Preoperative Steroids in Autoimmune Thyroid Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Preoperative Steroids in Autoimmune Thyroid Disease (NCT04542278), a randomized phase 4 study. The primary outcome is Thyroid Difficulty Scale Score.

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group, Phase 4 trial with no masking. It lists University of Wisconsin, Madison as the sponsor and Thyroid Diseases as the condition. The trial is marked completed. Its recorded start date is 2020-11-13, and its primary completion date is 2022-10-04. The listed enrollment is 30.

The two arms were called Steroids and No Steroids. The steroid-arm description specifies prednisone 20mg daily for 7 days before surgery, otherwise preoperative standard of care; the comparison arm is described as preoperative standard of care. The posted primary results include 14 participants in the steroid arm and 16 in the no-steroids arm. These are the arm counts recorded with the outcome results, not a basis for inferring how participants were assigned or why the counts differ.

The registry describes the project as a pilot and says that participants would be on study for up to 7 months. The primary outcome was assessed at the conclusion of the operation. Secondary outcome timeframes include immediately after surgery for parathyroid hormone and up to 4 weeks, on Post Operative Day 1, for recurrent laryngeal nerve injury. The registry lists nine outcome measures, while three secondary outcomes are posted in the record. Details of the original design assumptions, including a sample-size justification, are not reported in the registry record.

Randomization can help balance known and unknown participant characteristics between arms on average, but it does not guarantee balance in a small trial. With no masking, knowledge of assignment could also affect subjective measurement or other aspects of care. The record does not describe whether any procedures were used to limit such influence.

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 Thyroid Difficulty Scale Score, measured over surgery occurs up to 4 weeks, data collected a conclusion of operation.

The primary endpoint is the Thyroid Difficulty Scale score completed by the surgeon at the conclusion of the operation. It is a four-item survey covering vascularity, friability, mobility, and glandular size of the thyroid. The possible score range is 4-20, and higher scores indicate increased difficulty in thyroid surgery. The posted measure is a mean score with a standard deviation for each arm.

This endpoint is intended to represent the surgeon’s assessment of operative difficulty, the central question motivating the trial: whether a short preoperative course of prednisone is associated with a change in the inflammation of the gland and the perceived difficulty of surgery. The scale’s total score combines ratings across its listed features, rather than reporting each feature as a separate primary result. The registry record does not report further information about how the scale was validated, how the four items are combined, or whether scoring guidance was standardized between surgeons.

As a surgeon-rated measure, the score is not the same as an objective count of complications or an assessment of long-term health. Its meaning should therefore be kept specific: it describes the posted operative-difficulty score, not every aspect of surgery or recovery. The timing at the conclusion of the operation links the assessment to the procedure itself, while the absence of masking is relevant when considering the potential for expectations to influence a subjective rating.

7. Results Posted in the Registry

• No Steroids: mean 9.75 versus 9.80 on control; difference -0.05 (95% CI -2.84 to 2.74), standardized effect d = -0.01; z-test p = 0.972.

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.

Surgical Complications: Parathyroid Hormone (PTH) Level

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • No Steroids: mean 38.00 versus 52.00 on control; difference -14.00 (95% CI -40.90 to 12.90), standardized effect d = -0.38; z-test p = 0.308.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between No Steroids and the control. The interval gives the range of effects that remain compatible with the data.

Number of Participants With Surgical Complications: Recurrent Laryngeal Nerve (RLN) Injury

• 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:
  • No Steroids: 12.5% (2/16) versus 14.3% (2/14) on control; risk difference -1.8 percentage points (95% CI -26.3 to 22.7), risk ratio 0.88 (0.14–5.42); unadjusted two-proportion z-test p = 0.886.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between No Steroids and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover different aspects of surgery and postoperative assessment: ultrasound Doppler quantification of blood flow is listed as an outcome, while posted results include parathyroid hormone and recurrent laryngeal nerve injury. Their measures differ from the surgeon-rated primary score, so they offer related but not interchangeable information. They do not directly confirm or refute the primary result.

