Benfotiamine in Alzheimer's Disease: A Pilot Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Benfotiamine in Alzheimer's Disease: A Pilot Study (NCT02292238), a randomized phase 2 study. The primary outcome is Change From Baseline in ADAS-Cog Score.

2. Clinical Question

3. Trial Design

The registry identifies NCT02292238 as a Phase 2, randomized, quadruple-masked, parallel-group pilot study sponsored by Burke Medical Research Institute. The listed condition is Alzheimer's disease. Enrollment is reported as 71, and the record status is completed. The recorded study start is 2015-02-15 and the primary completion date is 2020-07-20.

The study plan describes people with amnestic mild cognitive impairment or mild Alzheimer's disease dementia. Its central aim was to investigate whether increasing brain thiamine availability with benfotiamine could slow cognitive decline and decline in brain glucose utilization. The planned regimen was 600 mg per day, given as 300 mg in the morning and 300 mg in the evening, for one year. Placebo was the comparator.

The primary outcome was change from baseline in ADAS-Cog score, with the timeframe stated as baseline to one year. Secondary outcomes included brain glucose utilization and measures of daily living, neuropsychiatric symptoms, clinical dementia rating, and selective reminding. The registry posts results for three secondary outcomes in addition to the primary result; other planned or listed outcomes do not have results in the posted record.

For the primary outcome, the registry reports 34 analyzed participants in the benfotiamine group and 36 in the placebo group. It states that one placebo-arm subject was excluded from analysis because that person took benfotiamine from a commercial vendor. These analysis counts should not be confused with the enrollment figure. The registry record does not report the original sample-size justification, detailed allocation procedures, a full protocol, or the complete statistical analysis plan.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 1 subject on the placebo arm was excluded from analysis due to taking benfotiamine from a commercial vendor.

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: 1 subject on the placebo arm was excluded from analysis due to taking benfotiamine from a commercial vendor.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in ADAS-Cog Score, measured over Baseline, 1 year.

The primary endpoint is change from baseline in the Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) score, assessed at baseline and one year. The registry describes ADAS-Cog as a brief neuropsychological assessment used to assess the severity of cognitive symptoms of dementia and notes that it is widely used in clinical trials. Its scale ranges from 0 to 70, with higher scores indicating greater cognitive dysfunction.

Change scores express how much the measured score shifted between the stated assessments. For this endpoint, the group means are reported as 1.39 for benfotiamine and 3.26 for placebo. The comparison estimate is the mean difference for benfotiamine relative to placebo, reported as -1.87 in the recalculation. Since higher ADAS-Cog scores indicate greater dysfunction, a positive change represents movement toward a higher score, while the negative contrast indicates a lower mean change in the benfotiamine group than in the placebo group. This describes the observed sample contrast, not proof that the treatment caused it.

General: a cognitive scale can provide a standardized way to compare measured change between trial groups. The scale's use in research does not make a particular observed difference clinically important automatically. The registry record does not report a threshold for clinical importance, and the posted result must be read alongside its uncertainty and the limits of the analysis.

7. Results Posted in the Registry

• Benfotiamine: mean 1.39 versus 3.26 on control; difference -1.87 (95% CI -4.48 to 0.74), standardized effect d = -0.34; z-test p = 0.161.

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.

Change From Baseline in Brain Glucose Utilization

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Benfotiamine: mean -0.02 versus -0.01 on control; difference -0.01 (95% CI -0.05 to 0.03), standardized effect d = -0.12; z-test p = 0.619.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Benfotiamine and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Alzheimer's Disease Cooperative Study-Activities of Daily Living (ADCS-ADL) Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Benfotiamine: mean -1.93 versus -3.16 on control; difference 1.23 (95% CI 0.79 to 1.67), standardized effect d = 1.30; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Benfotiamine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change From Baseline in Neuropsychiatric Inventory (NPI) Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Benfotiamine: mean 6.69 versus 9.23 on control; difference -2.54 (95% CI -4.00 to -1.08), standardized effect d = -0.80; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Benfotiamine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The posted secondary results cover brain glucose utilization, ADCS-ADL, and NPI. Their recalculated contrasts do not form a single consistent pattern across all measures, and the scales capture different aspects of cognition, daily functioning, and neuropsychiatric symptoms. They should not be treated as interchangeable confirmations of the primary endpoint. The primary outcome comparison remains the prespecified central result in the registry's stated study aims; secondary outcomes provide additional, exploratory context rather than a replacement for that primary comparison.

