A Preliminary Study of Choline and Betaine Supplementation Among Adults Exposed to Arsenic in Bangladesh: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Preliminary Study of Choline and Betaine Supplementation Among Adults Exposed to Arsenic in Bangladesh (NCT01749982), a randomized phase 4 study. The primary outcome is Change in Urinary % Monomethyl Arsenic.

2. Clinical Question

The trial asked whether daily choline bitartrate, betaine, or their combination could be studied in adults exposed to arsenic in Bangladesh, with placebo as the comparator. Its primary endpoint was the change from baseline to week 8 in urinary monomethyl arsenic as a percentage of total urinary arsenic. The posted results describe this change separately for the placebo group and each supplement group.

The study rationale concerns arsenic metabolism and urinary excretion. The registry summary explains that arsenic methylation involves two steps and that choline and betaine contribute to the methyl pool. It also notes prior observational findings linking plasma choline and betaine concentrations with arsenic methylation. The rationale motivates testing supplementation, but an association from prior observations does not establish what this randomized pilot will show.

The registry describes additional pilot aims involving acceptance of supplements, monitoring for potential side effects, and practical difficulties with daily follow-up. Results for these aims are not reported in the registry record. The posted primary outcome therefore addresses only one part of the broader study rationale.

3. Trial Design

This was a randomized, triple-masked, factorial pilot study. The registry lists the phase as Phase 4 and the study status as completed. Columbia University is named as sponsor. The recorded enrollment is 60. The study start is recorded as 2012-12 and primary completion as 2013-03.

The four assigned regimens were placebo tablets, choline bitartrate 700 mg by mouth daily, betaine 1000 mg by mouth daily, and choline bitartrate 700 mg plus betaine 1000 mg daily. A factorial design evaluates more than one intervention component within a study structure. In general, this can support investigation of separate components and their combination, but the registry record does not report the statistical model or contrasts used to estimate such effects. The posted primary results are group-specific summaries rather than factorial effect estimates.

Randomization means that assignment was intended to be determined by chance. Triple masking indicates that the record identifies the study as triple-masked, but the identities of the masked parties and the practical masking procedures are not reported in the registry record. The record also does not provide allocation-concealment details.

The primary outcome timeframe was from baseline to 8 weeks after the start of the intervention, expressed as week 8 minus baseline. The outcome was the change in urinary percentage of monomethyl arsenic. Results are reported as medians with full ranges. The reported outcome group sizes are 15 for placebo, 15 for choline bitartrate, 15 for betaine, and 12 for choline bitartrate plus betaine. Reasons for the differing reported group sizes are not reported in the registry record.

The study was designated a pilot, and its stated aims included feasibility and preliminary data generation. The registry record does not report a protocol-level sample-size justification, a prespecified hypothesis for a particular pairwise comparison, or a detailed analysis plan. The posted record lists no analyses and does not post secondary outcome results.

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:

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 Change in Urinary % Monomethyl Arsenic, measured over From baseline to 8 weeks after the start of the intervention (week 8 - baseline).

The primary endpoint is the change in urinary monomethyl arsenic, expressed as a percentage of total urinary arsenic, from baseline to week 8. A change is defined in the record as the week 8 value minus the baseline value. A positive value therefore denotes an increase in that percentage over the stated interval, while a negative value denotes a decrease. This describes the direction of the recorded measure and does not itself indicate whether a change is beneficial or harmful.

The outcome focuses on one component of urinary arsenic composition rather than the total amount of arsenic in urine. The registry rationale describes arsenic methylation as part of the metabolism that facilitates urinary arsenic excretion and explains that choline and betaine contribute to the methyl pool. In that context, the urinary percentage of monomethyl arsenic is a measure relevant to the study's biological question. It is not a direct measure of clinical health or of arsenic-related disease.

The primary outcome is summarized with the median and full range for each group. The median identifies the middle of the observed group distribution, while the full range gives the lowest and highest reported values. These summaries do not show how values were distributed between the endpoints of the range, and they do not quantify uncertainty in the group median. Confidence intervals and formal comparisons are not reported for this endpoint in the registry record.

