. These gaps prevent evaluation of the intended primary comparison and its precision.

The adverse-event entries list zero participants at risk in each group as well as zero events, so they do not establish the safety experience of participants receiving an intervention. In addition, the record’s single-group model label sits alongside three listed interventions and randomized allocation. The record does not resolve this design discrepancy beyond stating that randomization was not performed.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Effects of a Supplement Derived From Palm Oil on Cholesterol Levels in the Blood — NCT02634606.
• Registry record last updated 2019-02-12; status last verified 2019-01.
• Results first posted 2019-02-12.
• 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.

17. Record Summary

The registry record concerns a Phase 3 study of palm-oil-derived gamma delta tocotrienols for people with hypercholesterolemia. Its intended primary endpoint was change in LDL cholesterol from baseline to 12 weeks across low-dose, high-dose, and sugar-pill groups, evaluated with mixed-effects regression models. The listed doses were 63mg and 127mg.

The study is listed as terminated, with enrollment of 13. The record states that randomization was not performed prior to termination. Although the design fields list randomized allocation and double masking, the record does not establish that these procedures occurred. No primary outcome analysis or LDL cholesterol results are reported in the registry record. The reason for termination, participant assignments, follow-up, analysis populations, and missing-data procedures are also not reported. The adverse-event fields list zero participants at risk and zero events in each group, which does not establish safety outcomes among treated participants.