Transabdominal and Transvaginal Sonography in Medical Abortion: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Transabdominal and Transvaginal Sonography in Medical Abortion (NCT03047551), a randomized na study. The primary outcome is Percentage of Subjects in Each Ultrasound Group With Additional Testing.

2. Clinical Question

3. Trial Design

The registry record describes NCT03047551, titled “Transabdominal and Transvaginal Sonography in Medical Abortion.” Columbia University is listed as the sponsor. The study is marked completed and has an enrollment of 634. It is classified as a randomized, parallel-group trial with no masking. Its phase is listed as not applicable.

Participants were assigned to transabdominal sonography or transvaginal sonography for assessment before medical abortion. The study summary describes a multicenter trial in practices that routinely use TVS. It anticipated enrolling 800 patients receiving care from about 30 providers over 6-8 months. The registry record reports enrollment of 634; it does not explain the difference between the anticipated enrollment and the recorded enrollment or provide an account of the number of providers who participated.

The recorded start is 2016-07, and primary completion is recorded as 2018-06. These dates describe the trial record and do not establish when every participant was enrolled or followed. The primary outcome timeframe is 1 day. The endpoint description focuses on whether a provider ordered additional testing after the patient history and assigned ultrasound, but before eligibility determination.

The posted outcome counts are 63 participants in the TAS group, whose listed group size is 317, and 14 participants in the TVS group, whose listed group size is 312. The registry record does not explain why the group denominators for this outcome differ from the overall enrollment. It also does not report the allocation sequence method, concealment procedures, or the trial's original sample-size justification.

4. Participant Flow and Accounting

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

Before assignment: Subjects were excluded from the analysis due to study withdrawal prior to receiving sonography.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Subjects in Each Ultrasound Group With Additional Testing, measured over 1 day.

The primary endpoint counts participants in each ultrasound group for whom a provider ordered additional testing after sonography and before determining eligibility for medical abortion. In the registry description, testing follows either patient history plus TAS or patient history plus TVS. The outcome is therefore a participant count, not a measure of the number or type of tests ordered, the time spent on assessment, or whether a participant ultimately received an abortion.

This endpoint is used to compare how often the two assessment approaches were followed by further evaluation. A count-based endpoint can be interpreted alongside the number of participants in each group, as the posted results do. The comparison is relevant to the study's stated question about how often clinicians request additional evaluation after one of the two ultrasound approaches.

The primary outcome timeframe is 1 day. The record does not provide a more detailed operational definition of “additional testing,” list the possible tests, or say whether orders were independently adjudicated. It also does not report the reason for each order or identify whether the additional evaluation changed the eligibility determination.

The study summary names patient satisfaction, measured with a questionnaire using a visual analog scale, as a second primary outcome. However, the registry record posts no secondary outcome results. Consequently, the available numerical comparison addresses additional testing only; it does not quantify satisfaction.

7. Results Posted in the Registry

• Transvaginal Sonography: 4.5% (14/312) versus 19.9% (63/317) on control; risk difference -15.4 percentage points (95% CI -20.3 to -10.4), risk ratio 0.23 (0.13–0.39); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted comparison concerns a binary outcome: each participant either had additional testing recorded or did not. For a binary endpoint, group results can be summarized as counts and risks, then compared using measures such as a risk difference, risk ratio, or odds ratio. General explanation: a risk difference subtracts one group's risk from the other's; its sign depends on which group is placed first. In this record the contrast is TVS relative to TAS, so the negative estimate indicates a lower reported risk in the TVS group.

What does the risk difference describe?

The risk difference is -0.154, with a 95% confidence interval from -0.203 to -0.104. It expresses the contrast on the risk scale, rather than as a relative comparison. The interval gives a range of values compatible with the observed data under the method used to calculate it. Because the full interval is negative, it is consistent with a lower risk estimate for TVS than TAS. The estimate is unadjusted in the posted comparison; the registry does not report a model adjusting for participant or provider characteristics.

Why are risk ratios and odds ratios also reported?