The record identifies the time window and measure, but further details about specimen collection, laboratory procedures, or handling of repeated measurements are not reported in the registry record. Those details would be needed to assess measurement procedures more fully.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary results use medians and full ranges. General statistical explanation: the median is a measure of location that is less affected by an unusually high or low observation than the arithmetic mean. It is useful when a distribution may be skewed or when extreme observations are present. The median describes a typical position in the observed sample, but it does not communicate how precise an estimate of the underlying group value is.

What does a full range describe?

The full range gives the lowest and highest observed values in a group. It provides a direct description of the observed spread, but it is sensitive to extreme observations and does not show the shape of the distribution. Two groups may have similar ranges while differing in the way their observations are concentrated within those ranges. Conversely, a range can be wide even when many observations lie close together. The reported ranges should therefore be read together with the medians, not as a substitute for individual-level distributions or uncertainty intervals.

General statistical explanation: a confidence interval estimates a range of values compatible with the data and a specified statistical procedure. It helps express uncertainty about an estimated parameter, such as a group difference. A full range is different: it describes observed sample extremes. The registry record reports full ranges for the group changes, but no confidence intervals for group medians or for treatment comparisons.

What is the primary contrast?

The endpoint summaries are presented separately for placebo, choline bitartrate, betaine, and their combination. A comparative treatment effect would usually be expressed as a contrast between groups, accompanied by an estimate of uncertainty and, where appropriate, a statistical test. Those comparative estimates are not reported in the registry record. It would be inappropriate to infer a formal treatment effect by comparing the listed medians informally, because the medians alone do not account for sampling variability, the factorial structure, or any analysis choices.

The study is labelled factorial. General statistical explanation: a factorial design can examine intervention components individually and in combination, and can also evaluate whether the components interact. Interaction means that the estimated effect of one component differs depending on whether another component is present. Such questions require specified contrasts and an analysis plan. The registry record does not report a model, interaction estimate, or factorial contrast, so the group summaries do not establish whether component effects were analyzed separately or jointly.

Why does variation matter?

General statistical explanation: estimates from small groups can be imprecise because each observation has a relatively strong influence on the summary. Greater variability also makes it harder to distinguish a real group contrast from random variation. Here, the posted group sizes are modest, and the full ranges span both negative and positive changes in each group. These features counsel against treating the group medians as definitive estimates. They do not, by themselves, establish whether a formal comparison would be statistically significant.

General statistical explanation: over-dispersion occurs when observed variability is greater than a statistical model assumes. If present, a model that ignores it may understate uncertainty and produce overly confident inferences. Whether over-dispersion is relevant depends on the outcome and model. The registry record does not report a fitted model or diagnostics, so it is not possible to determine whether the investigators assessed this issue for the primary endpoint.

What do the planning assumptions mean?

The record describes a planning illustration based on the observed effect, with two-sided significance level 0.05, equal allocation between two arms, and a normal approximation. It states that the effect to detect is a stated fraction of the observed effect. General statistical explanation: a planning calculation estimates the group size that a hypothetical future study might require under its assumptions. It is not a result from the trial and does not show the probability that this completed study detected an effect. It is not observed power.

The record notes that sample-size illustrations for non-time-to-event outcomes are expressed as participants per arm. The numerical planning outputs are not part of the registry entries for this interpretation, so no sample-size estimate is stated. These planning assumptions also do not establish that the original pilot was designed around a formal efficacy hypothesis. Its own sample-size justification is not reported in the registry record.

What can be inferred from missing inferential results?

The registry lists no analyses and provides no p-values, confidence intervals, or adjusted estimates for the primary outcome. That means the posted primary findings are descriptive. A p-value, in general, evaluates how compatible data are with a specified null model under stated assumptions; it is not the probability that a treatment works. A confidence interval communicates uncertainty around an estimate but also depends on the analysis model. Neither can be reconstructed reliably from medians and ranges alone.

