Milk Thistle in Pathological Gambling: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Milk Thistle in Pathological Gambling (NCT02337634), a randomized phase 2 study. The primary outcome is Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling (PG-YBOCS).

2. Clinical Question

The trial was intended to evaluate silymarin, the active ingredient identified in the official title, in people with gambling disorder. The primary measure is the Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling, which assesses gambling severity. Its score includes thoughts and urges as well as gambling behavior, with higher total scores indicating worse outcomes.

The registry describes assessments every two weeks during the 8-week study, but the reported primary outcome is one value: change from baseline to Week 8. The posted results compare the mean change among participants with Week 8 PG-YBOCS data. This is the specific question addressed by the primary comparison; it is not a comparison of every scheduled assessment or a summary of all outcomes collected.

The registry record describes the study hypothesis as expecting greater efficacy and tolerability with silymarin than with placebo. That stated hypothesis is not itself a result. The posted primary estimate and its uncertainty describe the observed data; they do not, on their own, establish the clinical importance of a difference.

3. Trial Design

This was a randomized, triple-masked, parallel Phase 2 study sponsored by the University of Chicago. The registry lists an enrollment of 55 and the study status as completed. The recorded start was 2015-01, and primary completion was 2022-12. The trial compared milk thistle with placebo.

In a randomized trial, assignment to groups is intended to make comparisons less dependent on participants’ pre-existing characteristics than a nonrandomized comparison would be. The registry identifies the allocation as randomized, but does not report the sequence-generation method, concealment procedures, or details of how assignment was implemented. Triple masking indicates the recorded masking category; which specific people were masked and how masking was maintained are not reported in the registry record.

The parallel model means the groups are compared as separate treatment groups rather than as a crossover in which participants receive both assignments. The milk thistle arm is described as capsule treatment with a dose range of 150mg BID to 300mg BID. The placebo arm is described as receiving a matched dosage of milk thistle daily. The record does not explain the dose-adjustment procedure, treatment adherence, or whether the placebo description reflects a matching procedure beyond the text recorded.

The 55-person enrollment figure is not the same as the number contributing to each endpoint summary. For the primary outcome, the registry identifies the population as participants with PG-YBOCS data at Week 8; the posted summaries include 21 participants in the placebo group and 16 in the milk thistle group. The reasons that the outcome counts differ from enrollment, and the reasons for any missing Week 8 data, are not reported in the registry record. The original sample-size justification, statistical plan, and interim-analysis plan are also not reported.

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: 12 enrolled participants were excluded from the study prior to randomization because they did not meet the inclusion criteria.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants who had PG-YBOCS data at Week 8

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: One participant assigned to placebo dropped out before being randomized, leaving placebo total at N = 25.

6. Primary Endpoint

The registry lists the primary outcome as Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling (PG-YBOCS), measured over Once every two weeks for the 8 weeks of the study.

The primary endpoint is change from baseline to Week 8 on the Yale Brown Obsessive Compulsive Scale Modified for Pathological Gambling, abbreviated PG-YBOCS. The measure assesses gambling severity, with higher scores indicating worse outcomes. It combines a thought and urge subscale with a behavior subscale. Each subscale has 5 items, each rated from 0-4, and the total score can range from 0-40.

The registry describes PG-YBOCS assessments every two weeks over the 8-week study. Although the measure was scheduled repeatedly, the primary result is a single change score from baseline to Week 8. A change score expresses how much a participant’s score shifted over the specified period; the reported group means summarize those changes for participants with Week 8 data.

As a general statistical rationale, a scale focused on the condition being studied can make a trial’s main comparison directly relevant to the target symptom domain. The registry record does not state why this measure was selected as the primary endpoint, nor does it report a threshold for a clinically important change. Thus, the score’s direction and scale are described, but the clinical meaning of a particular change cannot be determined from the scale range alone.

The reported mean changes were negative in both groups. Given that higher PG-YBOCS scores indicate worse outcomes, a negative change represents a decrease in the measured score on average. That description is limited to the scale and the participants with Week 8 data; it does not establish what caused the changes or how meaningful they were to participants.

7. Results Posted in the Registry

• Milk Thistle: mean -11.86 versus -11.76 on control; difference -0.10 (95% CI -4.98 to 4.78), standardized effect d = -0.01; z-test p = 0.968.