The risk ratio is 0.226, with a 95% confidence interval from 0.129 to 0.394. It expresses the risk in one group relative to the other. The odds ratio is 0.189, with a 95% confidence interval from 0.104 to 0.346. Odds and risks are related but not interchangeable quantities, so these estimates should not be interpreted as though they were the same measure. They are alternative summaries of the group contrast, not separate trial outcomes.

General explanation: the risk difference is often useful when the question concerns the absolute gap between groups, while relative measures describe the contrast in proportional terms. The choice of measure affects how a result is communicated. No single measure gives information about the reasons for the outcome or establishes the practical value of the observed contrast.

What does the p-value tell the reader?

The posted unadjusted p-value is <0.001. General explanation: a p-value evaluates how compatible the data are with a specified null model, assuming that model and the analysis conditions. A small p-value is evidence against that null model, but it does not measure effect size, the probability that the result is due to chance, or the probability that a hypothesis is true. The magnitude and precision of the estimate are better considered through the effect estimate and its confidence interval.

The registry does not report the complete statistical analysis plan, including the test used for this p-value, whether any covariates were considered, or how the analysis handled all randomized participants. The displayed comparison is unadjusted. The registry record also reports no secondary outcome results, despite naming patient satisfaction as a second primary outcome in the study summary.

What assumptions affect binary-outcome comparisons?

General explanation: these comparisons rely on clear outcome definitions, valid group denominators, and appropriate handling of participants whose outcome status is unavailable. Random allocation supports comparison of assigned groups, but it does not by itself settle questions about missing outcomes, departures from assigned procedures, or the analysis population. If participants were clustered within providers or practices, dependence among outcomes could matter for uncertainty estimates; the registry record does not describe whether clustering was accounted for. This is a general statistical consideration, not a claim that clustering changed this result.

The posted comparison is not labeled as adjusted, and the record does not describe missing-data methods, sensitivity analyses, interim analyses, or multiplicity control. Those methods are therefore not reported in the registry record. The available estimates support description of the posted group contrast, while these reporting gaps limit assessment of how fully the analysis addressed possible sources of uncertainty.

10. Confidence Intervals and Precision

The confidence interval for the risk difference runs from -0.203 to -0.104 around an estimate of -0.154. A confidence interval conveys the precision of an estimate: narrower intervals indicate less sampling uncertainty under the method used, while wider intervals indicate more. It is not a range containing every participant's outcome, and it does not account automatically for possible bias, measurement problems, or limitations in the analysis.

For the relative measures, the risk ratio is 0.226, with a confidence interval from 0.129 to 0.394, and the odds ratio is 0.189, with a confidence interval from 0.104 to 0.346. These intervals describe uncertainty on their respective scales. Their interpretation depends on the chosen measure and the direction of comparison. The registry reports the comparison as TVS relative to TAS.

All three posted intervals exclude the respective null value: zero for the risk difference and one for the risk ratio and odds ratio. This is consistent with the reported unadjusted p-value of <0.001. General explanation: interval estimates and p-values often convey related information when they arise from compatible methods, but neither addresses every source of uncertainty in a trial.

The registry record does not identify the confidence-interval method or state whether the calculation accounted for provider or practice structure. It also does not report intervals for patient satisfaction, which has no posted result. Precision in the posted numerical contrast should therefore be distinguished from broader certainty about the study's conduct, outcome measurement, and relevance beyond the trial setting.

11. 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 event counts. They illustrate the participant numbers associated with detecting a stated fraction of the observed effect under the listed planning assumptions. For planning for 100% of the observed effect, the illustration gives 70 participants per arm for 80% power and 93 participants per arm for 90% power. For planning for 75% of the observed effect, the corresponding figures are 142 and 190 participants per arm. For planning for 50% of the observed effect, they are 356 and 476 participants per arm.

In plain language, the 80% power column gives the number of participants per arm associated with an 80% chance of detecting the specified effect under the assumptions, if that effect is present. The 90% power column gives the corresponding planning number for a 90% chance. These are planning illustrations calculated from the observed effect, not the trial's original sample-size justification. They are not observed power.

The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is the stated fraction of the observed effect. The registry record does not report the trial's own original design assumptions or sample-size justification. Smaller target effects require more participants under these planning assumptions, which is why the figures increase as the planned fraction of the observed effect decreases.

