Comparing Opioid Prescription Patterns in Total Joint Arthroplasty Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparing Opioid Prescription Patterns in Total Joint Arthroplasty Patients (NCT03236155), a randomized phase 4 study. The primary outcome is Morphine Milligram Equivalents (MME) of Oxycodone Tablets Taken by Patient.

2. Clinical Question

3. Trial Design

The registry describes NCT03236155 as a completed, randomized, triple-masked, parallel-group phase 4 trial sponsored by Rush University Medical Center. Enrollment is listed as 304. The record gives a start date of 2017-07-01 and a primary completion date of 2018-07-01. It does not report the method used to generate the random sequence, how allocation was concealed, or which people were masked.

The two groups differed in the prescription arrangement at hospital discharge. One group received 90 5 mg OxyIR pills in a single prescription. The other received 30 5 mg OxyIR pills at discharge, with two refills available if requested. The record describes equal total opioid availability as part of the trial's purpose, but it does not report how often refills were requested or received in the posted primary outcome results.

The primary outcome was the mean MME of oxycodone tablets taken by the patient during 30 days. Its stated population was opioid-naive patients undergoing THA or TKA. The registry description says patients self-reported the number of oxycodone doses taken and that an investigator would verify this by counting pills in the bottle. Details about the timing of outcome collection, adherence to assigned prescription arrangements, and the handling of patients without outcome data are not reported in the registry record.

The record lists eight secondary outcomes, including inpatient opioid utilization, pain scores, patient-reported outcomes, complications, compliance, and unused tablets. Posted numerical secondary results are available for inpatient opioid utilization. Results for the other listed outcomes are not reported in the registry record.

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

Analysis population for the primary outcome, as defined in the registry: Opioid naive patients undergoing THA or TKA between August 2017 and April 2018 and Morphine Milligram Equivalents (MME) of Oxycodone Tablets Taken by Patient

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.

Population: Minimum Age: 18 Years Maximum Age: none Sex: All Gender Based: No Ethnicity Based: no

6. Primary Endpoint

The registry lists the primary outcome as Morphine Milligram Equivalents (MME) of Oxycodone Tablets Taken by Patient, measured over 30 days.

The primary endpoint is the mean morphine milligram equivalents of oxycodone tablets taken by each patient over 30 days. MME expresses opioid amounts on a common scale; the record identifies the unit as MME but does not report a conversion method. The outcome is about medication taken, not merely medication prescribed or dispensed.

The registry says patients were asked to report how many 5 mg oxycodone doses they had taken. It also describes checking those reports against the patient's pill bottle by counting the remaining tablets. The record does not explain how disagreements between self-report and pill counts were handled, or how the dose counts were converted to MME.

A mean is used to describe the average outcome in each group. In the posted results, the 90-pill group had a mean of 1090.0 MME with a standard deviation of 536.4, while the 30-pill group had a mean of 777.1 MME with a standard deviation of 414.2. The standard deviation describes variability among observed patient outcomes; it is not the uncertainty interval for the group mean. A group mean provides a compact comparison, although it does not show how individual patients' amounts were distributed or whether the group values were concentrated around the mean.

This endpoint directly addresses the stated primary question about whether the prescription arrangement was associated with the amount taken during the specified follow-up. It does not by itself measure pain relief, refill experience, or whether unused pills were safely stored or disposed of.

7. Results Posted in the Registry

• 30 (5mg OxyIR) Pills Prescribed: mean 777.10 versus 1,090 on control; difference -312.60 (95% CI -421.33 to -203.87), standardized effect d = -0.66; 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.

Postoperative Inpatient Opioid Utilization

• 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:
  • 30 (5 mg OxyIR) Pills Prescribed: mean 72.20 versus 64.60 on control; difference 7.60 (95% CI -3.13 to 18.33), standardized effect d = 0.16; z-test p = 0.165.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between 30 (5 mg OxyIR) Pills Prescribed and the control. The interval gives the range of effects that remain compatible with the data.

The registry names eight secondary outcomes spanning inpatient opioid utilization, pain scores, patient-reported outcomes, complications, compliance, and unused tablets. Numerical results are posted for inpatient opioid utilization; results for the other listed secondary outcomes are not reported in the registry record. Consequently, the overall secondary-outcome pattern cannot be assessed across those other domains.