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.

Gambling Symptom Assessment Scale (G-SAS)

• 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:
  • Milk Thistle: mean -19.72 versus -15.54 on control; difference -4.18 (95% CI -10.17 to 1.81), standardized effect d = -0.45; z-test p = 0.171.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Milk Thistle and the control. The interval gives the range of effects that remain compatible with the data.

Sheehan Disability Scale (SDS)

• 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:
  • Milk Thistle: mean -8.00 versus -9.94 on control; difference 1.94 (95% CI -5.16 to 9.04), standardized effect d = 0.20; z-test p = 0.592.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Milk Thistle and the control. The interval gives the range of effects that remain compatible with the data.

Quality of Life Inventory (QOLI)

• 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:
  • Milk Thistle: mean 1.30 versus 10.92 on control; difference -9.62 (95% CI -39.64 to 20.40), standardized effect d = -0.28; z-test p = 0.530.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Milk Thistle and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary measures spanning gambling symptoms, global clinical ratings, anxiety, disability, depression, perceived stress, and quality of life. The posted numerical secondary summaries include the Gambling Symptom Assessment Scale, Sheehan Disability Scale, and Quality of Life Inventory. These are related to the trial’s broader assessment of gambling-related symptoms and functioning, but they measure distinct constructs from the primary PG-YBOCS endpoint.

The directions of the posted secondary mean differences are not uniform. The G-SAS estimate is more negative for milk thistle than placebo, while the SDS and QOLI estimates point in the other direction under their reported comparison direction. This mixed pattern does not provide a single consistent secondary signal. Their uncertainty intervals include zero, and their unadjusted p values do not establish a reliable difference on any of these measures.

Secondary outcomes can add context to a primary result by examining related symptoms or broader functioning. They do not replace the primary endpoint, and differences across scales may reflect differences in what the measures capture, the participants contributing data, and the precision of each estimate. The primary PG-YBOCS comparison remains the principal registry result for the trial’s stated main question.

Unadjusted secondary results carry less weight when many outcomes are examined because repeated testing can increase the chance of observing an apparently notable result by chance. This is a general concern about multiplicity, not evidence that any particular result is false. The registry record does not report a testing hierarchy or multiplicity adjustment. Nor does it provide analysis details for all listed secondary measures. Accordingly, the secondary results are best treated as descriptive and exploratory rather than as independent confirmation of the primary result.

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

• Clinical Global Impression-Improvement and Severity Scales (CGI-I and CGI-S)
• Gambling Symptom Assessment Scale (G-SAS)
• Hamilton Anxiety Rating Scale (HAM-A)
• Sheehan Disability Scale (SDS)
• Hamilton Depression Rating Scale (HAM-D)
• Perceived Stress Scale (PSS)
• Quality of Life Inventory (QOLI)

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is continuous: it is a numerical scale score, and the registry reports means and standard deviations. A mean is the arithmetic average for the participants included in a group summary. A standard deviation describes how dispersed individual scores are around that group mean. It is not the uncertainty interval for the mean difference, and it does not show the full distribution of participant scores.

What does the mean difference represent?

The comparison is expressed as milk thistle minus placebo. The arm means are -11.86 for milk thistle and -11.76 for placebo, and the reported mean difference is -0.1. A negative difference means the milk thistle group’s average change was more negative than the placebo group’s average change, using this comparison direction. Because the scale assigns higher scores to worse outcomes, both group means indicate average decreases in score; the between-group estimate concerns how those average changes differ, not whether either group changed from baseline.

A change score uses each participant’s baseline and Week 8 scores to describe movement over time. The group mean change summarizes those participant-level changes. It is different from comparing only the Week 8 scores, and the registry explicitly identifies the primary result as change from baseline to Week 8. The record does not state whether the comparison was adjusted for baseline score or other characteristics.

What does the confidence interval add?

A confidence interval gives a range of values reasonably compatible with the data and the method used to calculate it. The primary interval runs from -4.984 to 4.784. It includes zero, the value corresponding to no mean difference, and values on both sides of zero. The interval is not a list of participant outcomes, nor does it assign a probability to each possible true difference. Its interpretation depends on the statistical method and assumptions used for the calculation.

