The Effect of Hormonal Add-Back Therapy in Adolescents Treated With a GnRH Agonist for Endometriosis: A Randomized Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Effect of Hormonal Add-Back Therapy in Adolescents Treated With a GnRH Agonist for Endometriosis: A Randomized Trial (NCT00474851), a randomized phase 2 study. The primary outcome is Bone Mineral Density.

2. Clinical Question

The comparison concerns the incremental contribution of the estrogen-containing regimen when both groups receive norethindrone acetate. It is not a comparison of add-back therapy with no add-back therapy. The registry identifies bone mineral density as the primary endpoint and total body bone mineral content as a secondary outcome. The primary endpoint is continuous: it records an amount of bone mineral density and its change over time, rather than classifying participants into categories. The posted arm summaries report mean changes, so interpretation centers on the difference between group averages and the uncertainty around that difference. The registry record does not report a clinical threshold for a meaningful difference, nor does it report a quality-of-life outcome result.

3. Trial Design

This is a completed phase 2, randomized, parallel-group trial sponsored by Boston Children's Hospital. The registry lists endometriosis as the condition and describes two drug regimens. Participants were assigned to norethindrone acetate plus conjugated estrogens or norethindrone acetate plus placebo. Both regimens were taken for the 12 months of study participation. The registry describes the study as quadruple masked, but the registry record does not report which roles were masked or how masking was maintained.

The registry lists 53 participants enrolled. For the primary endpoint, the posted summaries include 25 participants in the intervention group and 26 in the placebo group. These endpoint counts do not explain what happened to every enrolled participant; reasons for any difference between enrollment and outcome summaries are not reported in the registry record. The registry identifies a baseline-to-12-month timeframe for the primary outcome, an adjusted mean change in total body areal bone mineral density. Total body bone mineral content is also listed as a secondary outcome over the same timeframe.

The trial began in 2007-08, reached primary completion in 2013-10, and is recorded as completed. It is randomized and uses a parallel-group structure, so the posted comparison is between groups assigned to different regimens rather than a within-person crossover. The registry record does not report the original sample-size justification, the trial’s planned statistical model in sufficient detail, or whether interim analyses were planned. Those omissions limit how fully the reported results can be connected to the trial’s original design and analysis decisions. The registry identifies a phase 2 study; that designation alone does not establish the precision, clinical importance, or generalizability of the reported effect.

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 Bone Mineral Density, measured over Baseline to 12 months.

The primary endpoint is adjusted mean change in total body areal bone mineral density (aBMD), measured in g/cm^2, from baseline to 12 months. Bone mineral density is a continuous measure of mineral content relative to projected bone area. Reporting change over time makes the outcome directly relevant to the trial’s question about maintaining skeletal health during treatment. The endpoint summarizes the average change within each randomized group, while the between-group contrast describes how those average changes differ.

A continuous endpoint retains information about the magnitude of measured change rather than reducing results to a yes-or-no outcome. The posted summaries report an intervention-group mean change of 0.01 and a placebo-group mean change of -0.11, each with a standard error of 0.006. These are group-level summaries, not descriptions of every participant’s experience. The registry record does not report the instrument or measurement procedures used, a threshold for a clinically important change, or individual participant trajectories.

Bone mineral density is the named primary endpoint, while total body bone mineral content is listed as a secondary measure. The registry’s stated purpose also mentions quality of life, but a quality-of-life endpoint result is not reported in the registry record. Consequently, the posted primary result addresses a bone-density measure; it cannot by itself establish what happened to quality of life or other aspects of the study’s stated aims.

7. Results Posted in the Registry

• Intervention Group: mean 0.01 versus -0.11 on control; difference 0.12 (95% CI 0.10 to 0.14), standardized effect d = 3.96; z-test p = <0.001.

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.

Total Body Bone Mineral Content (BMC)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Intervention Group: mean 37.00 versus 15.00 on control; difference 22.00 (95% CI -7.14 to 51.14), standardized effect d = 0.41; z-test p = 0.139.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Intervention Group and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcome results add a different statistical picture from the primary endpoint, but they do not independently settle the primary question. The sponsor’s reported secondary analysis and the simple recalculation do not yield the same statistical signal. This is a reminder that the method and analysis specification matter, especially when the registry does not provide enough detail to reproduce the sponsor’s model.

