Antimullerian Hormone Versus Antral Follicle Count for Determination of Gonadotrophin Dosing in IVF: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Antimullerian Hormone Versus Antral Follicle Count for Determination of Gonadotrophin Dosing in IVF (NCT02739269), a randomized na study. The primary outcome is Percentage of Subjects Having Desired Ovarian Response.

2. Clinical Question

The registry frames the study as a comparison of two approaches to selecting gonadotrophin dose in IVF. Its stated hypothesis was that AMH-based dosing would result in a more optimal ovarian response than AFC-based dosing. The posted primary outcome operationalizes that response using the number of oocytes retrieved, rather than a broader measure of IVF success.

This endpoint compares whether participants fell within a specified range of retrieved oocytes. It does not, by itself, describe other outcomes such as pregnancy or live birth; those results are not reported in the registry record. The clinical question is therefore specifically about the measured ovarian response under the two dosing approaches, not a general comparison of all aspects of IVF treatment.

3. Trial Design

The registry describes a prospective, randomized, parallel-group trial comparing serum AMH measurement with AFC measurement as the basis for gonadotrophin dosing in IVF. The enrollment was 200. The registry lists the trial status as completed, with a start date of 2016-04-07 and a primary completion date of 2018-04-20.

The AMH group had 100.0 participants and the AFC group had 100.0 participants. The allocation is listed as randomized, and masking is listed as quadruple. The registry does not identify which people or roles were masked, so the precise implementation of masking is not reported in the registry record. The phase is listed as NA.

The posted primary outcome was assessed at one time point, at oocyte retrieval. The analysis population was intention-to-treat. Participants who did not proceed with ovarian stimulation or oocyte retrieval were counted as not having the desired response. This rule makes the outcome incorporate failure to reach retrieval as well as the number of oocytes among those who did reach retrieval.

The registry also lists one secondary outcome: the percentage of participants requiring a gonadotrophin dose step-up or step-down upon first ultrasound tracking. Its timeframe is the 8th day of ovarian stimulation. No further detail about dose adjustment criteria or the process for determining the initial dose is reported in the registry record. The record does not describe the original sample-size justification, trial protocol assumptions, interim analyses, or additional design features beyond those stated above.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intention-to-treat basis. Those not proceeding with ovarian stimulation or oocyte retrieval were regarded as not having desired ovarian response

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 Percentage of Subjects Having Desired Ovarian Response, measured over One single time point, i.e. at the time of oocyte retrieval.

The primary endpoint is the percentage of subjects having the desired ovarian response. The registry defines that response as having a number of oocytes retrieved between 6 and14, measured at the time of oocyte retrieval. Results are recorded as participant counts in the AMH and AFC groups.

This endpoint is used to compare how often the two dosing approaches were associated with retrieval in the prespecified range. A range-based binary outcome treats participants within that range as having the desired response and those outside it as not having it. The registry further specifies that participants who did not proceed with ovarian stimulation or oocyte retrieval were regarded as not having the desired response. That intention-to-treat rule is important when interpreting the result: the outcome is not restricted to participants who reached retrieval.

The endpoint is a measure of ovarian response, not a direct measure of every outcome relevant to IVF. It does not report the number of pregnancies, live births, or other downstream outcomes; these are not reported in the registry record. Nor does the binary result show how far an individual participant’s retrieval count was from the specified range. The result answers the narrower question of whether the participant met the recorded response definition.

7. Results Posted in the Registry

• AFC Group: 54.0% (54/100) versus 49.0% (49/100) on control; risk difference 5.0 percentage points (95% CI -8.8 to 18.8), risk ratio 1.10 (0.84–1.44); unadjusted two-proportion z-test p = 0.479.

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.