The posted comparison also gives an unadjusted p value of 0.968. As a general explanation, a p value measures how surprising data at least as incompatible with a null hypothesis would be if that hypothesis and the test assumptions held. It is not the probability that the null hypothesis is true, and it does not indicate the probability that the observed result will replicate. The registry record does not report the complete test specification or a prespecified decision threshold for the primary comparison.

Why do participant counts and variability matter?

The primary summaries include 21 participants in the placebo group and 16 in the milk thistle group. Group sizes and the spread of observations affect how precisely a mean difference can be estimated. Greater variability generally makes a given sample less informative about the group means, while more observations generally improve precision, provided the study and analysis assumptions are appropriate. This is a general statistical relationship, not a claim about a specific unreported feature of this trial.

The endpoint population is restricted to participants who had PG-YBOCS data at Week 8. The registry does not explain why other enrolled participants are absent from the primary outcome summaries or how any missing scores were addressed. If participants with missing outcomes differ systematically from those with observed outcomes, a complete-case summary can represent a different population from the randomized groups. Whether that issue affected these estimates cannot be assessed from the record.

What is not specified about the analysis?

The registry record does not report a detailed statistical analysis plan, the exact model or test used for the primary comparison, covariate adjustment, assumptions checks, or a missing-data method. The displayed comparison is an unadjusted mean difference with an interval and unadjusted p value. No adjusted primary estimate is reported in the record. It is therefore important not to attribute a particular modeling strategy or analysis population to the trial without documentation.

General statistical methods can make a comparison more precise or address features of the data, but they do not remove the need to understand the design and analysis decisions. For example, an adjusted model may account for specified baseline measures, and a repeated-measures model may use multiple observations over time. The registry says outcomes were assessed every two weeks but reports only baseline-to-Week-8 change as the primary result. It does not report whether the repeated assessments were used in a model.

The comparison and its interval are useful as a concise description of the posted endpoint results. Their interpretation remains conditional on the participants included, the unreported details of the analysis, and the information available in the registry record. No reconstruction of the original analysis plan or its assumptions is possible from the reported summary statistics alone.

11. Confidence Intervals and Precision

The primary mean-difference estimate is -0.1 units, with a confidence interval from -4.984 to 4.784. The point estimate is close to zero, but the interval extends in both directions and includes zero. It is therefore compatible with a range of possible between-group differences, rather than locating the difference precisely at the point estimate.

A confidence interval is a way to express uncertainty in an estimate. In repeated studies conducted under the same assumptions, intervals constructed by a method with a stated coverage level would capture the true parameter at that long-run rate. The registry record does not state the confidence level or full calculation method, so the interval should be read as the posted uncertainty interval without assigning it an unstated coverage level.

The interval is wider than the point estimate’s proximity to zero might suggest on its own. The group summaries include 21 placebo participants and 16 milk thistle participants, and their standard deviations are 7.18 and 7.75. Participant counts and within-group variation contribute to precision, but those details alone do not explain every feature of the interval or establish whether the study was adequately sized.

The interval includes effects in either direction, so it does not settle whether the groups differed in a way that matters clinically. The registry does not report a minimum clinically important difference for PG-YBOCS, and the uncertainty interval cannot be translated into clinical importance without such a benchmark. The appropriate reading is that the estimate is uncertain across a range of values, including no difference, rather than treating a near-zero point estimate as conclusive evidence of equivalence.

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, not a record of the trial’s original design calculation. For these continuous outcomes, the figures are participants per arm, not total events. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The original sample-size assumptions for the trial are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to detect the specified effect with 80% power under those assumptions. The 90% power column gives the corresponding number per arm for 90% power. In general terms, power is the chance that a statistical test will detect an effect of the specified size if that effect is present and the planning assumptions hold. It is not the probability that a treatment is effective.

Rows for 100%, 75%, and 50% of the observed effect illustrate how the target effect size changes the planning requirement. A smaller target effect is harder to distinguish from random variation, so more participants are needed under the same assumptions. The resulting numbers can change substantially with the effect fraction and the outcome’s variability. The displayed figures are not observed power; observed power is not a useful substitute for interpreting the estimate and its confidence interval.