The available inpatient comparison concerns opioid use from surgery to hospital discharge, with the timeframe described as 1 day - 2 weeks. Its mean-difference interval includes zero, and the unadjusted p-value is 0.165. This result does not provide clear evidence of a difference in inpatient use. It also does not show that the groups were equivalent. The primary endpoint, by contrast, concerns oxycodone taken over 30 days. Differences in timeframe and outcome mean the inpatient result is not a direct substitute for the primary comparison.

The secondary result does not establish a mechanism for the primary finding. In particular, it does not report whether the prescription arrangement changed refill requests or outpatient dispensing. Nor can the missing secondary results show whether the primary comparison was accompanied by differences in pain, patient-reported outcomes, complications, compliance, or unused pills.

Unadjusted secondary results generally carry less weight when many outcomes may be examined, because multiple comparisons can increase the chance of apparently notable findings. General statistical explanation: a prespecified multiplicity strategy can help control that risk, but the registry does not report whether one was planned or used. The record also does not report a hierarchy for secondary outcomes. The single posted secondary comparison should therefore be interpreted as supportive descriptive evidence, not as a comprehensive account of all secondary questions.

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

• Postoperative Patient Reported Outcomes of Opioid naïve Patients Undergoing THA or TKA Between August 2017 and April 2018
• Complications
• Compliance
• Preoperative Patient Reported Outcomes
• Postoperative Outpatient Pain Scores
• Postoperative Inpatient Opioid Utilization
• Postoperative Inpatient Pain Scores
• Number of Tablets Received From Pharmacy That Went Unused

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a numerical amount of medication use, represented as MME. The registry reports each group's mean and standard deviation and gives a mean-difference comparison. General statistical explanation: the mean describes the center of the observed values, while the standard deviation describes how spread out those values are around the mean. Neither summary alone reveals the full shape of the outcome distribution.

What does the mean difference represent?

The comparison is framed as the 30-pill group relative to the 90-pill group. Its estimate is -312.6 MME. The negative sign indicates that the estimated mean for the 30-pill group is lower under that comparison's direction. The posted group means and the comparison estimate are registry-reported values; the estimate should be interpreted as reported rather than reconstructed from rounded group summaries.

A mean difference is expressed on the same scale as the outcome. Here, that scale is MME. It gives an absolute contrast in average use, unlike a standardized effect, which expresses a contrast relative to a measure of variability. The registry also lists a standardized effect estimate of -0.657, but it does not report the analysis method used to obtain the primary comparison. The effect estimate does not describe the degree of overlap between individual patients' outcomes.

What does the confidence interval add?

The 95% confidence interval for the primary mean difference runs from -421.3 to -203.9. General statistical explanation: a confidence interval represents uncertainty in an estimate under a specified statistical procedure. A narrower interval indicates greater precision on that scale than a wider interval, but precision is not the same as freedom from bias or clinical importance. The registry does not report the method used to construct this interval.

Both reported endpoints of the primary interval are negative, which is consistent with the direction of the estimated difference. The interval does not show the range in which most individual patients' values fall. For that, individual-level distributions or suitable quantiles would be needed; these are not reported in the registry record.

How should the p-value be read?

The primary p-value is reported as <0.001 and is labeled unadjusted. General statistical explanation: a p-value evaluates how unusual data at least as incompatible with a null hypothesis would be if that hypothesis and the test assumptions held. It does not measure the probability that the observed result arose by chance, and it does not quantify the probability that the treatment arrangements are equivalent. The registry does not report the specific test, its assumptions, or any adjusted p-value.

The numerical p-value and confidence interval address statistical uncertainty about the group contrast, not whether the amount difference matters to patients. Clinical importance would require relevant context, such as a stated threshold for a meaningful difference or links to pain and other outcomes. Such a threshold is not reported in the registry record.

Why do standard deviations matter?

The primary standard deviations are 536.4 and 414.2 MME. They describe substantial observed spread relative to the group means, but no additional calculation is needed to interpret them. General statistical explanation: variability affects how precisely group averages can be estimated. A wide spread can make an average comparison less precise unless there is enough information in the data to offset it. The standard deviation is not a confidence interval and should not be read as the uncertainty around the mean.

