Impact of Metformin in Teens With Polycystic Ovary Syndrome (PCOS) on Oral Contraceptive Therapy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Impact of Metformin in Teens With Polycystic Ovary Syndrome (PCOS) on Oral Contraceptive Therapy (NCT00283816), a randomized phase 3 study. The primary outcome is Reduction in Abdominal Fat as Measured by Waist Circumference..

2. Clinical Question

The registry frames the study around the metabolic impact of oral contraceptives with or without metformin in adolescents with PCOS. Its brief summary notes that oral contraceptives may improve menstrual cycles and symptoms, while raising concern about cholesterol, and that metformin may affect metabolic state. Those statements describe the rationale, not findings from this trial. The posted primary outcome is waist circumference, and the secondary outcomes include change in weight, total testosterone, and sex hormone binding globulin (SHBG).

The record also lists a lifestyle management program intended to reduce weight. It does not report how that program fits into the randomized comparison, whether it was offered to both arms, or how its effects were handled in the analysis. The clinical question that can be read directly from the posted arm descriptions and primary outcome is therefore the comparison of metformin with placebo for change in waist circumference, in the context of the study’s stated oral-contraceptive question. The record does not provide a separate estimate for the effect of oral contraceptives or the lifestyle program.

3. Trial Design

The registry classifies this as a phase 3, randomized, double-masked, parallel-group study and marks it completed. It lists an enrollment of 36. The two outcome arms are labeled metformin and placebo; each has 16 participants for the primary outcome and the posted secondary outcomes. Those outcome counts do not explain the difference between the enrollment figure and the arm counts, and reasons for any difference are not reported in the registry record.

The trial’s official title describes obese adolescents with PCOS and contrasts oral contraceptives with or without metformin. The brief summary similarly frames metformin as an addition to oral-contraceptive therapy. The arm descriptions identify metformin and placebo, but the registry record does not detail the oral-contraceptive regimen, its duration, or whether other care was identical between arms. It also lists a lifestyle management program, quality-of-life questionnaire, oral glucose tolerance test, blood work, abdominal ultrasound, and DEXA scan as interventions or procedures. Their timing and relationship to the randomized comparison are not reported in the registry record.

The primary outcome was measured at baseline and 24 weeks. The secondary outcomes are also described as changes between baseline and 24 weeks. The record gives no further visit schedule, eligibility details, allocation method, masking procedures, or account of how participants were followed. The study’s own original sample-size assumptions and statistical analysis plan are not reported in the registry record. Although the registry marks the study completed, that status alone does not explain how the trial was conducted or how its analyses were selected.

The enrollment and outcome arm counts should be read as separate pieces of registry information, rather than as evidence about why participants may not appear in the posted outcomes. No participant flow, analysis population definition, or explanation of missing outcome data is reported. These gaps limit assessment of how closely the reported results represent all participants enrolled in the study.

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

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 Reduction in Abdominal Fat as Measured by Waist Circumference., measured over baseline and 24 weeks.

The primary endpoint is change in waist circumference, measured in centimeters, between baseline and 24 weeks. The registry describes waist circumference as a measure of abdominal adiposity. It labels the outcome as a mean and gives a standard deviation for each arm, so the posted arm results describe average change and the spread of individual changes around each average.

The outcome description defines change as pre minus post intervention. That direction matters when reading the signs: a negative value under this definition is not automatically a reduction. The registry title calls the outcome a reduction in abdominal fat, but the actual measure is waist circumference, not a direct measure of abdominal fat. The posted values should therefore be interpreted according to the stated pre-minus-post convention, without relabeling a negative value as a decrease in waist circumference.

Waist circumference is used as a proxy measure of abdominal adiposity. A circumference measure can summarize body size around the abdomen, but it does not directly quantify fat tissue or establish a change in metabolic health. The endpoint also does not, by itself, show whether participants experienced changes in PCOS symptoms or other clinical outcomes. Those questions would require outcomes designed to measure them.

The comparison is between the mean change in the metformin arm and the mean change in the placebo arm. This contrast estimates the difference between the groups’ average changes, rather than the change within either group alone. The registry does not report an adjusted analysis or additional detail about endpoint measurement procedures.

7. Results Posted in the Registry

• Metformin: mean -3.90 versus -1.40 on control; difference -2.50 (95% CI -9.30 to 4.30), standardized effect d = -0.25; z-test p = 0.471.