12. 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 = 0.199; treatment proportion p₁ = 0.045; average p̄ = 0.122.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4625 = 0.9065.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4496 = 0.3784.
4. n = (0.9065 + 0.3784)² / (-0.1539)² = 69.7, rounded up to 70 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 <- 0.198738170347003                          # control proportion
p1 <- 0.0448717948717949                          # 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(3047551)
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 = 70 (80% power), 93 (90% power)
 75% of observed effect: n per arm = 142 (80% power), 190 (90% power)
 50% of observed effect: n per arm = 356 (80% power), 476 (90% power)
power.prop.test: 69.7 
Simulated power at n = 70 per arm: 0.833 (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.

13. Multiplicity, Missing Data and Analysis Populations

The study summary identifies two primary outcomes: the proportion of participants requiring additional evaluation after sonography and patient satisfaction measured by a visual analog scale questionnaire. The registry record posts a result for additional testing, but it reports no secondary outcome results. It does not provide a posted numerical result for satisfaction.

When a study evaluates multiple outcomes or makes multiple comparisons, multiplicity can affect the interpretation of statistical evidence. General explanation: if several hypotheses are tested, the chance of at least one apparently unusual result can increase unless the analysis plan addresses that issue. The registry record does not report whether multiplicity adjustments were planned or used, how the two named primary outcomes were treated together, or whether any additional comparisons were prespecified.

The record reports zero analyses and zero posted secondary outcomes. It does not provide participant-level outcome availability, missing-data counts, or methods for handling missing outcome measurements. The overall enrollment is listed as 634, while the primary outcome group denominators are listed as 317 for TAS and 312 for TVS. The record does not explain the difference between these figures. It is therefore not possible to determine from the registry record whether this reflects unavailable outcomes, an analysis population choice, or another reason.

The registry record does not report whether the primary comparison followed an intention-to-treat approach, whether participants were analyzed according to the ultrasound group assigned, or whether a per-protocol or other population was used. It also does not describe exclusions after randomization, protocol deviations, interim analyses, or sensitivity analyses. These items should not be inferred from the group counts or the posted p-value. Their absence limits assessment of how missingness, analysis populations, and multiplicity were handled.

14. 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.

For serious adverse events, the registry lists 0 affected participants among 317 at risk in the TAS group and 0 among 312 at risk in the TVS group. For other adverse events, it likewise lists 0 affected participants among 317 at risk in the TAS group and 0 among 312 at risk in the TVS group. Deaths are listed as 0 in both groups. No serious adverse event terms are listed.

These are the counts recorded in the registry. A zero count means that no affected participants were recorded in the listed group for the specified category; it does not prove that the event is impossible or that the procedure has no risks. The registry record does not describe the adverse-event collection process, the criteria used to classify events, or the duration of safety monitoring. The primary outcome timeframe is 1 day, but the record does not state that this timeframe defines the safety follow-up period. The adverse-event figures should therefore be read as registry reporting, not as a comprehensive estimate of all possible harms.

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

The result shows a difference in the posted additional-testing outcome between the randomized ultrasound groups: the reported count was 63 among 317 in the TAS group and 14 among 312 in the TVS group. The unadjusted risk difference, expressed as TVS relative to TAS, was -0.154, with a 95% confidence interval from -0.203 to -0.104. The registry also reports an unadjusted p-value of <0.001. Together, these statistics describe the observed group contrast and its sampling precision under the methods used.

The endpoint measures whether additional testing was ordered before eligibility determination. It does not measure the type, burden, appropriateness, or consequences of that testing. The finding does not establish which approach is preferable for a particular person, nor does it determine whether an individual is eligible for medical abortion. It should not be read as medical advice.

The result also does not answer every question raised by the study summary. Patient satisfaction is named as a second primary outcome, but the registry record posts no result for it. The record does not provide results for other clinical outcomes, describe the reasons for additional testing, or show whether additional evaluation changed subsequent care.