These illustrations use effects estimated from the posted data, so they are sensitive to the uncertainty in those observed effects. They should not be treated as recommendations for a future study or as evidence that the completed trial should have enrolled a particular number. A future design would need a prespecified target effect and justified assumptions; those original design choices are not reported in the registry record.

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 = -0.100; pooled standard deviation σ = 7.430.
2. n = 2 × (1.96 + 0.842)² × 7.430² / -0.100² = 86,651.0, rounded up to 86,651 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 <- -0.0999999999999996                      # observed difference in means
sigma <- 7.42964236785902                     # 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(2337634)
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 = 86,651 (80% power), 116,002 (90% power)
 75% of observed effect: n per arm = 154,047 (80% power), 206,225 (90% power)
 50% of observed effect: n per arm = 346,604 (80% power), 464,005 (90% power)
power.t.test: 86651.9 
Simulated power at n = 86,651 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.

14. Multiplicity, Missing Data and Analysis Populations

The registry names multiple secondary outcomes in addition to the primary PG-YBOCS endpoint. When many outcomes or comparisons are considered, the chance of at least one result appearing unusual by chance can increase. This is the general statistical problem of multiplicity. Researchers may address it by prespecifying a hierarchy, adjusting tests, or identifying secondary analyses as exploratory. The registry record does not report which, if any, approach was planned or used.

The primary comparison is presented with an unadjusted p value. The record does not state a significance threshold for the primary endpoint, nor does it describe whether the secondary comparisons were part of a confirmatory testing strategy. Without those details, the p values should not be interpreted as if each outcome were the sole planned test. No claim can be made from the record that multiplicity was or was not controlled in analyses beyond the posted unadjusted results.

Missing-data details are also limited. The primary population is defined as participants with PG-YBOCS data at Week 8, while enrollment is recorded as 55 and the endpoint summaries include 21 placebo participants and 16 milk thistle participants. The reasons for the difference in counts are not reported. The record does not describe the amount or pattern of missing data at each visit, the method used to handle missing scores, or whether a sensitivity analysis assessed the effect of missing outcomes.

The analysis population is not described as intention-to-treat, per-protocol, or another defined group. The record does not explain how participants who discontinued treatment, changed dose, or lacked endpoint data were treated in the analysis. These are not reasons to assume the results are biased; they are uncertainties that limit assessment of how well the analyzed participants represent everyone randomized.

The registry record also does not report the statistical analysis plan, a hierarchy among endpoints, interim analyses, or the rationale for the original sample size. These omissions prevent a detailed assessment of prespecification and analysis choices. The available interpretation should therefore remain tied to the posted summaries, with the unreported procedures clearly distinguished from the results that are recorded.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent non-serious adverse event terms

The serious adverse event summaries list 26 participants at risk in the placebo group and 17 in the milk thistle group. There were 0 serious adverse events affecting participants in either group, and 0 deaths in either group. These figures describe the serious-event counts recorded in the registry; they do not establish that no other harms occurred.

The registry also reports other, non-serious adverse events affecting 3 placebo participants and 4 milk thistle participants. Those counts are distinct from the serious adverse event figures. The record does not provide event terms for serious adverse events, since no serious-event terms are listed, and does not give a detailed description of the non-serious events in the registry record for this comparison.

The numbers at risk for the safety summaries differ from the primary endpoint group counts. Safety and efficacy populations can have different denominators, and the registry does not explain those differences. The serious-event counts should be read as descriptive counts for the recorded at-risk groups, not as a precise comparison of safety between the assignments. No conclusion about relative safety follows from zero serious events in both groups alone.

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

The primary results describe a close pair of average PG-YBOCS changes among participants with Week 8 data. The mean difference for milk thistle versus placebo was -0.1, and the confidence interval ran from -4.984 to 4.784. That interval includes zero and values on both sides, so the estimate is not precise enough to exclude a range of possible between-group differences.

This result does not show that the groups were equivalent. Equivalence requires a prespecified margin defining how large a difference could be while still counting as practically equivalent, along with an analysis designed to test that question. No equivalence margin or equivalence analysis is reported in the registry record. A nonsignificant or near-zero estimate alone cannot establish equivalence.