Percentage of Subjects Requiring Step-up or Step-down of Gonadotrophin Dose Upon First Ultrasound Tracking

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • AFC Group: 36.0% (36/100) versus 44.0% (44/100) on control; risk difference -8.0 percentage points (95% CI -21.5 to 5.5), risk ratio 0.82 (0.58–1.15); unadjusted two-proportion z-test p = 0.248.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between AFC Group and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary outcome, concerning whether participants required a gonadotrophin dose step-up or step-down upon first ultrasound tracking. Its reported analysis and the simple unadjusted calculation from the posted summary counts lead to different p-values. That discrepancy in the numerical results makes the secondary finding less straightforward to interpret than the primary comparison.

The secondary outcome addresses dose adjustment during stimulation, whereas the primary outcome concerns whether the retrieved oocyte count met the specified response range. They are related to the dosing strategies but measure different events at different points in treatment. A difference in dose adjustment does not, by itself, establish a difference in the primary ovarian-response outcome or in outcomes not reported in the registry record.

The secondary result should carry less weight than the primary result in an overall interpretation. Secondary outcomes may be numerous, and testing several outcomes increases the chance of seeing an apparently notable result by chance. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting secondary outcomes. The reported secondary p-value should therefore not be treated as if it were a standalone confirmatory finding.

The overall pattern is that the primary endpoint does not show a clear difference, while the one posted secondary endpoint has an unresolved gap between the registry’s reported test and the simple unadjusted recalculation. Neither finding establishes a general advantage of one dosing basis. Further details about secondary-outcome analysis and multiplicity are not reported in the registry record.

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

• Percentage of Subjects Requiring Step-up or Step-down of Gonadotrophin Dose Upon First Ultrasound Tracking

9. Statistical Interpretation

10. Statistical Methods Explained

The primary comparison is between two groups in a randomized, parallel design. The outcome is binary: each participant either met the registry’s definition of desired ovarian response or did not. The posted group values are counts of participants with the event, alongside the number assigned to each group. The comparison therefore concerns the proportion with the event, rather than a mean number of oocytes.

What does the risk difference represent?

The risk difference is the event proportion in the AFC group minus the event proportion in the AMH group. Its estimate is 0.05, so the observed contrast is in the direction of a higher response proportion in the AFC group. A risk difference is an absolute measure: it expresses the separation between proportions on the probability scale. It is not a relative increase, and it does not say how the result would change in a different population.

The measure is useful because it keeps the comparison close to the outcome as recorded: the difference in the proportion meeting the response definition. The estimate is only one part of the result. Its uncertainty interval and the statistical test help show how much information the observed counts provide about the underlying group contrast.

How should the p-value be read?

The registry reports a chi-squared test for the primary outcome, with p = 0.479. In general, a p-value measures how compatible the observed data are with a specified null model, assuming that model and the test’s conditions. It is not the probability that the null hypothesis is true, and it is not a measure of clinical importance. A value such as this does not establish equality; it indicates that the data do not provide strong evidence against the no-difference model by this test.

The registry does not report additional details about the test implementation, such as whether a continuity correction was used. The posted comparison is unadjusted and does not list covariates in the estimate. Further statistical analysis details are not reported in the registry record.

Why does the interval matter?

A confidence interval describes a range of effect estimates compatible with the observed data under the method used. In general, intervals that include effects on both sides of no difference indicate uncertainty about the direction of the underlying contrast. Here, the risk-difference interval includes negative and positive values. Its span is consistent with the lack of a clear difference in the primary comparison, while also showing that the estimate is not precise enough to rule out meaningfully different contrasts.

The registry record does not state the confidence level attached to the posted interval. For that reason, the bounds should be read as the interval reported with the recalculated comparison, without assigning a confidence level not given in the record. The p-value and interval address related but distinct aspects of uncertainty; neither should be used alone to describe the entire result.

What does intention-to-treat mean for this endpoint?

The primary outcome is specified on an intention-to-treat basis. The registry explicitly states that participants not proceeding with ovarian stimulation or oocyte retrieval were regarded as not having the desired response. This rule preserves those participants in the outcome definition rather than limiting analysis to those who reached retrieval. In general, intention-to-treat analyses aim to retain participants according to their randomized group, although the record does not provide a detailed account of all analysis procedures or missing-data handling beyond the stated rule.