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 in Weight Post Minus Pre Intervention.

• 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:
  • Metformin: mean -1.90 versus -1.20 on control; difference -0.70 (95% CI -3.89 to 2.49), standardized effect d = -0.15; z-test p = 0.667.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Metformin and the control. The interval gives the range of effects that remain compatible with the data.

Total Testosterone Change

• 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:
  • Metformin: mean -57.70 versus -27.60 on control; difference -30.10 (95% CI -48.28 to -11.92), standardized effect d = -1.15; z-test p = 0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Metformin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Sex Hormone Binding Globulin (SHBG)

• 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:
  • Metformin: mean -73.40 versus -58.00 on control; difference -15.40 (95% CI -38.35 to 7.55), standardized effect d = -0.46; z-test p = 0.188.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Metformin and the control. The interval gives the range of effects that remain compatible with the data.

The three posted secondary outcomes address different aspects of the study’s metabolic question: change in weight, total testosterone, and SHBG. They are related to the primary waist-circumference measure, but they are not substitutes for it. A pattern across them can provide context about whether the observed group contrasts align across outcomes; it cannot establish that one outcome explains another or that a change in one measure caused a change in another.

Across the secondary outcomes, the estimated contrasts do not all carry the same statistical evidence. One interval excludes zero, while the other secondary intervals include zero. The corresponding unadjusted p-values likewise differ. This is a mixed pattern rather than a uniform signal across all outcomes, and it does not remove the uncertainty in the primary comparison. No sponsor-reported analyses are listed in the registry record, so these comparisons are recalculations from the posted arm summaries.

There are multiple outcomes under consideration: three secondary outcomes are posted alongside the primary endpoint. When many outcomes are examined, some may appear more statistically notable by chance, even if there is no underlying effect. General statistical principle: multiplicity procedures can control the chance of false-positive findings across a family of tests, but their use and the family of tests must be specified. The registry does not report whether an adjustment, hierarchy, or other multiplicity strategy was planned.

For that reason, unadjusted secondary results carry less weight than a result from a prespecified primary comparison, particularly when the analysis plan is unavailable. They may be useful for generating hypotheses or identifying outcomes for further study, but they should not be presented as independent confirmation of the primary finding. The registry also does not report whether all intended secondary outcomes were measured, how missing observations were handled, or whether these outcomes were selected before results were known. Those gaps further limit the strength of conclusions drawn from the secondary pattern.

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

• Change in Weight Post Minus Pre Intervention.
• Total Testosterone Change
• Change in Sex Hormone Binding Globulin (SHBG)

9. Statistical Interpretation

10. Statistical Methods Explained

What does the mean difference represent?

The primary outcome is continuous: waist circumference change is measured in centimeters. For a continuous outcome, each arm’s mean gives the average observed change among the participants contributing to that arm’s outcome summary. The comparison subtracts the placebo mean from the metformin mean. The recalculated unadjusted mean difference is -2.5 cm. It describes the separation between the posted arm averages; it is not an individual participant’s change and it is not a measure of how many participants improved.

The registry defines the primary change as pre minus post. Preserving that convention is essential for interpreting the sign. A negative mean difference indicates that the metformin arm’s pre-minus-post average is lower than the placebo arm’s pre-minus-post average. Translating this into a claim about a reduction or increase in post-intervention waist circumference would require keeping the subtraction order in view. The endpoint title does not override the definition.

What do the standard deviations describe?

A standard deviation describes how much individual observations vary around an arm’s mean. It is not the uncertainty in the mean itself, and it is not the range containing every participant’s value. The primary outcome standard deviations are 8.6 cm for metformin and 10.9 cm for placebo. The wide spread relative to the mean changes indicates that individual changes varied considerably within the groups. General statistical principle: greater variability makes it harder to estimate a group mean precisely when the number of observations is limited.

The posted comparison uses arm sizes of 16.0 and 16.0 and a pooled standard deviation of 9.818 for its recalculation. Pooling combines the within-arm variability under the assumptions used for this simple comparison. It does not adjust for other characteristics or account for a more complex repeated-measures model. Whether a different model was prespecified or used by the investigators is not reported in the registry record.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the observed estimate under the statistical method used. For the primary mean difference, the interval is -9.303 to 4.303 cm. Because zero lies inside that interval, the data do not distinguish the estimated contrast from no difference at conventional levels using this unadjusted calculation. The interval also includes values in both directions, so it does not settle which arm had the more favorable average change under a clinically meaningful interpretation.