Nor does the result establish that milk thistle works or does not work. A statistical comparison of average changes is not a direct determination of clinical benefit. The registry does not state a threshold for meaningful PG-YBOCS change, and the endpoint summaries include participants with Week 8 data rather than a documented full analysis population. The observed decreases in both groups are descriptive and cannot be attributed to one cause from these summaries.

The result is also not a complete account of the trial. The registry lists secondary outcomes and serious adverse event information, but does not provide a full statistical analysis plan, missing-data methods, or detailed analyses for every outcome. These gaps limit how broadly the posted figures can be interpreted. The study’s stated aim and hypothesis provide context, but they are distinct from evidence in the numerical results.

A careful reading keeps three points separate: the estimated average change in each group, the uncertainty around the between-group estimate, and the clinical importance of any difference. The posted data describe the first two. A clinical importance threshold and fuller analysis details are not reported, so the third cannot be resolved from the registry record alone.

17. Why This Trial Matters Statistically

This trial is statistically informative as an example of why a point estimate should be read together with its uncertainty. The primary mean difference is close to zero, but the confidence interval ranges across values on both sides of zero. A single point estimate does not communicate how much uncertainty remains; the interval makes that uncertainty central to interpretation.

The endpoint population also matters. The registry records enrollment of 55, while the primary outcome summaries include 21 placebo participants and 16 milk thistle participants. The primary results therefore summarize participants with Week 8 data, as the record specifies, rather than every enrolled participant. Without information about missing data and analysis populations, readers cannot determine how the endpoint summaries relate to all randomized participants.

The trial also illustrates the distinction between a study’s objective and the evidence needed to answer it. The registry states a hypothesis about silymarin, but the primary estimate and its interval remain the basis for interpreting the posted comparison. The hypothesis does not change the statistical meaning of an unadjusted estimate or resolve its uncertainty.

Finally, the presence of several secondary measures makes multiplicity relevant. A broad outcome battery can describe several dimensions of participants’ experience, but secondary findings need context about prespecification and adjustment. Those details are not reported in the registry record. The record supports a measured description of the primary result and available safety counts; it does not support stronger claims than the design, endpoint population, and reported precision allow.

18. Limitations

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

The primary endpoint summaries cover participants with PG-YBOCS data at Week 8, not necessarily everyone randomized. The registry does not explain why the reported endpoint counts are 21 in the placebo group and 16 in the milk thistle group, despite recorded enrollment of 55. It also does not describe missingness by visit, reasons for missing outcomes, or methods used to address missing data. This limits assessment of potential selection effects in the endpoint summaries.

Details needed to evaluate the analysis are not reported, including the complete statistical plan, the primary test specification, any covariate adjustment, and the analysis population definition. The record describes repeated assessments but reports one primary change score. It does not explain whether information from the intermediate assessments was incorporated into an analysis. The original sample-size justification and any interim-analysis plan are likewise not reported.

The registry does not state a clinically important difference for the PG-YBOCS or explain how a score difference should be mapped to meaningful changes in symptoms or functioning. Therefore, even though the scale direction and range are described, the clinical importance of the estimated difference cannot be determined from the scale summary alone.

Multiple secondary measures are listed, but a testing hierarchy or multiplicity adjustment is not reported. The posted secondary results should consequently be interpreted cautiously. The record also provides limited detail about masking implementation, treatment adherence, dose changes, and the nature of non-serious adverse events. These limitations do not determine the direction of any bias, but they constrain what can be concluded from the registry summaries.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Milk Thistle in Pathological Gambling — NCT02337634.
• Registry record last updated 2026-07-07; status last verified 2024-02.
• Results first posted 2024-09-04.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry record describes a completed randomized, triple-masked, parallel Phase 2 study of milk thistle versus placebo for gambling disorder, sponsored by the University of Chicago. Enrollment was 55, with the recorded study start at 2015-01 and primary completion at 2022-12.

The primary outcome was baseline-to-Week-8 change on the PG-YBOCS among participants with Week 8 data. The posted mean difference for milk thistle versus placebo was -0.1, with a confidence interval from -4.984 to 4.784 and an unadjusted p value of 0.968. The interval includes zero and values in either direction.

The record lists multiple secondary measures and reports serious adverse events affecting 0 participants in each group, with 0 deaths in either group. Original design assumptions, missing-data procedures, analysis-population details, and a full statistical plan are not reported in the registry record.