Secondary outcomes can help describe whether the overall pattern is coherent across related measures, but they are generally less conclusive than a prespecified primary endpoint. The registry lists total body bone mineral content as a secondary outcome and does not report a multiplicity adjustment or a testing hierarchy. When multiple outcomes are examined, interpreting each unadjusted p-value as confirmatory can overstate the strength of evidence. A secondary result may be useful for generating questions or informing future research, but it should not be treated as an independent replication of the primary result.

The primary and secondary bone measures are related aspects of skeletal health, but they are not interchangeable outcomes. Agreement or disagreement between them would not, by itself, establish the broader clinical meaning of the intervention. The registry does not report a quality-of-life result, despite quality of life being included in the study’s stated purpose. Nor does it provide information that would clarify whether the outcomes were analyzed in a prespecified order or how missing values were handled.

Accordingly, the secondary analysis is best read as supportive context rather than decisive evidence. Its weight is limited by its secondary status, the absence of a reported multiplicity plan, and the divergence between the sponsor’s analysis and the simple recalculation. These considerations apply in addition to the uncertainty about clinical importance and the lack of a reported threshold for meaningful change.

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

• Total Body Bone Mineral Content (BMC)

9. Statistical Interpretation

10. Statistical Methods Explained

The posted primary outcome is a continuous measure summarized as mean change. A mean represents the average observed change within a group. The between-group mean difference compares those averages; here, the recalculated difference is 0.12, with the intervention-group average change higher than the placebo-group average change. The sign and size of a difference depend on the order of subtraction, so the important interpretation is which group average is higher and how uncertain the contrast is. The summaries are adjusted mean changes, while the recalculation is described as an unadjusted calculation from posted summary statistics. Those labels distinguish the registry’s reported summaries from the assumptions used in the simple comparison.

What does the standard error describe?

The primary group summaries report standard errors of 0.006. A standard error describes the precision of an estimated group mean: it reflects how much the estimate would vary across repeated samples under the statistical model. It is not the same as the spread of individual participant measurements. The registry reports standard errors rather than standard deviations for these endpoint summaries. The recalculation uses the posted summary information and does not have access to individual participant records.

What does the confidence interval add?

A confidence interval gives a range of values compatible with the data and the method used, under the model’s assumptions. It helps readers assess precision as well as whether the interval includes a null difference of zero. For the primary recalculation, the interval excludes zero. That is a statement about the recalculated comparison, not the sponsor’s RMANOVA result. The registry record does not report a confidence interval for the sponsor’s analysis. An interval does not show the probability that a particular effect is true, and it does not address whether the effect is clinically important.

Why are there two reported p-values for an endpoint?

The sponsor’s primary analysis is identified as RMANOVA and has a reported p-value of 0.10. Separately, the recalculated unadjusted comparison has a p-value of <0.001. These values arise from distinct analyses. RMANOVA is a repeated-measures analysis of variance approach; in general, such methods assess patterns across repeated observations while accounting for their relationship. The registry does not report enough model details, estimates, or assumptions to determine exactly how the sponsor’s model was specified or why its result differs from the simple recalculation. The recalculated analysis uses posted summary statistics and should not be substituted for the sponsor’s own model.

A p-value measures how incompatible data are with a specified null hypothesis under the analysis assumptions. It is not the probability that the null hypothesis is true, nor does it measure the probability that a result will replicate. A smaller p-value is not, by itself, evidence that an effect is larger or more clinically consequential. Here, the difference between the sponsor’s p-value and the recalculated p-value is a reason to keep the analyses distinct rather than select one without understanding its model and role in the trial plan.

What can be learned from the secondary analysis?

The secondary bone mineral content outcome is also continuous and summarized as mean change. The recalculated mean difference is 22.0 g, and its interval extends from -7.137 to 51.14; the recalculated p-value is 0.139. The sponsor’s RMANOVA p-value for that outcome is 0.02. As with the primary result, these are results from different analytical approaches. The interval in the simple recalculation includes zero, so it is compatible with no between-group difference under that calculation, as well as with differences in either direction. The sponsor’s p-value alone does not provide an effect estimate or interval in the registry record.

Secondary outcome analyses are often considered exploratory, particularly when several outcomes or comparisons are examined. The registry lists two outcome measures and two analyses, but it does not report a multiplicity adjustment or a hierarchy of testing. General statistical reasoning is that examining multiple outcomes increases the chance of observing a seemingly notable result by chance if each is interpreted in isolation. Without a reported multiplicity plan, secondary p-values merit less confirmatory weight than a result supported by a prespecified primary analysis.