The registry reports a mean-based endpoint and mean-based comparison. It does not report whether the outcome distribution was assessed for skewness, whether a transformation or robust method was considered, or whether sensitivity analyses were conducted. Medication-use quantities can in general be unevenly distributed, but whether that applied in this trial is not reported in the registry record.

What do the reported secondary statistics show?

For inpatient opioid utilization, the registry reports group means of 64.6 and 72.2 morphine equivalents, with standard deviations of 38.9 and 55.9. The reported mean difference for the 30-pill group relative to the 90-pill group is 7.6, with a 95% confidence interval from -3.133 to 18.33 and an unadjusted p-value of 0.165. The interval includes zero, so the posted estimate does not establish a difference at conventional significance thresholds. It also does not establish that the groups are equivalent. The registry does not report the test used for this secondary comparison.

The remaining secondary endpoints are named in the record, but their numerical results are not reported in the registry record. The record also does not describe an analysis plan that explains how these outcomes were selected, compared, or interpreted alongside the primary endpoint.

What is not documented about the analysis?

The registry does not report the primary analysis population, missing-data strategy, covariate adjustment, model specification, assumptions checks, or interim analyses. It does not report the original sample-size justification or the trial's own planned power. These omissions limit evaluation of how the posted comparison relates to the trial's planned analysis. They do not change the numerical values recorded for the primary outcome, but they constrain conclusions about robustness and the scope of inference.

11. Confidence Intervals and Precision

The primary comparison has a reported 95% confidence interval from -421.3 to -203.9 MME for the mean difference, defined as the 30-pill group relative to the 90-pill group. Because both endpoints are below zero, the interval is consistent with a lower mean in the 30-pill group. The estimated difference is -312.6 MME, and the unadjusted p-value is <0.001.

General statistical explanation: a confidence interval gives a range of values compatible with the data under the interval's statistical procedure. Its width conveys precision on the outcome's scale. It is not a prediction interval for individual patients and does not mean that a specified proportion of individuals will have outcomes inside it. The construction method for the registry interval is not reported in the registry record.

The interval gives a range of plausible group mean differences, but does not settle whether the magnitude is clinically important. The registry does not state a threshold for a meaningful difference. It also does not report individual-level distributions, which would be needed to assess how much patients' outcomes overlap across groups.

The secondary inpatient-utilization comparison is less precise in relation to zero: its 95% confidence interval runs from -3.133 to 18.33 morphine equivalents. This interval includes zero and permits values in either direction. It is not evidence of equivalence; an equivalence assessment would require an appropriate design and prespecified margin, neither of which is reported in the registry record.

Precision should be considered alongside the uncertainty in the conduct and analysis. The registry does not report missing-data handling, the analysis population, or the interval-estimation method. Those gaps prevent a fuller assessment of how sensitive the reported precision may be to analysis choices.

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 participants or total events. The registry's planning illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is a stated fraction of the observed effect. The trial's own original design assumptions and sample-size justification are not reported in the registry record.

For the primary outcome, planning for 100% of the observed effect corresponds to 37 participants per arm for 80% power and 49 participants per arm for 90% power. Planning for 75% of the observed effect corresponds to 65 participants per arm for 80% power and 87 participants per arm for 90% power. Planning for 50% of the observed effect corresponds to 146 participants per arm for 80% power and 195 participants per arm for 90% power.

In plain language, 80% power means that, under the listed planning assumptions and if the specified effect were the true effect, the design would have an 80% chance of detecting it using the stated statistical threshold. The 90% column uses a higher chance of detection under those same assumptions. Smaller effects need more participants to be detected with the same power, as reflected in the planning rows.

For inpatient opioid utilization, planning for 100% of the observed effect corresponds to 644 participants per arm for 80% power and 862 participants per arm for 90% power. At 75% of that observed effect, the figures are 1,144 and 1,532 participants per arm. At 50%, they are 2,574 and 3,445 participants per arm, respectively. These are planning illustrations based on observed effects, not observed power. They do not establish what power the trial had under its original design.