General statistical explanation: a confidence interval is not a statement that a particular percentage of individual participants’ results fall within its limits. Nor does it assign a probability to the true effect after the data have been observed. Its width reflects estimation uncertainty under the chosen method. The registry does not state the confidence level or detail the method used for an investigator-reported interval; the posted intervals are recalculated summaries.

What does the p-value tell us?

The primary unadjusted p-value is 0.471. In general, a p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, the probability that the result happened by chance, or a measure of clinical importance. A larger p-value does not prove that two treatments have identical effects.

The posted p-values are labeled unadjusted. They do not account for examining multiple outcomes or comparisons. The registry lists three secondary outcomes in addition to the primary outcome, but does not report a multiplicity procedure. Consequently, nominal evidence for an individual secondary outcome should not be treated as if it were the sole planned test. The study’s prespecified testing hierarchy and decision rules are not reported in the registry record.

Why does over-dispersion matter in some analyses?

General statistical explanation: over-dispersion occurs when observed variability is greater than the model assumes. It can make standard errors too small and confidence intervals too narrow if the model does not accommodate the extra variation. It is especially relevant for count outcomes analyzed with models that expect a particular relationship between the mean and variance. The outcomes posted for this trial are continuous mean changes, and the registry does not describe a count model or report an over-dispersion assessment. This concept is therefore not a reported feature of the trial’s primary analysis.

What is recalculated, and what is not known?

The comparison estimates and intervals are simple unadjusted recalculations from the posted arm summaries. They do not claim to reproduce an investigator model, adjust for baseline characteristics, or account for missing outcome observations. No sponsor-reported analyses are listed. The registry record does not report the original analysis plan, the handling of repeated measurements, covariate adjustments, or an intention-to-treat definition.

Because the record gives group means, standard deviations, and outcome counts, a basic group comparison can be described. That does not reveal whether the trial used a different model in its own analysis or whether its investigators reached a different conclusion. The absence of reported analyses is an important qualification: the recalculations aid interpretation of the posted summaries, but they should not be represented as the sponsor’s original inferential results.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of -2.5 cm, with a confidence interval from -9.303 to 4.303 cm. The interval spans zero and extends in both directions. It therefore leaves substantial uncertainty about the direction and size of the between-arm contrast. The point estimate is only one value within a wider range compatible with the posted summary data.

Confidence intervals describe uncertainty in an estimated group contrast under the method used; they do not describe the spread of individual participant outcomes. The primary arm standard deviations, 8.6 cm and 10.9 cm, describe that within-arm spread. The distinction matters: variability among participants and uncertainty about the average difference are related but not interchangeable quantities.

The primary comparison is based on 16 participants in each outcome arm. That is a limited basis for estimating a mean difference when the reported standard deviations are substantial. General statistical principle: limited information and high variability tend to produce less precise estimates. The interval’s width is consistent with a result that does not sharply distinguish no difference from differences in either direction.

The registry does not report a clinically important difference for waist circumference. Statistical compatibility with a range of values therefore cannot be translated into a judgment about clinical importance. Nor does an interval containing zero establish that the treatments are equivalent; equivalence requires a prespecified margin and a method designed to test that question. No such margin is reported in the registry record.

These intervals and p-values are unadjusted recalculations from the posted summaries. They should be distinguished from any sponsor analysis, which is not listed in the registry record. The record also does not report confidence-interval methods or whether baseline values were incorporated through an adjusted model. Those omissions limit assessment of the precision and robustness of the intended analysis.

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 table and curve are planning illustrations, not a report of the trial’s original design calculations. For these continuous outcomes, the figures mean participants per arm, not a total across arms or total events. The registry’s planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses a stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column describes the planning number of participants per arm for a hypothetical study to have an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% column gives the corresponding planning number for a 90% chance. These are design targets, not probabilities that the current study’s result is correct or that a treatment effect exists. The figures do not revise the observed trial results.

The rows show planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. A smaller effect is harder to distinguish from random variation, so the planning estimates require more participants per arm. For the primary outcome, the figures range from 243 participants per arm for 80% power when planning for 100% of the observed effect to 969 participants per arm when planning for 50% of that effect. These values are illustrations based on the observed estimate, not recommendations for a future trial; future design choices would depend on a justified target effect and design assumptions. This is not observed power.