Overall, the statistical information supports a description of the group-level changes and observed ranges. It does not support a precise comparative estimate or an inferential conclusion about treatment effects. Further interpretation would require the prespecified analysis methods, participant-level data or appropriate summary statistics, and reporting of the planned contrasts and their uncertainty.

10. Multiplicity, Missing Data and Analysis Populations

The registry describes a factorial study with placebo, choline bitartrate, betaine, and combined supplementation. Several possible questions could arise from this structure: comparisons of individual regimens with placebo, comparisons of the supplement components, or an assessment of whether the components interact. The registry record does not report which comparisons were prespecified or how multiplicity was handled. When several hypotheses are tested, general statistical practice considers the increased chance of at least one apparently notable result arising by chance. Without the analysis plan, it is not possible to know whether or how this concern was addressed.

The record lists no analyses and no posted secondary outcome results. No primary-outcome p-value, confidence interval, or adjusted estimate is reported. Thus, multiplicity cannot be evaluated from the posted inferential results, and the group medians should not be treated as the outcome of a formal set of comparisons.

For the primary endpoint, the registry gives outcome group sizes of 15 for placebo, 15 for choline bitartrate, 15 for betaine, and 12 for the combined regimen. It does not report why the endpoint group sizes differ from each other or from the enrollment figure. The number of participants with missing endpoint values, reasons for missingness, and any imputation or other missing-data procedure are not reported in the registry record. No conclusion about attrition or its causes can be drawn from the group counts alone.

The analysis population is also not specified. The registry record does not say whether the primary summaries include all randomized participants, only participants who completed the intervention, or another defined group. It does not report whether participants were analyzed according to assigned regimen or actual supplementation. These omissions matter because choices about analysis population and missing data can affect interpretation, particularly when group summaries are based on different reported numbers.

11. Safety Results

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

Other (non-serious) adverse events

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

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events for each group. The serious-event at-risk count is 15 for placebo, 15 for choline bitartrate, 15 for betaine, and 15 for choline bitartrate plus betaine. These are the posted registry figures for the listed groups. They do not establish that serious events were impossible or that the interventions are generally free of serious risks.

The record also reports other adverse events in each group: 5 for placebo, 5 for choline bitartrate, 2 for betaine, and 2 for the combined regimen. The corresponding other-event at-risk count is 15 for each group. The registry does not provide event terms in the registry record recorded for this analysis, nor does it report a statistical comparison of adverse-event frequency between groups. The counts therefore describe the posted safety entries and should not be read as evidence that one regimen is safer than another.

Deaths are not reported in the registry record. Details about event severity, timing, relationship to study treatment, or participant-level recurrence are also not reported in the registry record. The study summary says that participants were to be monitored for potential side effects, but the posted figures alone do not show the monitoring procedures or completeness of reporting.

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

The result establishes what the registry reports for the primary biomarker endpoint: group medians and full ranges for change in urinary monomethyl arsenic percentage from baseline to week 8. The values describe variation in the outcome across participants within each listed regimen. The endpoint is connected to the study's interest in arsenic methylation, as described in the registry rationale.

The result does not establish that any supplement changed the endpoint relative to placebo. Although the group medians differ, the registry does not provide a between-group estimate, confidence interval, or statistical test. An apparent numerical contrast between group summaries is not itself evidence of a statistically reliable difference. The reported ranges show observed extremes, not uncertainty around comparative effects.

The measure also does not establish a change in total arsenic exposure, arsenic-related illness, or long-term health outcomes. It is a percentage of total urinary arsenic, and it is reported over the stated interval. The registry record does not report clinical outcomes that would connect this biomarker result to a health benefit or harm.

The study's pilot aims include assessing supplement acceptance, monitoring potential side effects, identifying daily follow-up difficulties, and generating preliminary data. The primary outcome figures do not answer all of those questions. Results for acceptance and follow-up feasibility are not reported in the registry record. The safety entries report zero serious adverse events in the listed groups and counts of other events, but the record does not provide the detail needed to assess event types or comparative safety.