Across the posted secondary comparisons, the estimates remain uncertain. The continuous parathyroid hormone comparison has an interval that includes no difference, while the nerve-injury comparison is based on a small number of recorded events and has a broad interval. Taken together with the primary result, the secondary results do not form a precise, consistent pattern that would establish a clear effect across surgical difficulty, biochemical measurement, and injury outcomes. They also cannot be collapsed into a single overall effect because they measure different outcomes.

These secondary comparisons are unadjusted, and no reported analyses are listed for them. When multiple outcomes are assessed, the chance of observing an apparently notable result by chance can increase if each result is considered separately. A multiplicity adjustment or a prespecified hierarchy can address that issue, but the registry does not report such an approach. Accordingly, isolated secondary findings deserve less weight than a primary result that was designated in advance, and even the primary result must be interpreted with its uncertainty and design limitations in view.

The record lists three secondary outcomes as posted, but the trial entry lists nine outcome measures overall. The relationship between all listed measures and the posted secondary results is not fully described. Missing outcome details and the absence of reported analyses limit the ability to assess whether the posted results represent the complete set of planned comparisons. The secondary evidence is therefore best treated as exploratory and descriptive rather than as a basis for confirmatory conclusions.

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

• Ultrasound Doppler Quantification of Blood Flow
• Surgical Complications: Parathyroid Hormone (PTH) Level
• Number of Participants With Surgical Complications: Recurrent Laryngeal Nerve (RLN) Injury

9. Statistical Interpretation

10. Statistical Methods Explained

The main comparison concerns a continuous outcome: the Thyroid Difficulty Scale score. The registry reports the mean and standard deviation in each randomized arm, then gives a mean difference comparing No Steroids with Steroids. The direction is important: a negative mean difference means the posted average in the no-steroids arm is lower than the posted average in the steroid arm. The difference is an unadjusted comparison of the posted group summaries; it is not reported as adjusted for baseline characteristics or other variables.

What does the mean and standard deviation describe?

General statistical context: a mean gives the arithmetic average of observed scores in a group. A standard deviation describes how spread out individual scores are around that group average. It is not the uncertainty in the mean itself. Uncertainty in a mean comparison also depends on how many observations contribute to each group and on the variability of the observations. The registry reports a standard deviation for each arm, rather than a confidence interval around each arm’s mean.

The primary comparison reports a mean difference and a confidence interval. The confidence interval estimates a range of values compatible with the data and the model used to calculate it. It is not a range containing a specified proportion of individual participant scores, and it does not assign a probability to any one treatment effect. For this trial, the interval crosses zero, so it includes both a lower mean in the no-steroids arm and a higher mean in that arm relative to the steroid arm. Its breadth indicates substantial uncertainty about the size and direction of the difference.

What does the p-value test?

General statistical context: a p-value measures how unusual the observed data, or data more extreme, would be under a null hypothesis and the assumptions of the statistical method. It is not a measure of clinical importance, the probability that the null hypothesis is true, or proof that groups are equivalent. A large p-value can occur when the groups are similar, but it can also occur when an estimate is imprecise. The primary comparison’s unadjusted p-value of 0.972 should therefore be considered together with its estimate and interval, not used alone.

The registry labels the primary measure as a mean with standard deviation and reports a mean-difference comparison. It does not report the detailed statistical test specification, any adjustment variables, or a missing-data method. The record also lists zero reported analyses. Further details of the analysis plan are not reported in the registry record. The displayed comparison should not be interpreted as confirming that a prespecified analysis plan was followed when the record does not describe that plan.

Why does the unmasked design matter?

General statistical context: randomization is intended to reduce systematic differences between groups at assignment, but small samples can still have chance imbalances. Masking is intended to limit the influence of treatment knowledge on assessments or care. Here, allocation is listed as randomized and masking as none. Because the primary outcome is a surgeon-rated scale, awareness of assignment could potentially influence ratings. The registry does not report whether the surgeons who scored the outcome were also masked in any other way, and no such practice should be assumed.