The endpoint combines two circumstances in one binary outcome: whether retrieval occurred and whether the retrieval count fell within the designated range. This makes the recorded result relevant to the overall probability of achieving the defined response under the assigned strategy, but it does not separate the contribution of not proceeding from the count among those who did proceed.

What do the recalculated values add?

The recalculated primary risk difference and p-value are simple, unadjusted calculations from the posted summary counts. They offer a direct comparison on the absolute-proportion scale, but they do not replace an analysis using participant-level data or any adjustment specified in a full statistical analysis plan. For the primary outcome, the recalculated p-value agrees with the registry’s reported p-value. The record does not provide a participant-level dataset or additional adjusted result.

For the secondary outcome, the simple unadjusted recalculation does not match the registry’s reported p-value. The record does not explain the difference. A calculation from summary counts cannot determine whether the difference reflects a distinct analysis choice, a reporting issue, or another factor. The registry’s stated test and p-value should therefore be presented as reported, alongside the recalculation rather than silently substituted for it.

General statistical interpretation also distinguishes statistical significance from effect size and precision. A test result alone does not describe whether a contrast is important, and a nonsignificant result is not proof of equivalence. In this trial, the point estimate, interval, outcome definition, and analysis population all contribute to interpretation; the available summary statistics do not support conclusions beyond the recorded comparison.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.05 for AFC compared with AMH. Its reported interval extends from a negative value to a positive value, crossing the value that represents no group difference. This means the posted results are compatible with more than one direction of effect. The exact interval bounds have more than three decimal places in the registry record and are not reproduced.

The interval is broad relative to the point estimate, which signals that the observed counts do not pin down the size or direction of the underlying contrast closely. The p-value for the primary comparison is 0.479, consistent with the interval’s inclusion of no difference. This is not evidence that the groups are identical; a result that does not distinguish the groups may reflect limited information as well as a small underlying contrast.

The registry does not state the confidence level for the interval. That level is not reported in the registry record, so the interval should not be labeled with a confidence percentage. The interval’s bounds are also not a range in which an individual participant’s response is expected to fall; they describe uncertainty about the group-level risk difference under the calculation used.

Interpretation should emphasize both the direction of the estimate and the uncertainty around it. The observed response count is higher in the AFC group, but the interval includes contrasts in the opposite direction. The available precision is therefore insufficient to treat the observed difference as a dependable estimate of a single underlying direction or magnitude.

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 entries are participants per arm, not total events. They illustrate how large a future two-arm study might need to be under specified assumptions, using the observed effect as a starting point. The assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted is the stated fraction of the observed effect.

The 80% power column gives the planned number of participants per arm for a design with an 80% chance of detecting the assumed effect under the planning model, if that effect is the true one and the assumptions hold. The 90% power column describes the corresponding planning target for a 90% chance. These are design illustrations, not observed power for the completed trial.

The entries consider planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller target effects require more participants per arm to detect under the same planning assumptions. The figures are therefore sensitive to the chosen effect fraction and should not be read as a sample-size recommendation without considering the intended future study and its assumptions.

The trial’s own original design assumptions, including any original power target or sample-size calculation, are not reported in the registry record. The planning figures are based on the posted effect and do not establish what the investigators used when designing this trial.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.490; treatment proportion p₁ = 0.540; average p̄ = 0.515.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7068 = 1.3853.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7059 = 0.5941.
4. n = (1.3853 + 0.5941)² / (0.0500)² = 1,567.2, rounded up to 1,568 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
p0 <- 0.49                          # control proportion
p1 <- 0.54                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(2739269)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 1,568 (80% power), 2,098 (90% power)
 75% of observed effect: n per arm = 2,789 (80% power), 3,733 (90% power)
 50% of observed effect: n per arm = 6,278 (80% power), 8,404 (90% power)
power.prop.test: 1567.2 
Simulated power at n = 1,568 per arm: 0.805 (20000 replicates)

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

14. Analyses Reported by the Sponsor

For the primary outcome, the registry reports a chi-squared p-value of 0.479. The simple unadjusted recalculation from the posted summary counts also gives p = 0.479, and the recalculated risk difference is 0.05 for AFC compared with AMH. The matching p-values indicate that the basic comparison from the posted counts is consistent with the registry’s reported primary test at the displayed precision.