Some recalculated secondary comparisons have unadjusted p-values below 0.001, while the brain glucose utilization comparison has an unadjusted p-value of 0.619. These are unadjusted results and cannot establish a coherent treatment effect across outcomes. The registry also reports sponsor analyses for the posted secondary outcomes whose p-values and contrast values do not align with the recalculated summaries for ADCS-ADL and NPI. The record does not explain those differences, and the analyses should not be blended or presented as if they were the same calculation.

General: testing multiple outcomes creates more opportunities for a nominally small p-value to occur, even when there is no real effect. Adjusting for multiplicity or specifying an outcome hierarchy can help control how strongly secondary findings are interpreted. The registry record does not report a multiplicity adjustment or a hierarchy for these results. For that reason, unadjusted secondary findings carry less weight than a clearly prespecified primary comparison and should be viewed as hypothesis-generating rather than definitive.

The study plan also names other secondary measures, but results for all listed measures are not posted in the record. The available set therefore does not provide a complete account of every planned secondary objective. Taken together, the secondary results add context but do not resolve the uncertainty in the primary outcome.

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

• Change From Baseline in Brain Glucose Utilization
• Change From Baseline in Alzheimer's Disease Cooperative Study-Activities of Daily Living (ADCS-ADL) Score
• Change From Baseline in Neuropsychiatric Inventory (NPI) Score
• Change From Baseline in Clinical Dementia Rating (CDR) Score
• Change From Baseline in Buschke Selective Reminding Test (SRT) Score

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary comparison concerns a continuous outcome: change in ADAS-Cog score. The registry gives a mean and standard deviation for each group, then a mean difference comparing benfotiamine with placebo. The recalculated contrast is unadjusted, meaning it is based on the posted group summaries rather than a reported covariate-adjusted model.

What does a mean difference represent?

General: a mean difference subtracts one group's average outcome from the other group's average outcome. Its sign depends on the order of subtraction. In the recalculation, the estimate is expressed as benfotiamine minus placebo, so a negative estimate means the observed group mean change was lower with benfotiamine. The sponsor's reported net mean difference is presented with the opposite sign from that recalculation; the registry does not explain its sign convention.

What does the confidence interval add?

General: a confidence interval describes the range of values compatible with the data and the statistical procedure, under its assumptions. It communicates uncertainty around an estimate rather than providing a guarantee that the true value lies in the interval. A wider interval indicates less precision than a narrower one. For the primary result, the reported interval extends from -4.484 to 0.744 and includes zero, reflecting uncertainty about both the size and direction of the mean difference.

How should the p-value be read?

General: a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be, if that null hypothesis and the test assumptions were true. It is not the probability that the null hypothesis is true, and it does not measure the probability that a result occurred by chance. The recalculated unadjusted p-value is 0.161. That does not meet the conventional 0.05 threshold used in the posted planning illustration, but a threshold alone does not resolve whether an effect is important or absent.

The sponsor-reported primary p-value is 0.125, from a two-sided t-test. The recalculated p-value is 0.161. The registry does not provide enough information to determine why the values differ, including whether the sponsor used a different analysis population, variance assumption, or calculation. Those possibilities should not be presumed. The reported result and recalculation should therefore remain separately identified.

Why does variability matter?

General: standard deviations describe variation among individual outcomes around a group mean. Greater variation makes it harder to estimate a group average precisely when the other features of a study are unchanged. The primary endpoint's posted standard deviations are 5.63 for benfotiamine and 5.52 for placebo. Those values describe spread in the observed changes; they are not standard errors of the mean difference. The confidence interval is the more direct summary of uncertainty in the group contrast.

The registry reports a randomized, quadruple-masked, parallel-group design. Randomization is intended to make treatment groups comparable on average, while masking is intended to reduce certain biases in trial conduct and assessment. These design labels do not guarantee that all sources of bias or uncertainty are eliminated. Details of random sequence generation, allocation concealment, masking procedures, and protocol adherence are not reported in the registry record.

The record notes one exclusion from the primary analysis: a placebo-arm subject took benfotiamine from a commercial vendor. It does not describe an intention-to-treat analysis, a per-protocol analysis, imputation for missing outcomes, or sensitivity analyses. Consequently, the target population and the effect of exclusions or missing measurements cannot be fully assessed from the posted information.