More broadly, the posted results should be understood as descriptive pilot data. They can inform the record of the observed biomarker changes and identify the types of outcomes that were collected. They do not establish the magnitude or precision of a treatment effect, validate the biomarker as a clinical surrogate, or show whether a future study would reproduce these findings. Such judgments would require fuller methods and inferential reporting than is available in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically informative because its design brings together a placebo comparison and two supplement components, including their combination, in a factorial pilot. That arrangement raises questions about component effects and possible interaction. The registry identifies the factorial design, but does not report the corresponding statistical model or contrasts. The distinction is important: four group summaries do not automatically answer every question that a factorial design can pose.

The primary outcome also illustrates why the form of reporting matters. Medians and full ranges provide a compact description of changes in a biomarker, but they are not comparative effect estimates and do not quantify uncertainty in a contrast. Readers can see the reported central values and observed extremes, while the absence of confidence intervals and tests leaves the precision of group differences unresolved. This is a useful reminder that descriptive statistics and inferential statistics answer different questions.

The pilot designation provides context for the scale and purpose of the record. Its stated aims include feasibility, monitoring potential side effects, and preliminary data generation, not only measurement of a biomarker. Yet the registry does not report results for the feasibility aims or a sample-size justification. The posted primary outcome therefore offers only a partial account of the study's stated objectives.

The record also illustrates how participant counts need context. The endpoint summaries use different reported group sizes, but the reasons are not given. Without analysis-population definitions and missing-data reporting, it is not possible to determine what those differences mean for interpretation. Transparent reporting of these details would help distinguish a descriptive result from a more complete analysis.

14. 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 does not report a detailed protocol or statistical analysis plan. In particular, it does not identify a prespecified primary comparison, the model used for the factorial design, or whether an interaction between choline and betaine was assessed. This limits interpretation of the four group summaries as estimates of specific treatment effects.

The primary outcome is reported with medians and full ranges, without confidence intervals, p-values, or other measures of uncertainty for group contrasts. No participant-level distributions or additional descriptive statistics are reported in the registry record used. The posted summaries therefore do not show the shape of the outcome data or the precision of the differences among groups.

The registry records enrollment of 60, while the reported primary outcome group sizes are 15, 15, 15, and 12. The reason for the different endpoint group sizes is not reported in the registry record. Eligibility for analysis, loss to follow-up, missing measurements, and the handling of any missing values are likewise not described. The analysis population and treatment of adherence are not reported.

The study's time frame for the primary endpoint ends at week 8. Longer-term biomarker measurements and clinical outcomes are not reported in the registry record. The stated pilot objectives include supplement acceptance and challenges with daily follow-up, but results for those objectives are not reported. Safety reporting includes event counts but not event terms, timing, or relationship to assigned regimen. These gaps limit both the biological and feasibility interpretation of the posted results.

Finally, the registry's brief summary describes the study rationale but does not supply enough detail to assess laboratory methods, participant characteristics, or the process used to maintain masking. These limitations do not negate the recorded summaries; they define what can be concluded from them.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Preliminary Study of Choline and Betaine Supplementation Among Adults Exposed to Arsenic in Bangladesh — NCT01749982.
• Registry record last updated 2014-08-28; status last verified 2014-08.
• Results first posted 2014-08-28.
• 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.

18. Record Summary

The registry record describes a completed, randomized, triple-masked factorial pilot sponsored by Columbia University, with a recorded enrollment of 60. It lists choline bitartrate, betaine, their combination, and placebo tablets. The primary endpoint is the median change in urinary monomethyl arsenic as a percentage of total urinary arsenic from baseline to week 8, with full ranges reported by group.

The posted endpoint group sizes are 15 for placebo, 15 for choline bitartrate, 15 for betaine, and 12 for the combined regimen. The record reports zero serious adverse events in each group and other adverse-event counts of 5, 5, 2, and 2, respectively. It lists no analyses and no secondary outcome results. Comparative estimates, confidence intervals, p-values, details of missing data, and the analysis population are not reported in the registry record. The outcome summaries are therefore descriptive and do not establish a comparative treatment effect.