The registry’s primary result is the principal reported analysis. The recalculated estimate adds an absolute risk-difference measure, while the registry’s posted analysis supplies the named test and p-value. Neither provides evidence of a clear difference on the primary endpoint, and neither demonstrates that the approaches are equivalent.

For the secondary outcome, the registry reports a chi-squared p-value of 0.002, whereas a simple unadjusted recalculation from the posted counts gives p = 0.248. The record does not explain this difference. The registry does not provide enough detail to determine why these analyses differ, so both should be distinguished rather than treating the recalculation as a replacement for the registry’s reported result. The primary result is not affected by this secondary-outcome difference.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary outcome and one secondary outcome. It does not report whether the investigators adjusted for testing more than one outcome, whether a hierarchy of outcomes was prespecified, or whether any other comparisons were included in a formal analysis plan. Multiplicity handling is therefore not reported in the registry record.

In general, when several statistical tests are conducted, the chance of at least one apparently notable result can increase even if there is no true effect on any outcome. Adjustment methods or a prespecified testing order can address that issue, but the record does not identify such a method. The secondary p-value should be interpreted with that limitation in mind and should not be regarded as having the same confirmatory status as the primary comparison.

The primary outcome population is described as intention-to-treat, with participants who did not proceed with ovarian stimulation or oocyte retrieval regarded as not having the desired response. This gives a specific rule for those circumstances. However, the registry does not detail other missing-data procedures, the number of participants with missing measurements, or the handling of any other deviations from the planned outcome assessment. Those details are not reported in the registry record.

The record does not provide a full statistical analysis plan, participant-level analysis details, or information about interim analyses. The original sample-size justification and design assumptions are also not reported. These omissions limit assessment of whether the posted analyses followed a prespecified plan and how much flexibility may have existed in selecting analyses or outcomes.

These reporting gaps do not change the posted primary counts or its reported test, but they affect how confidently the full analysis process can be evaluated. The clearest reading is to distinguish the specified primary outcome from the secondary result, report the latter cautiously, and avoid assuming multiplicity or missing-data procedures that are not described.

16. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

The registry reports no serious adverse events among the 100 participants at risk in the AMH group and none among the 100 participants at risk in the AFC group. It also reports no other adverse events in either group and deaths of 0 in each group.

These are the safety counts recorded in the registry, not evidence that adverse events could not occur. The record does not provide further detail about event definitions, surveillance procedures, severity categories beyond the listed serious and other adverse-event groupings, or follow-up completeness. The absence of posted events should therefore be read as a description of the recorded data, not as a guarantee about safety.

The safety counts do not determine the primary efficacy comparison. They are separate observations and should not be used to infer whether either dosing approach has a benefit or whether it is appropriate for an individual. Additional safety information is not reported in the registry record.

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

The result describes how often participants in each randomized group met the registry’s definition of desired ovarian response: a retrieval count between 6 and14, with participants who did not proceed to stimulation or retrieval counted as not having that response. The posted counts are 49.0 in the AMH group and 54.0 in the AFC group. The estimated risk difference is 0.05 in the AFC-minus-AMH direction, with p = 0.479 for the registry’s reported primary analysis.

The data do not establish a clear difference between the groups on this endpoint. The interval for the risk difference includes both negative and positive values, and the test does not provide strong evidence against a no-difference model. At the same time, this is not proof of equivalence or proof that the approaches have identical effects. The uncertainty leaves a range of possible group contrasts compatible with the posted results.

The result is specific to the defined ovarian-response outcome and the trial as recorded. It does not establish the relative effects of the strategies on pregnancy, live birth, or other outcomes that are not reported in the registry record. It also does not reveal the experience of an individual participant, since a group-level proportion cannot predict an individual outcome.