For secondary outcomes, the registry posts group means and comparison estimates for brain glucose utilization, ADCS-ADL, and NPI. The measures use different units and variability summaries, so their raw values should not be compared directly across outcomes. The posted recalculated contrasts are also unadjusted. General: an unadjusted comparison does not account for testing multiple outcomes or multiple hypotheses. When several comparisons are examined, the chance of at least one apparently strong result can increase. The registry record does not report a multiplicity strategy or identify a hierarchy governing interpretation of secondary outcomes.

The summary-statistic recalculations are not a substitute for a complete trial analysis. The registry does not report adjusted analyses, subgroup analyses, interim analyses, or the original sample-size justification. No specific model beyond the sponsor-reported two-sided t-tests is available for the results described. Interpretation should therefore remain tied to the reported group summaries and their stated uncertainty.

11. Confidence Intervals and Precision

The primary mean difference is -1.87 ADAS-Cog points for benfotiamine relative to placebo. Its 95% confidence interval runs from -4.484 to 0.744. Because the interval crosses zero, the data are compatible with a negative mean difference, no difference, or a positive mean difference under the calculation used. The estimate's negative direction is not enough to establish a treatment effect.

General: precision is about how tightly an estimate is located. A confidence interval communicates this uncertainty in the same units as the outcome, which can make the range more interpretable than a p-value by itself. Here, the interval spans values on both sides of zero, so the direction is uncertain. It also includes values whose magnitudes differ from the point estimate, indicating that the estimate is not highly precise.

The trial's analysis groups are relatively small, and the observed standard deviations are 5.63 and 5.52. The registry does not report a confidence interval for each group mean, but it does report the interval for the between-group contrast. The contrast interval is the relevant uncertainty summary for comparing the groups; the separate group means alone do not establish whether the groups differ.

The sponsor-reported primary analysis gives p = 0.125, while the recalculation gives p = 0.161. The registry-reported analysis does not include a confidence interval. Since the registry does not explain why the p-values differ, neither should be substituted for the other. Their difference does not change the key reading of the recalculated interval: it includes zero, and the evidence does not precisely establish the direction or magnitude of the contrast.

These intervals and tests address sampling uncertainty under their respective methods. They do not cover every source of uncertainty, such as missing data, outcome measurement, adherence, or limitations in the study design. Those matters are incompletely described in the registry record.

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 calculation. They give participants per arm for the continuous outcomes; they are not total event counts. The assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Each planning row uses the stated fraction of the observed effect as the effect to detect. The registry record does not report the trial's own original design assumptions.

The 80% power column describes the illustrative number of participants per arm for a future study designed to have an 80% chance of detecting the chosen effect under the stated assumptions, if that effect and assumptions are correct. The 90% power column gives the corresponding planning figure for a 90% chance. Power is a design property under specified assumptions; it is not a measure of whether the current trial's result is true.

The rows that plan for 100%, 75%, and 50% of the observed effect show how the assumed effect size changes the planning requirement. A smaller target effect is harder to distinguish from random variation, so more participants are needed to attain the same planned power. For the primary endpoint, the posted illustration lists 140 participants per arm for 80% power at 100% of the observed effect, and 558 participants per arm for 80% power at 50% of the observed effect. The larger figures in the 90% column reflect the higher target probability of detecting the assumed effect.

These figures are not observed power. They are conditional planning examples based on the observed effect and stated assumptions, not a retrospective measure of what the trial achieved. Because the observed effect itself is uncertain, the resulting planning numbers are also illustrative rather than a definitive prescription for a future study.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -1.870; pooled standard deviation σ = 5.574.
2. n = 2 × (1.96 + 0.842)² × 5.574² / -1.870² = 139.5, rounded up to 140 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 <- -1.87                      # observed difference in means
sigma <- 5.5736534907809                     # 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(2292238)
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 = 140 (80% power), 187 (90% power)
 75% of observed effect: n per arm = 248 (80% power), 332 (90% power)
 50% of observed effect: n per arm = 558 (80% power), 747 (90% power)
power.t.test: 140.4 
Simulated power at n = 140 per arm: 0.803 (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.