The secondary outcomes use different statistical forms. Parathyroid hormone is reported as a continuous measure with group means, standard deviations, and a mean difference. Recurrent laryngeal nerve injury is reported as a participant count, with a risk-difference comparison. Those outcomes cannot be interpreted as if they were additional versions of the primary score: they measure different clinical features, have different scales, and carry different sources of uncertainty. The registry’s posted secondary analyses are unadjusted.

Why can over-dispersion matter in other settings?

General statistical context: over-dispersion arises when count data vary more than an assumed statistical model expects. If that extra variability is ignored, standard errors and confidence intervals can be too small, creating an overly confident impression. The primary outcome is a continuous score, not a count outcome, so over-dispersion is not the central issue for its reported mean comparison. The registry does not report a model-based count analysis for the recurrent laryngeal nerve outcome, and no such model should be inferred.

Finally, a statistically nonsignificant comparison is not a test of equivalence or noninferiority unless such a design and analysis are specified. The registry does not report an equivalence margin, a noninferiority margin, or the corresponding analysis. The posted results should be read as ordinary unadjusted comparisons, with uncertainty, rather than as a formal demonstration that the arms are the same.

11. Confidence Intervals and Precision

The primary mean difference is -0.05, and its 95% confidence interval extends from -2.844 to 2.744. Because the interval includes zero and reaches in both directions, the observed comparison is compatible with a lower mean score in the no-steroids arm, little or no difference, or a higher mean score in that arm. The estimate alone is not precise enough to settle which of those possibilities best describes the underlying difference.

General statistical context: a confidence interval communicates both the estimated effect and the uncertainty around it. A narrower interval indicates greater precision under the method’s assumptions; a wider interval leaves more possible effect sizes compatible with the data. The interval does not mean that each possible value is equally likely, nor does it measure the spread of individual scores. Individual-level spread is described by the arm standard deviations, which are 3.8 for Steroids and 4.0 for No Steroids.

The primary interval should be interpreted in light of the pilot enrollment and the unmasked, surgeon-rated outcome. The registry does not report a precision target or a clinically important difference for the scale. Without such context, the interval cannot be classified as ruling out differences that matter to patients or clinicians. A p-value of 0.972 does not make the interval narrower or remove the uncertainty it displays.

The secondary intervals also communicate uncertainty, but they apply to different measures and should not be compared as if they were on one common scale. The parathyroid hormone mean-difference interval includes values on both sides of zero. The recurrent laryngeal nerve risk-difference interval is broad relative to the small number of recorded events. The precision of these secondary comparisons is limited, and the registry does not report adjusted analyses for them.

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 for 80% power and participants per arm for 90% power. They are not total sample sizes. In general, power is the probability that a study will meet its statistical significance criterion if a particular effect truly exists and the planning assumptions hold. The 80% and 90% columns therefore describe different target probabilities for detecting the assumed effect, not probabilities that the posted results are correct.

The planning assumptions listed in the registry are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect used for each illustration is a stated fraction of the observed effect. Smaller assumed effects require more participants to detect under the same planning assumptions, which is why the figures rise as the planning effect moves from 100% to 75% to 50% of the observed effect.

These are planning illustrations calculated from the observed effects, not the trial’s original design assumptions. The trial’s own sample-size justification is not reported in the registry record. For the outcomes in this trial, the figures are participants per arm; the general note that time-to-event planning displays use total events does not apply to these outcomes. The figures are not observed power. They should be read as hypothetical planning requirements under the stated assumptions, not as a retrospective score of whether the completed trial was adequately powered.