The reported analyses do not make individual participant-level data available. The record does not report the exact model specification, covariates, handling of missing measurements, analysis population, or diagnostics. Such details matter because analysis choices can affect estimates and uncertainty. The summaries allow a limited comparison of group-average changes and a simple recalculation, but they do not permit a complete reproduction of the sponsor’s analysis. The sponsor’s own result should be understood in the context of its model, while the recalculation should be understood as a separate, unadjusted reading of the posted summary values.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate an estimated difference. For the primary endpoint, the recalculated mean difference is 0.12. Its recalculated interval excludes zero, indicating that the range from this simple analysis does not include no difference between group means. The interval is based on the posted summary statistics, not on the sponsor’s RMANOVA model. The registry record does not report a confidence interval for the sponsor’s analysis, so the recalculated interval should not be mistaken for the sponsor’s model-based uncertainty estimate.

The primary endpoint’s standard errors are 0.006 in each group. Standard errors describe uncertainty in the estimated means, not the variability among individual participants. They therefore do not show whether participants had similar changes or whether some experienced changes far from the group average. Individual-level distributions are not reported in the registry record.

For the secondary bone mineral content outcome, the recalculated interval ranges from -7.137 to 51.14 and includes zero. That interval is compatible with no difference under the simple calculation and also with differences in either direction. The breadth of an interval is relevant to interpretation: it indicates the range of effect sizes that remain compatible with the data and method. The sponsor’s reported RMANOVA p-value for this secondary outcome is 0.02, but the sponsor’s effect estimate and confidence interval are not reported in the registry record.

Intervals do not resolve questions of clinical importance. The registry record does not report a threshold for a clinically meaningful difference in either bone measure. Nor do these intervals account for uncertainty about missing outcomes, model selection, or multiplicity unless those features are incorporated in the underlying analysis. The primary recalculation is statistically precise under its assumptions, but the gap between it and the sponsor’s reported p-value makes the distinction between analyses especially important. Precision should be considered alongside the analysis method, endpoint hierarchy, and incomplete reporting of the original statistical plan.

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 planning figures are participants per arm, not total events. They illustrate the sample size that a hypothetical two-arm study would require under the stated calculation assumptions. They are not the original design target for this trial: the registry record does not report the trial’s own sample-size assumptions or justification.

The illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect. The 80% power column gives the number of participants per arm for a design with an 80% chance of detecting the assumed effect at the specified significance level, if the assumptions hold. The 90% power column uses a higher chance of detection and therefore generally requires more participants per arm. For both columns, the figures are planning illustrations computed from the observed effect, not observed power.

The entries change with the assumed effect size. Planning for 100% of the observed effect gives 2 participants per arm for 80% power and 2 for 90% power. Planning for 75% of the observed effect gives 2 participants per arm for 80% power and 3 for 90% power. Planning for 50% of the observed effect gives 5 participants per arm for 80% power and 6 for 90% power. The smaller the effect a study aims to detect, the more participants it generally needs, because a subtler difference is harder to distinguish from random variation.

These figures should not be read as evidence that the actual trial was adequately powered, nor as a reconstruction of its design. They depend on the observed effect and the calculation assumptions; a planning exercise based on an observed estimate can be optimistic or otherwise unrepresentative of a future study. The registry record does not report the original sample-size justification, and these planning figures do not replace it.

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 = 0.120; pooled standard deviation σ = 0.030.
2. n = 2 × (1.96 + 0.842)² × 0.030² / 0.120² = 1.0, rounded up to 2 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 <- 0.12                      # observed difference in means
sigma <- 0.0303045763365663                     # 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(474851)
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 = 2 (80% power), 2 (90% power)
 75% of observed effect: n per arm = 2 (80% power), 3 (90% power)
 50% of observed effect: n per arm = 5 (80% power), 6 (90% power)
power.t.test: 2.4 
Simulated power at n = 2 per arm: 0.557 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s reported primary analysis is RMANOVA, with a p-value of 0.10. The simple recalculation from the posted arm summaries gives a mean difference of 0.12, an interval that excludes zero, and a p-value of <0.001. The two results should be reported as distinct analyses, not blended into a single conclusion. The registry record does not include the sponsor’s model-based effect estimate or confidence interval, and it does not describe the RMANOVA specification in enough detail to explain the difference fully.