14. Analyses Reported by the Sponsor

For the primary outcome, the registry reports the sponsor's analysis as a two-sided t-test with a net mean difference of 1.869 and p = 0.125. It does not provide a confidence interval for that analysis. The recalculation from the posted group summaries expresses the contrast as benfotiamine minus placebo: -1.87, with a 95% confidence interval from -4.484 to 0.744 and p = 0.161.

The estimates have similar magnitudes but opposite displayed signs, consistent with a possible difference in subtraction order; the registry does not state the sign convention. The p-values also differ. The record does not explain the difference, so it is not possible to identify its cause from the posted information. The recalculation is an unadjusted calculation from the posted summary statistics and should not be represented as the sponsor's analysis.

The registry does not flag a discrepancy for this comparison. No sponsor-reported confidence interval is available to compare with the recalculated interval. The distinction matters: the reported analysis describes what the registry attributes to the sponsor, while the recalculation is a separate summary-statistic analysis. Neither should be silently substituted for the other, and the registry record does not provide enough detail to reconcile them.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple secondary outcomes in addition to the primary ADAS-Cog result. It reports results for brain glucose utilization, ADCS-ADL, and NPI, while the record also names CDR and Buschke Selective Reminding Test outcomes. Results for all listed secondary measures are not posted.

General: multiplicity arises when researchers evaluate more than one outcome, comparison, time point, or subgroup. The more tests that are considered, the greater the opportunity for a result to appear statistically notable by chance. A prespecified hierarchy or statistical adjustment can help clarify which findings should be treated as confirmatory and how to interpret the others. No multiplicity adjustment or testing hierarchy is reported in the registry record. The posted secondary p-values are therefore unadjusted and should not be read as if they were protected against multiple testing.

The registry does not report how missing outcome data were handled, whether all randomized participants were included in each analysis, or whether imputation or sensitivity analyses were performed. It notes that one placebo-arm subject was excluded from the primary analysis because of commercial benfotiamine use. The effect of this exclusion, or of any other missing measurements, is not reported.

The analysis population is not fully described beyond the reported group counts and that specific exclusion. The registry does not state whether the primary analysis followed an intention-to-treat principle, whether additional participants were excluded, or how deviations from treatment were handled. Without these details, the influence of missingness, exclusions, and adherence on the estimates cannot be assessed.

The registry record also does not report interim analyses, a complete statistical analysis plan, or the original sample-size justification. These absences limit the ability to determine how outcome selection, missing data, and multiple comparisons were handled in the trial's full analysis. They should be treated as reporting gaps rather than evidence that a particular procedure was or was not used.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: surgery (8.3%), stroke (5.9%).

The registry reports serious adverse events affecting 3 participants in each group. The corresponding groups at risk are reported as 34 for benfotiamine and 36 for placebo. The listed top serious adverse event terms are Surgery at 8.3% and Stroke at 5.9%. These terms describe the most frequently listed serious events in the record; the registry excerpt does not link each term to a specific treatment group.

For other adverse events, the registry reports 25 affected participants among 34 at risk in the benfotiamine group and 23 among 36 at risk in the placebo group. Deaths are reported as 0 in each group. These counts describe the posted safety record and do not, by themselves, show whether a treatment caused an event or establish a comparative safety profile.

The registry record does not provide event-level details, timing, severity beyond the serious-event classification, relatedness assessments, or statistical comparisons for these figures. The counts should be read as descriptive rather than as evidence that one treatment has more or fewer harms. The study's size and the information posted do not support a comprehensive assessment of uncommon or long-term adverse events.

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

The primary result describes the average one-year change in ADAS-Cog score in the two analyzed groups and the estimated difference between those averages. The benfotiamine group had a mean change of 1.39 and the placebo group had a mean change of 3.26. The recalculated contrast is -1.87, with a confidence interval that includes zero. The data therefore leave uncertainty about the direction and size of the between-group difference.

This is not proof that benfotiamine slowed cognitive decline, and it is not proof that benfotiamine had no effect. A non-significant result does not demonstrate equivalence. Establishing equivalence or ruling out effects of a particular size would require a prespecified framework and suitable precision; the registry does not report such a framework.

The trial's pilot designation and the imprecision of the primary estimate matter when interpreting the result. The observed group means are estimates from this trial, not guaranteed outcomes for other people or settings. The result does not establish what would happen over longer follow-up, and the registry record does not report a clinical-importance threshold for the score difference.