The trial was randomized and unmasked, but the registry record does not report important details of allocation procedures, the analysis population, missing-data handling, or adjustment for provider or practice structure. These omissions limit evaluation of potential sources of bias and uncertainty. The recorded group denominators also differ from the overall enrollment, and the registry does not explain why. The numerical comparison is informative about the posted endpoint in this trial, while interpretation beyond that endpoint requires caution.

Finally, zero adverse events recorded in the listed categories does not establish an absence of risk. The safety collection and follow-up methods are not described in the registry record. The result is best understood as a comparison of additional-testing counts in the study context, not as a complete assessment of the two procedures or their broader consequences.

16. Why This Trial Matters Statistically

This trial poses a statistical question about what happens after an assessment: whether providers order additional testing following one of two ultrasound approaches. That is distinct from comparing diagnostic accuracy or patient outcomes. Naming the endpoint precisely helps prevent the result from being interpreted as evidence about outcomes that were not measured or posted.

The randomized comparison offers a direct group contrast, and the registry reports that contrast in several forms: a risk difference, risk ratio, odds ratio, and unadjusted p-value. Looking at the effect estimate together with its confidence interval makes the size and precision of the posted comparison more informative than relying on a p-value alone. The counts also keep the analysis tied to the number of recorded outcomes and the group denominators.

The record highlights why completeness matters in clinical-trial interpretation. The summary names patient satisfaction as a second primary outcome, but no result is posted for it. The overall enrollment and the listed primary-outcome denominators are not the same, and the record does not explain the reason. Analysis-population choices, missing data, and multiplicity are also not described. Those gaps do not erase the posted comparison, but they limit the ability to assess the full analysis and the range of questions addressed.

The planning illustrations are another useful statistical distinction: they describe hypothetical future sample sizes under stated assumptions, rather than the observed power of this completed trial. Keeping planning calculations separate from trial results avoids treating a retrospective illustration as part of the study's original design.

17. 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 leaves important design and reporting details unspecified. It does not report the randomization sequence method or allocation concealment, and the trial was unmasked. Because clinicians and participants could know which ultrasound approach was used, the record does not establish whether awareness affected decisions to order additional testing. The endpoint is based on provider orders, so consistent interpretation of what qualified as additional testing matters; operational details and independent adjudication are not reported.

The study summary anticipated enrollment of 800, while the completed record lists enrollment of 634. No explanation for this difference is reported. The primary outcome group denominators are 317 and 312, and the registry does not explain why these denominators differ from the overall enrollment. It does not report missing-outcome counts, post-randomization exclusions, or the analysis population.

The registry record does not provide the trial's original sample-size justification, the full statistical analysis plan, details of adjustment for provider or practice structure, or methods for multiplicity and missing data. It also does not report patient satisfaction results, although satisfaction is named as a second primary outcome. The adverse-event collection and follow-up methods are not described.

The study summary describes practices that routinely use TVS and characterizes the trial as multicenter. Results may be specific to the participating practice context; the registry record does not provide enough detail about sites, providers, or participant characteristics to assess how broadly the results apply. These limitations concern the information available in the record and should not be treated as proof that any particular design or analysis procedure was inadequate.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Transabdominal and Transvaginal Sonography in Medical Abortion — NCT03047551.
• Registry record last updated 2019-09-30; status last verified 2019-09.
• Results first posted 2019-09-30.
• 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.

21. Record Summary

NCT03047551 is a completed randomized, parallel-group, unmasked trial sponsored by Columbia University. It compared transabdominal sonography with transvaginal sonography for eligibility assessment before medical abortion. The record lists enrollment of 634 and primary outcome timeframe of 1 day.

The posted primary outcome counts are 63 among 317 participants in the TAS group and 14 among 312 in the TVS group. The risk difference for TVS relative to TAS is -0.154, with a 95% confidence interval from -0.203 to -0.104; the unadjusted p-value is <0.001. The registry record posts no secondary outcome results, including for the patient-satisfaction outcome named in the study summary.

Serious adverse events, other adverse events, and deaths are each recorded as 0 in both groups. The registry does not report the trial's original sample-size justification, detailed missing-data methods, or the rationale for the difference between overall enrollment and the listed outcome denominators.