The recalculation is an unadjusted comparison based on the posted summary statistics. It is not a re-creation of the sponsor’s repeated-measures model, and it may not account for the same information or assumptions. The arm summaries are described as adjusted mean changes, which makes the difference between the posted summaries and the unadjusted recalculation particularly important to state clearly. A recalculated p-value cannot supersede a reported analysis merely because it is smaller.

The registry record lists no discrepancy flag, but that does not resolve the methodological gap between the two analyses. Readers should retain the sponsor’s RMANOVA result as the reported sponsor analysis and treat the recalculation as a separate statistical comparison. The record does not report enough information to decide whether the difference reflects model specification, data handling, or another analytical choice.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures and two analyses: primary bone mineral density and secondary total body bone mineral content. It does not report a strategy for controlling multiplicity, such as an adjustment across outcomes or a hierarchy specifying which tests would be considered confirmatory. General statistical reasoning is that testing multiple outcomes can increase the chance of at least one apparently notable result arising by chance when each test is interpreted separately. Without a reported plan, the secondary p-value should therefore be read cautiously and should not be treated as confirmation of the primary result.

The registry identifies the primary endpoint, which helps distinguish it from the secondary outcome, but it does not report whether additional outcomes were measured or considered in the study’s full analysis plan. A quality-of-life result is not reported in the registry record, despite quality of life appearing in the study’s purpose. The record also does not report whether the primary analysis was specified before the data were examined or whether any interim analyses were planned. These details should not be inferred from the posted results.

Missing-data handling is not reported in the registry record. The registry lists 53 participants enrolled, while the primary outcome summaries include 25 participants in the intervention group and 26 in the placebo group. Those counts do not explain why the endpoint summaries include those numbers or whether measurements were unavailable, excluded, or handled by a model. The record does not describe the analysis population, imputation, assumptions about missingness, or sensitivity analyses. Without these details, it is not possible to assess how incomplete outcomes may have influenced the estimates.

The record also does not provide enough information about the RMANOVA specification to assess its assumptions or reproduce the sponsor’s analysis. In general, repeated-measures methods need to account appropriately for correlations among repeated observations and rely on assumptions about the outcome and its covariance structure. The exact approach used in this trial, any diagnostics, and any adjustments are not reported in the registry record. These reporting gaps limit independent evaluation; they do not establish that the analysis was inappropriate, but they prevent a full assessment of its robustness.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: depression (19.2%), hospitalization for pelvic/abdominal pain (12.0%), fracture (12.0%), rectal bleeding (4.0%).

The registry reports serious adverse event counts for the two groups. In the intervention group, 25 participants were at risk and 8 were affected; in the placebo group, 26 were at risk and 5 were affected. These are group-level counts in the posted record. They do not establish that the regimens caused the events, and the registry’s aggregate figures do not describe timing, duration, severity by event, or individual participant histories.

The listed serious event terms include Depression at 19.2%, Hospitalization for pelvic/abdominal pain at 12.0%, Fracture at 12.0%, and Rectal bleeding at 4.0%. The registry record does not make clear in these summaries how those percentages map to group-specific counts. Death counts are not reported in the registry record; a missing death value should not be interpreted as zero.

Other adverse events affected 15 of 25 participants in the intervention group and 13 of 26 participants in the placebo group. These counts are distinct from the serious adverse event counts. The registry record does not provide enough information to compare event types, exposure time, or causal relationships between groups. These figures describe reported events during the study and should not be used alone to infer comparative safety.

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

The registry describes a randomized comparison of two add-back regimens in adolescents treated with a GnRH agonist for endometriosis. The posted primary summaries show a higher average change in bone mineral density in the intervention group than in the placebo group. The simple recalculation yields a mean difference of 0.12 and an interval excluding zero, while the sponsor’s RMANOVA analysis reports a p-value of 0.10. These are not interchangeable results: one is a recalculation from posted summaries, and the other is the sponsor’s reported repeated-measures analysis.

These statistics do not establish whether the observed difference is clinically important. The registry record does not report a threshold for meaningful change, individual participant outcomes, or the sponsor’s model-based effect estimate and confidence interval. The results also do not show what happened to quality of life, even though it appears in the study’s stated purpose. The available endpoint summaries describe group averages, not the full distribution of outcomes or the experience of every participant.

The findings should not be generalized automatically to other populations, treatment durations, or clinical settings. The record describes adolescents with endometriosis treated with a GnRH agonist and the two regimens used during 12 months of study participation. It does not provide information about longer-term outcomes or other treatment comparisons. The registry record also does not report the original sample-size justification, missing-data methods, or all details of the sponsor’s statistical plan.