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 = -0.050; pooled standard deviation σ = 3.908.
2. n = 2 × (1.96 + 0.842)² × 3.908² / -0.050² = 95,917.8, rounded up to 95,918 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 <- -0.0500000000000007                      # observed difference in means
sigma <- 3.90841582814755                     # 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(4542278)
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 = 95,918 (80% power), 128,407 (90% power)
 75% of observed effect: n per arm = 170,521 (80% power), 228,279 (90% power)
 50% of observed effect: n per arm = 383,672 (80% power), 513,627 (90% power)
power.t.test: 95918.8 
Simulated power at n = 95,918 per arm: 0.802 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity concerns arise when a study evaluates several outcomes or makes several statistical comparisons. General statistical context: if each comparison is judged separately at a chosen significance threshold, the chance of at least one apparently unusual result can increase across the set. Researchers may address this through an adjustment, a hierarchy of outcomes, or a clear distinction between confirmatory and exploratory analyses. The registry record does not report a multiplicity adjustment, a testing hierarchy, or another method for controlling this issue.

The record identifies a primary outcome and lists secondary outcomes, including measures related to blood flow, parathyroid hormone, and recurrent laryngeal nerve injury. It lists nine outcome measures but posts three secondary outcomes. The registry does not explain the status of the other listed measures or whether all were intended for formal hypothesis testing. No reported analyses are listed for the primary or secondary comparisons. This limits assessment of which analyses were planned, which were completed, and how the results were selected for posting.

Missing-data handling is not reported in the registry record. The posted arm counts show the number of participants contributing to the reported results, but the record does not describe whether any participants had missing measurements, how missingness was defined, or whether any imputation or other approach was used. It also does not report an analysis population such as intention-to-treat or per-protocol. Those omissions matter because the way participants with incomplete data are handled can affect group comparisons.

The registry also does not report the detailed statistical analysis plan, covariate adjustment, or any interim analyses. These should not be inferred from the posted estimates. The absence of such information does not establish that a particular method was or was not used; it means the methods cannot be verified from the registry record. In consequence, the comparisons are most safely read as the unadjusted results that are posted, with the limitations of incomplete analysis documentation kept in view.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The adverse-event record lists zero serious adverse events in the steroid arm among 14 at risk and zero in the no-steroids arm among 16 at risk. It also lists zero deaths in each arm. No serious adverse-event terms are listed. These entries describe what was recorded in this trial; they do not establish that serious events cannot occur or that the arms have equivalent safety.

The record lists other adverse events in 9 participants in the steroid arm and 2 in the no-steroids arm. These are counts, not an adjusted comparison, and the registry record does not provide event details or a statistical analysis for them. A count alone does not show the nature, severity, timing, or relationship of an event to treatment. The registry also does not report additional methods for assessing or attributing adverse events.

Because this was a pilot trial with limited enrollment, the absence of recorded serious events is especially limited as evidence about uncommon harms. Safety interpretation should remain descriptive: the registry reports no serious events or deaths, and it reports other adverse events in both arms. Those observations do not support a broad conclusion about comparative safety.

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

The primary result is a comparison of mean surgeon-rated thyroid difficulty scores after random assignment to preoperative prednisone or preoperative standard of care without steroids. The posted estimate is close to zero, but the confidence interval is wide and includes differences in either direction. It therefore does not provide a precise estimate of how much the groups might differ in the wider population represented by the trial.

These findings do not establish that prednisone improves or worsens the operative experience, and they do not establish that it has no effect. The study is described as a pilot, the posted primary comparison is unadjusted, and the outcome is a surgeon rating in a trial with no masking. The results apply specifically to the recorded Thyroid Difficulty Scale measure and should not be extended automatically to other aspects of surgery, recovery, or longer-term outcomes.

The secondary results address distinct measures and are also imprecise. They do not turn the primary comparison into a definitive conclusion, and the absence of a reported multiplicity approach further limits confirmatory interpretation. Likewise, the adverse-event entries describe events recorded in this trial but are not evidence that serious harms are impossible or that safety is equivalent between arms.