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 = -312.600; pooled standard deviation σ = 475.585.
2. n = 2 × (1.96 + 0.842)² × 475.585² / -312.600² = 36.3, rounded up to 37 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 <- -312.6                      # observed difference in means
sigma <- 475.585406846444                     # 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(3236155)
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 = 37 (80% power), 49 (90% power)
 75% of observed effect: n per arm = 65 (80% power), 87 (90% power)
 50% of observed effect: n per arm = 146 (80% power), 195 (90% power)
power.t.test: 37.3 
Simulated power at n = 37 per arm: 0.796 (20000 replicates)

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

14. Multiplicity, Missing Data and Analysis Populations

The registry reports an unadjusted p-value for the primary comparison and an unadjusted p-value for the posted inpatient secondary comparison. It does not report a plan for multiplicity adjustment, a hierarchy among the primary and secondary outcomes, or a prespecified strategy for interpreting multiple tests. It is therefore not possible to determine from the registry record whether the posted p-values were intended as confirmatory tests or how the family of outcomes was handled.

General statistical explanation: when several outcomes or comparisons are tested, the chance of at least one apparently positive result can rise if each test is interpreted independently. Adjustments or a prespecified outcome hierarchy can address this issue, though their appropriateness depends on the study question and analysis plan. No particular adjustment should be assumed because it is not reported in the registry record.

The registry lists nine outcome measures in total and eight secondary outcomes. It provides numerical results for the primary endpoint and inpatient opioid utilization, while results for the other listed secondary endpoints are not reported in the registry record. This incomplete numerical record makes it difficult to judge the full set of analyses or whether selective reporting may have affected the available picture. The registry does not explain why results for the remaining outcomes are absent.

Missing-data methods are also not reported in the registry record. The record does not state how many participants had a primary outcome available, whether any outcome values were missing, or how missing values were handled. The listed arm counts for the primary outcome are 143 and 161, but the registry does not explain whether these counts represent all randomized participants, an analysis population, or another reporting basis. No intention-to-treat or per-protocol analysis population is specified.

The registry also does not report interim analyses, covariate adjustment, sensitivity analyses, or the original analysis plan. These omissions prevent assessment of whether the final analysis followed a prespecified approach. They do not by themselves demonstrate that analyses were inappropriate; they identify information needed for a fuller statistical appraisal.

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.

The serious adverse event figures list 143 people at risk in the 90-pill group and 161 in the 30-pill group. The registry records zero serious adverse events in each group. It also records zero other adverse events and zero deaths in both groups. These are the adverse-event counts posted for the groups, not evidence that no other harms could occur.

General statistical explanation: observing zero events does not prove that the underlying event risk is zero. The interpretation depends on the amount and duration of safety follow-up, how events were defined and collected, and how complete the reporting was. The registry does not give further details about safety surveillance, the follow-up period for adverse events, or event definitions beyond the serious and other adverse-event categories.

The record lists no top serious adverse-event terms. The zero counts should therefore be read narrowly as the registry's reported figures for this trial, rather than as a comprehensive comparison of safety under all circumstances. The available figures do not support a conclusion about rare events.

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

The primary posted result is a randomized-group comparison of mean oxycodone use, expressed in MME, over 30 days. The registry reports a lower mean in the group prescribed 30 pills at discharge with two refills available if requested than in the group prescribed 90 pills at discharge. The estimated mean difference is -312.6 MME, with a 95% confidence interval from -421.3 to -203.9 and an unadjusted p-value of <0.001.

This finding describes the outcome measured and recorded in the registry. It does not show that every patient in the 30-pill group took less, and it does not establish how individual outcomes were distributed. Nor does the primary endpoint directly measure pain, function, patient satisfaction, refill burden, or unused medication. The listed secondary outcomes include several of these broader topics, but numerical results for most of them are not reported in the registry record.

The trial purpose describes a comparison in which total opioid availability could be equal despite differences in pill quantity per prescription. The record does not report how often refills were requested, how many refills were obtained, or the total amount dispensed in each group. The primary result therefore should not be interpreted as evidence about those unreported details.

The comparison is not a statement about all patients undergoing joint arthroplasty, other procedures, or other prescription arrangements. The registry specifies opioid-naive patients undergoing THA or TKA for the primary outcome, but details needed to assess generalizability—such as eligibility criteria and setting beyond the sponsor—are not reported in the registry record.