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 = -2.500; pooled standard deviation σ = 9.818.
2. n = 2 × (1.96 + 0.842)² × 9.818² / -2.500² = 242.1, rounded up to 243 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 <- -2.5                      # observed difference in means
sigma <- 9.81758626139847                     # 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(283816)
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 = 243 (80% power), 325 (90% power)
 75% of observed effect: n per arm = 431 (80% power), 577 (90% power)
 50% of observed effect: n per arm = 969 (80% power), 1,297 (90% power)
power.t.test: 243 
Simulated power at n = 243 per arm: 0.796 (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 lists a primary endpoint and three secondary outcomes, but does not report how the investigators planned to address multiple testing. When several outcomes are tested, the chance of at least one apparently notable result can increase. This is a general statistical issue; it does not establish that any particular finding in this trial is false. Methods such as adjustment procedures or a prespecified hierarchy can clarify how evidence across outcomes should be interpreted. Whether such methods were used is not reported in the registry record.

The posted p-values are unadjusted recalculations. No sponsor-reported analyses are listed, and the registry does not provide a statistical analysis plan, endpoint hierarchy, or criteria for interpreting the secondary outcomes. The secondary comparisons should therefore be treated cautiously and not as if each had been the only planned test. The record also does not report whether the listed outcomes represent all outcomes measured during the study.

Missing-data handling is another unresolved issue. The outcome summaries list 16 participants per arm, while the registry enrollment is 36. The record does not explain the relationship between those figures, give a participant flow, or state whether outcome data were unavailable for any enrolled participants. It does not report imputation, complete-case analysis, or sensitivity analyses for missing values. No conclusions about the reasons for the outcome counts or the impact of missing data can be drawn from the registry record.

The analysis population is also not defined. The record does not say whether the reported summaries follow an intention-to-treat approach, include only participants who completed treatment, or use another population definition. It does not report adherence, protocol deviations, or withdrawals. These details matter because analysis choices can affect the interpretation of a randomized comparison. Without them, the posted outcome summaries can be described, but their relationship to the original randomization and to all enrolled participants remains uncertain.

Finally, no adjusted analysis is reported. Whether the investigators intended to adjust for baseline measurements or other variables, and whether they used a repeated-measures method, is not reported in the registry record. The unadjusted recalculations should not be mistaken for a complete account of the trial’s analysis plan.

15. 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 1% reporting threshold.

The registry reports serious adverse event counts for 18 participants at risk in the metformin arm and 18 at risk in the placebo arm. It records zero serious adverse events in each arm. The other adverse event counts are also recorded as zero in both arms, with 18 participants at risk in each arm. These are the figures entered in the registry record, not a guarantee that no event occurred outside the registry’s reported categories or observation period.

The deaths fields are empty rather than populated with a count. Deaths are therefore not reported in the registry record; an empty field should not be interpreted as a reported zero. The record does not give event definitions, monitoring duration, severity details beyond the serious-event category, or a narrative account of events. It also does not describe how adverse events were solicited or attributed.

Safety counts should be read with their denominators and reporting context. The at-risk counts are not the same as the outcome analysis counts, and the record does not explain the difference between those sets of counts. The posted zero counts provide a description of reported events in the recorded groups, but limited event information and lack of follow-up detail restrict broader conclusions.

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

The posted primary result is an uncertain estimate of the difference in mean waist-circumference change between the metformin and placebo arms. Its confidence interval includes zero and extends in both directions. The result therefore does not provide a precise estimate that clearly distinguishes the arms on this endpoint. The point estimate and p-value should be read alongside the interval, rather than used alone to make a categorical claim.

This finding does not establish that the arms are equivalent. An equivalence conclusion would require an explicit margin defining differences small enough to count as equivalent, together with a suitable analysis. No such margin or equivalence analysis is reported. Conversely, the uncertainty does not establish that an important difference exists. The registry does not give a threshold for a clinically meaningful waist-circumference change.

The endpoint is a proxy measure of abdominal adiposity, not a direct assessment of all metabolic effects, PCOS symptoms, or quality of life. The secondary outcomes offer additional measures, but their unadjusted comparisons and the absence of reported multiplicity handling mean they should not be treated as definitive confirmation. The registry’s rationale about possible effects of oral contraceptives and metformin is background, not evidence that these effects occurred in the trial.