The registry’s stated hypothesis favored AMH-based dosing, but the primary result does not confirm that hypothesis. Nor does the observed numerical direction in favor of AFC establish that AFC-based dosing is better. The study comparison should be understood as uncertain for the primary endpoint, rather than as a definitive finding for either strategy.

These results are statistical evidence about the posted trial outcome, not medical advice. A treatment decision requires considerations outside the scope of this registry comparison. The record does not supply all details needed to assess broader clinical outcomes or the full analysis plan.

18. Why This Trial Matters Statistically

This trial is statistically informative because it compares two ways of basing gonadotrophin dosing within a randomized parallel-group design and defines a concrete binary outcome for ovarian response. The primary outcome is linked to oocyte retrieval and includes a stated rule for participants who did not proceed to retrieval. That makes the analysis population and event definition central to interpreting the comparison.

The result also illustrates why an observed difference and evidence for a difference are not interchangeable. The AFC group has a higher posted response count, and the estimated risk difference is positive for AFC relative to AMH. Yet the p-value is 0.479 and the uncertainty interval crosses no difference. The point estimate gives the direction and scale of the observed contrast; uncertainty measures show that the data do not tightly determine the underlying contrast.

The trial also offers a useful contrast between primary and secondary evidence. The primary reported p-value agrees with a simple calculation from the posted counts. The secondary reported p-value does not agree with the simple unadjusted recalculation, and the registry does not explain why. This highlights the value of transparent analysis details, especially when more than one endpoint is considered.

Finally, planning calculations based on an observed effect are illustrations, not evidence that the completed trial had a particular probability of finding a result. The registry does not report its original sample-size justification. Separating planning scenarios from observed trial results prevents an estimated design requirement from being mistaken for an analysis of the study’s achieved power.

19. Limitations

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

The registry record provides summary counts and selected analyses rather than participant-level data. It does not report the confidence level attached to the primary interval, the full statistical analysis plan, covariate-adjusted estimates, or the detailed implementation of the chi-squared test. These omissions limit independent assessment of the analysis beyond the simple unadjusted calculation from the posted counts.

The primary outcome is a binary measure based on whether the retrieval count fell within the specified range. That summarizes whether the response met the definition but does not describe the distribution of retrieval counts or how far counts outside the range were from it. The outcome also combines failure to proceed to stimulation or retrieval with counts that do not meet the response definition, according to the stated intention-to-treat rule.

The registry lists one secondary outcome, but its reported p-value differs from the simple unadjusted recalculation. No explanation for that difference is reported. The record also does not describe a multiplicity strategy, detailed missing-data procedures beyond the primary outcome rule, or interim analyses. The original sample-size justification and design assumptions are not reported.

The registry’s masking field says quadruple, but the masked roles and practical implementation are not described. The record does not post outcomes such as pregnancy or live birth, so the primary result cannot be extended to those outcomes. These limitations concern what can be inferred from the registry record and do not establish that particular procedures were absent; further details are not reported in the registry record.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Antimullerian Hormone Versus Antral Follicle Count for Determination of Gonadotrophin Dosing in IVF — NCT02739269.
• Registry record last updated 2019-02-27; status last verified 2019-02.
• Results first posted 2019-02-27.
• 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

NCT02739269 is a completed prospective randomized parallel-group trial comparing serum AMH measurement with AFC measurement as the basis for gonadotrophin dosing in IVF. The registry lists quadruple masking and enrollment of 200.

The primary outcome was the proportion of participants meeting the desired ovarian-response definition, based on retrieval of 6 to 14 oocytes and assessed at oocyte retrieval. The registry reports 49.0 participants with the event in the AMH group and 54.0 in the AFC group. The risk difference for AFC compared with AMH was 0.05, and the registry-reported primary p-value was 0.479.

The posted primary result does not establish a clear group difference. The registry includes one secondary dose-adjustment outcome, for which the reported p-value differs from the simple unadjusted recalculation. Further details about multiplicity, the original sample-size rationale, and several analysis procedures are not reported in the registry record.