The result also does not establish that one prescription arrangement is clinically preferable. The registry does not provide a threshold for a clinically meaningful change, a complete report of patient-centered secondary outcomes, or enough detail about missing data and analysis populations to resolve all questions about robustness. The posted estimate is informative about the recorded comparison, but its scope should remain tied to that outcome, population description, and follow-up period.

17. Why This Trial Matters Statistically

This trial raises a statistical question about prescription design and measured use: does changing the amount available at each prescription opportunity relate to the amount patients report taking over a defined follow-up? Its randomized parallel-group design gives the comparison a structure suited to studying differences between assigned prescription arrangements. The primary endpoint focuses on medication taken rather than only the amount prescribed.

The reporting illustrates why an estimate, its uncertainty, and the measurement process should be considered together. The primary mean difference is accompanied by a confidence interval and an unadjusted p-value, while the endpoint description explains that self-report was to be checked against pill counts. Those details help readers understand both the statistical contrast and how the outcome was intended to be measured.

The record also shows the limits of interpreting a statistically clear primary comparison in isolation. The secondary outcomes include pain, patient-reported outcomes, complications, compliance, and unused tablets, but most do not have numerical results in the registry record. The available inpatient opioid-utilization comparison is a different outcome over a different timeframe. Without the remaining results, the broader pattern across medication use and patient experience cannot be assessed.

Finally, the record distinguishes observed results from planning illustrations. The sample-size displays use the observed effects and stated planning assumptions; they are not evidence of the original design rationale or observed power. The original sample-size justification, analysis plan, missing-data approach, and multiplicity strategy are not reported. Recognizing these distinctions helps keep the interpretation anchored to the posted evidence rather than to assumptions about unreported methods.

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 registry record leaves important design and analysis details unspecified. It does not report eligibility criteria in full, recruitment setting, allocation concealment, the randomization procedure, or which parties were masked. Although masking is listed as triple, the identities of the masked parties and the practical masking procedures are not reported in the registry record.

The primary endpoint relies on patient self-report with a pill-bottle count described as verification. The registry does not explain how discrepancies between those sources were resolved, whether all participants brought bottles, or how missing or incomplete reports were treated. It also does not report a conversion method for expressing oxycodone use as MME.

The analysis population, missing-data strategy, statistical test, covariate adjustment, and assumptions checks are not reported. The registry gives arm counts for the primary comparison but does not explain the relationship between those counts and the overall enrollment. The original sample-size justification and trial-specific power assumptions are also not reported; the displayed planning calculations instead use observed effects.

Most secondary outcome results are absent from the registry record. This limits assessment of whether the primary comparison was accompanied by differences in pain, patient-reported outcomes, complications, compliance, or unused medication. Multiplicity handling and a hierarchy for interpreting secondary outcomes are also not reported.

Safety figures record zero serious adverse events, other adverse events, and deaths in both groups, but detailed event definitions and safety follow-up procedures are not reported. The 30-day primary outcome and described inpatient timeframe do not by themselves establish longer-term patterns. These limitations do not negate the posted primary estimate; they define what can and cannot be evaluated from the registry information.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Comparing Opioid Prescription Patterns in Total Joint Arthroplasty Patients — NCT03236155.
• Registry record last updated 2023-07-20; status last verified 2023-06.
• Results first posted 2023-07-20.
• 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

NCT03236155 is listed as a completed, randomized, triple-masked, parallel-group phase 4 trial sponsored by Rush University Medical Center, with enrollment of 304. It compared a single 90-pill prescription at hospital discharge with a 30-pill prescription at discharge and two refills available if requested.

The primary endpoint was mean MME of oxycodone tablets taken over 30 days by opioid-naive patients undergoing THA or TKA. The registry reports a lower mean in the 30-pill group, with a mean difference of -312.6 MME, a 95% confidence interval from -421.3 to -203.9, and an unadjusted p-value of <0.001.

Numerical results are also posted for inpatient opioid utilization; results for the other listed secondary outcomes are not reported in the registry record. Adverse-event figures record zero serious events, other events, and deaths in both groups. The original sample-size justification, analysis population, missing-data methods, and multiplicity plan are not reported.