The results apply to the trial groups and measurements described in the registry record. The record does not provide details needed to judge generalizability, such as fuller eligibility criteria or demographic characteristics. It also does not clarify how the listed lifestyle program related to treatment assignment, or whether it influenced outcomes. The completed status and phase classification do not fill in those gaps.

Finally, the posted comparisons are recalculated from arm-level summaries, and no sponsor-reported analyses are listed. They do not account for any unreported adjustments, missing-data procedures, or analysis-population choices. The available data support a cautious description of the recorded group summaries, not a broad claim about treatment effects beyond the outcomes and context recorded.

17. Why This Trial Matters Statistically

This trial illustrates why a point estimate, a p-value, and a confidence interval answer different questions. The primary mean difference gives the estimated separation between group averages, the p-value describes compatibility with a null model under specified assumptions, and the interval displays the range of values compatible with the estimate under the calculation. Taken together, they show that the primary comparison is imprecise and does not resolve the direction of the difference.

It also demonstrates why endpoint definitions deserve close attention. The registry calls the outcome a reduction in abdominal fat but defines the recorded change as pre minus post and measures waist circumference. Reading the sign without checking the subtraction order could reverse the interpretation. Distinguishing a proxy measure from a direct measure is equally important: waist circumference is used as an indicator of abdominal adiposity, but is not itself a direct measure of fat tissue.

The outcome-arm counts are limited, while the standard deviations indicate notable variation in individual changes. General statistical principle: variation and limited information constrain precision. The broad primary interval makes this visible without implying that the point estimate is meaningless or that a particular alternative value is true.

The registry also illustrates the gap between posted summaries and a full statistical report. No sponsor analyses, analysis plan, multiplicity approach, or missing-data strategy are listed. Simple recalculations can make the arm summaries easier to interpret, but they cannot recover unreported modeling decisions. Likewise, planning illustrations based on observed effects do not reveal whether the original trial was designed around a justified target effect. These distinctions help prevent descriptive registry results from being read as more definitive than the record supports.

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.

Several trial-specific gaps limit interpretation beyond the uncertainty visible in the primary estimate. The registry does not describe the oral-contraceptive regimen, even though the study title and summary frame the question as metformin added to oral-contraceptive therapy. It also lists a lifestyle management program without explaining its delivery, allocation, or role in the comparison. Without those details, the treatment contrast and the contribution of co-interventions cannot be fully reconstructed.

The record gives an enrollment of 36 and outcome counts of 16 in each arm but does not explain the relationship between them. It does not report a participant flow, reasons for unavailable outcomes, adherence, withdrawals, or an analysis population. The extent and possible impact of missing data therefore cannot be assessed. Nor does the record say whether the outcome summaries include all randomized participants.

Statistical reporting is limited. No sponsor-reported analyses are listed, and the original analysis plan, adjustment strategy, handling of repeated measurements, and multiplicity approach are not reported in the registry record. The posted comparisons are unadjusted recalculations, so they may not reflect the investigators’ intended model. The registry also does not give a clinically meaningful difference threshold for the primary endpoint.

Outcome interpretation has additional limits. Waist circumference is a proxy for abdominal adiposity, and the recorded change is pre minus post, a direction that can be easy to misread. The record does not detail measurement procedures or provide baseline and post-intervention values separately. Broader eligibility and participant-characteristic details are also not reported, restricting assessment of how well the results may apply beyond the described adolescent population with PCOS.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Impact of Metformin in Teens With Polycystic Ovary Syndrome (PCOS) on Oral Contraceptive Therapy — NCT00283816.
• Registry record last updated 2015-05-15; status last verified 2011-04.
• Results first posted 2011-07-13.
• 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 describes a completed phase 3, randomized, double-masked, parallel-group trial of metformin versus placebo in adolescents with PCOS. Its stated question concerns oral-contraceptive therapy with or without metformin. The primary outcome is waist-circumference change from baseline to 24 weeks, with the change defined as pre minus post. The outcome summaries list 16 participants in each arm.

The recalculated unadjusted primary mean difference is -2.5 cm, with a confidence interval from -9.303 to 4.303 cm and an unadjusted p-value of 0.471. The interval includes zero. Three secondary outcomes are posted, but no sponsor-reported analyses are listed. The record does not report the trial’s original analysis plan, multiplicity handling, missing-data procedures, or analysis population. Safety reporting records zero serious adverse events in each arm among 18 participants at risk per arm; deaths are not reported in the registry record.