Thus, the posted data support a careful description of the measured group-average changes and the distinct statistical results, not a broad claim about treatment benefit or lack of benefit. The sponsor’s analysis, the simple recalculation, uncertainty, endpoint hierarchy, and reporting gaps all matter when interpreting what the trial can establish.

18. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation depends on more than a single p-value. The primary endpoint has a recalculated unadjusted comparison with an interval excluding zero, while the sponsor’s RMANOVA analysis reports a p-value of 0.10. Because the registry does not provide the sponsor’s full model specification or model-based estimate, readers cannot reproduce the reported analysis or resolve the difference from the simple recalculation. Keeping the analyses separate is essential to an accurate account.

The example also shows why effect estimates and uncertainty should accompany significance tests. A p-value alone does not convey the size of a difference, its precision, or its clinical relevance. The recalculated mean difference and interval provide information about the estimated contrast under the simple method, while the sponsor’s reported analysis supplies a different model-based test result. Neither removes the need for a clinical threshold or an explanation of the analysis plan.

Secondary outcomes add another statistical consideration. The sponsor’s secondary result and its simple recalculation differ, while the registry does not report a multiplicity strategy. This highlights how endpoint status and the number of analyses affect the weight assigned to results. Secondary findings can inform interpretation, but they should not be elevated to confirmatory evidence without a clear plan.

Finally, the record makes visible the importance of transparent reporting. Details about missing outcomes, analysis populations, interim analyses, and the original sample-size justification are not reported in the registry record. These omissions do not prove a flaw in the trial, but they limit independent assessment of how robustly the evidence answers the stated question. Statistical literacy means recognizing both what the reported estimates show and what the record leaves unresolved.

19. Limitations

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

Several limitations specific to this trial follow from the information reported in the registry. The primary result has a notable difference between the sponsor’s RMANOVA p-value of 0.10 and the recalculated p-value of <0.001. The registry does not provide enough detail about the sponsor’s model, model-based estimate, or confidence interval to explain that difference. The simple recalculation is not a substitute for the sponsor’s analysis and relies on the posted summaries rather than participant-level data.

The registry reports 53 participants enrolled, while the primary endpoint summaries include 25 participants in the intervention group and 26 in the placebo group. The reasons those numbers differ are not reported in the registry record. Missing-data handling, analysis population definitions, and sensitivity analyses are also not reported, limiting evaluation of potential effects of unavailable or excluded observations. The record does not report the original sample-size justification, so the trial’s intended precision and design assumptions cannot be assessed from the posted planning illustration.

The registry does not report a clinical threshold for meaningful changes in bone mineral density or bone mineral content. It also does not provide the quality-of-life results mentioned in the study purpose, or longer-term outcome information. A multiplicity strategy for the secondary outcome is not reported, so its p-value should be treated as less confirmatory. The serious adverse event summaries provide counts and selected terms, but do not report enough detail to establish causal attribution, event timing, or the relationship between terms and group-specific counts.

These are limitations of what can be concluded from the registry record, not evidence that the trial necessarily had deficiencies in its conduct. The study’s population, comparator, follow-up timeframe, and reporting scope define how far its posted results can be interpreted.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: The Effect of Hormonal Add-Back Therapy in Adolescents Treated With a GnRH Agonist for Endometriosis: A Randomized Trial — NCT00474851.
• Registry record last updated 2015-03-30; status last verified 2015-03.
• Results first posted 2015-03-30.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

The registry describes a completed randomized phase 2 trial of add-back therapy in adolescents treated with a GnRH agonist for endometriosis. It compared norethindrone acetate plus conjugated estrogens with norethindrone acetate plus placebo over 12 months. The primary outcome was adjusted mean change in total body areal bone mineral density from baseline to 12 months.

The posted primary group summaries include 25 participants in the intervention group and 26 in the placebo group. The sponsor reports an RMANOVA p-value of 0.10; a separate unadjusted recalculation from the posted summaries gives a mean difference of 0.12 and p < 0.001. The record does not report enough analytical detail to reconcile these results. Total body bone mineral content is listed as a secondary outcome, with differences between its sponsor-reported and recalculated results as well.

The registry record does not report the original sample-size justification, missing-data methods, a multiplicity plan, or a quality-of-life result. The posted statistics describe group averages and should be interpreted with these reporting limits in mind.