General statistical context: a trial can produce an estimate near zero while leaving substantial uncertainty around the true effect. More data can narrow uncertainty, but whether further research is justified depends on considerations beyond these statistics, including the importance of the outcome and the feasibility of obtaining more precise evidence. The registry does not report a clinically important score difference or a formal equivalence objective, so neither should be inferred from the posted result.

17. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates the difference between a small point estimate and a precise conclusion. The primary mean difference is near zero, but its confidence interval remains broad. Reading the estimate together with its uncertainty avoids treating a large p-value as proof that the two arms are identical.

The record also highlights how outcome type affects interpretation. A surgeon-rated continuous score, a laboratory measurement, and a participant count for an injury each answer different questions and carry different uncertainties. A single overall narrative cannot be built by treating their estimates as if they measured the same construct. The secondary comparisons are particularly limited as confirmatory evidence because they are unadjusted and the registry does not report a multiplicity strategy.

Study design and reporting details also matter. Randomization is listed, but masking is not. This is relevant to a subjective primary assessment. The record does not describe the analysis plan, missing-data handling, or analysis populations, limiting the extent to which the posted comparisons can be independently assessed. The enrollment is characterized as a pilot, and the registry does not report the original sample-size justification.

Finally, the planning illustrations show how sample-size requirements depend on the effect assumed for planning. They are based on observed effects and stated statistical assumptions, not on the original trial design. Distinguishing these retrospective planning illustrations from the study’s actual design helps prevent them from being mistaken for evidence about what the trial was built to detect. Together, these features make the record useful for understanding uncertainty, measurement, and reporting limits in a small randomized clinical study.

18. Limitations

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

The enrollment is limited, and the registry describes the project as a pilot. The primary confidence interval is wide, so the estimate does not precisely determine the size or direction of any underlying difference. The record does not report the original sample-size justification or a target precision for the primary outcome.

The primary endpoint is completed by the surgeon, and masking is listed as none. That leaves potential for knowledge of assignment to influence a subjective assessment. The registry does not report whether a common scoring protocol was used across surgeons, whether the same surgeons assessed all participants, or whether any other steps were taken to reduce assessment bias. No conclusion about those procedures can be inferred from the record.

The statistical documentation is incomplete. The registry does not report detailed analysis methods, missing-data handling, an analysis population, covariate adjustment, or an approach to multiple comparisons. It lists nine outcome measures but posts three secondary outcomes, and does not explain the relationship between all listed outcomes and the posted analyses. These omissions limit assessment of the completeness and prespecification of the results.

The secondary findings are also constrained by their different measures and limited precision. The recurrent laryngeal nerve outcome has few recorded events, while the continuous secondary measure has an interval spanning values on either side of no difference. Adverse-event reporting is descriptive, with no serious events or deaths recorded, but the trial is not large enough to establish conclusions about rare events. Finally, the results concern the specific outcomes and timeframes recorded; broader clinical outcomes and longer-term effects are not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Preoperative Steroids in Autoimmune Thyroid Disease — NCT04542278.
• Registry record last updated 2023-12-26; status last verified 2023-12.
• Results first posted 2023-12-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT04542278 is a completed randomized, parallel-group, unmasked Phase 4 trial sponsored by University of Wisconsin, Madison. The registry lists an enrollment of 30 participants with autoimmune thyroid disease undergoing surgery. The intervention was prednisone 20mg daily for 7 days before surgery, alongside preoperative standard of care; the comparison was preoperative standard of care without steroids.

The primary outcome was the surgeon-completed Thyroid Difficulty Scale score at the conclusion of the operation. The posted mean difference comparing No Steroids with Steroids was -0.05, with a 95% confidence interval from -2.844 to 2.744 and an unadjusted p-value of 0.972. Secondary results include parathyroid hormone and recurrent laryngeal nerve injury. The record lists no reported analyses and does not report the original sample-size justification, missing-data methods, or a multiplicity approach.