The secondary outcomes do not remove the uncertainty. They include different kinds of measures, have unadjusted comparisons, and include differences between sponsor-reported and recalculated results that the registry does not explain. They should not be used to claim a definitive overall effect.

The findings are limited to the outcomes and analyses posted in the registry. The record does not provide enough information to assess all details of missing data, protocol adherence, analysis populations, or multiplicity. The most supportable reading is that the posted primary estimate favors a lower mean change in the benfotiamine group, but remains imprecise and does not establish a treatment effect.

18. Why This Trial Matters Statistically

This trial illustrates why an estimated effect should be read together with its uncertainty. The primary point estimate is negative in the benfotiamine-minus-placebo convention, but the confidence interval includes zero. Reporting only the direction of the means or only a p-value would leave out important information about how precisely the group contrast was estimated.

It also illustrates the distinction between a sponsor-reported analysis and an unadjusted recalculation from summary statistics. The reported primary p-value and recalculated p-value differ, and the registry does not explain why. For some posted secondary measures, the reported and recalculated comparisons diverge more substantially. Clear reporting of analysis populations, methods, contrast direction, and confidence intervals helps readers understand such differences.

The planning illustrations show how strongly sample-size calculations depend on the target effect. Planning to detect a smaller fraction of the observed effect calls for more participants per arm under the stated assumptions. These illustrations are useful for understanding planning sensitivity, but they are not evidence of the trial's achieved power or a substitute for the original design rationale, which is not reported.

Finally, multiple secondary outcomes, incomplete outcome reporting, and limited detail about missing data complicate interpretation. These are central statistical issues, not merely technical details: they affect how much weight a reader can place on isolated results and how well the findings can be reproduced or compared with future studies.

19. 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 provides limited information about the trial's statistical plan. It does not report the original sample-size justification, the full protocol, a multiplicity strategy, or interim analyses. The planning figures associated with the observed effects are illustrations and do not reveal the assumptions used to design this trial.

Details of outcome data completeness and analysis populations are also limited. The record identifies one exclusion from the primary analysis because a placebo-arm subject took commercial benfotiamine, but does not explain how other missing measurements or treatment deviations were handled. It does not report whether the analysis was intention-to-treat, whether imputation was used, or whether sensitivity analyses were conducted.

The sponsor-reported primary analysis and the recalculation differ in p-value, and the registry does not explain the sign convention or analytic difference. For ADCS-ADL and NPI, the sponsor-reported and recalculated comparisons also differ. The available information is insufficient to resolve these differences; the recalculations should therefore remain clearly labeled as unadjusted analyses of the posted summaries.

Results for all outcomes named in the record are not posted. The secondary outcomes have unadjusted comparisons, and no multiplicity adjustment or outcome hierarchy is reported. This limits how confidently secondary findings can be interpreted, especially when several outcomes are considered together.

The study is identified as a pilot, and the primary confidence interval is broad enough to include values on both sides of zero. The registry does not report a clinical-importance threshold for the ADAS-Cog contrast. Safety information is also descriptive and lacks detail about timing, relatedness, and event-level attribution. These limitations constrain conclusions about both the size of the outcome effect and the wider benefit-risk profile.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Benfotiamine in Alzheimer's Disease: A Pilot Study — NCT02292238.
• Registry record last updated 2022-06-28; status last verified 2022-06.
• Results first posted 2022-06-28.
• 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.

23. Record Summary

NCT02292238 is a completed Phase 2 randomized, quadruple-masked, parallel-group pilot study sponsored by Burke Medical Research Institute. It compared benfotiamine with placebo in a study plan concerning amnestic mild cognitive impairment or mild Alzheimer's disease dementia. The registry reports enrollment of 71.

The primary endpoint was change in ADAS-Cog score from baseline to one year. The recalculated mean difference was -1.87, with a 95% confidence interval from -4.484 to 0.744 and p = 0.161. The interval includes zero, and the estimate is imprecise. The registry's sponsor-reported primary analysis gives p = 0.125, with no reported confidence interval.

Secondary results and safety counts provide descriptive context, but unadjusted comparisons, incomplete methodological reporting, and differences between some sponsor-reported and recalculated values limit interpretation. The registry record does not establish a definitive treatment effect.