Risk Factor Control Before Orthopedic Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Risk Factor Control Before Orthopedic Surgery (NCT01837069), a randomized phase 4 study. The primary outcome is Number of Partipants That Experienced Death, Myocardial Infarction, Stroke, Transient Ischemic Attack, Myocardial Necrosis, or Venous Thromboembolism.

2. Clinical Question

3. Trial Design

The registry describes a phase 4, randomized, open-label, parallel-group trial titled Risk Factor Control Before Orthopedic Surgery. It lists NYU Langone Health as the sponsor. The study began in 2014-02, had a primary completion date of 2018-04-02, and is recorded as terminated. The reason for termination is not reported in the registry record.

The registry reports enrollment of 198. The treatment arm lists 103 participants and the control arm lists 95. Random allocation is recorded, but the method used to generate or conceal the allocation sequence is not reported in the registry record. Masking is listed as none, so the design is open-label. The registry does not describe whether endpoint assessment was blinded or how the component events were verified.

The treatment was multifactorial rather than a single drug intervention. It combined counseling with atorvastatin and conditional use of metoprolol and lisinopril based on elevated heart rate or systolic blood pressure, respectively. The comparator is described only as standard of care. The registry does not provide further detail about how those treatment conditions were assessed or how adherence was monitored.

The primary endpoint is a participant count for a composite of serious cardiovascular and thromboembolic events over 30 days. A modified composite of cardiovascular events is listed as a secondary outcome. Four outcome measures are recorded, while the posted analysis count is zero. The registry record therefore contains group outcome counts and a comparison, but does not report a fuller statistical-analysis record. The original protocol, detailed eligibility criteria, and original design assumptions 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

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 Number of Partipants That Experienced Death, Myocardial Infarction, Stroke, Transient Ischemic Attack, Myocardial Necrosis, or Venous Thromboembolism, measured over 30 days.

The primary endpoint counts participants who experienced death, myocardial infarction, stroke, transient ischemic attack, myocardial necrosis, or venous thromboembolism within 30 days. It is recorded in participants, with the outcome measure described as a number. The listed components cover several kinds of cardiovascular and thromboembolic events, and combining them produces a single outcome for the primary comparison.

In general, a composite endpoint can capture whether a participant experienced any event included in the definition. It can also make a trial’s main comparison more informative when the individual components are related to the clinical question. However, a composite count does not show how many cases came from each component, whether some components were more frequent than others, or whether the components had similar clinical importance. Component-specific results are not reported in the registry record.

The primary endpoint is reported at the participant level rather than as a total number of event episodes. The posted results list 10 participants in the treatment arm and 2 in the control arm. The registry does not state how events were adjudicated, whether recurrent events were counted separately, or whether endpoint status was available for every participant. Those details matter when interpreting a composite but cannot be established from the registry record.

7. Results Posted in the Registry

• Treatment: 9.7% (10/103) versus 2.1% (2/95) on control; risk difference 7.6 percentage points (95% CI 1.2 to 14.0), risk ratio 4.61 (1.04–20.51); unadjusted two-proportion z-test p = 0.025.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is binary at the participant level: each participant either experienced the composite endpoint during the stated 30-day timeframe or did not, as represented by the posted count. The registry reports group sizes and counts, then gives a treatment-versus-control comparison using a risk difference, risk ratio, odds ratio, confidence intervals, and an unadjusted p-value. The record does not describe a detailed statistical analysis plan.

What does the risk difference describe?

In general, a risk is the proportion of participants in a group who experience an outcome over a defined period. A risk difference compares the risks by subtracting the control risk from the treatment risk. A value above zero indicates a greater observed risk in the treatment arm under that orientation of the comparison; a value below zero indicates a smaller observed risk. Here, the reported risk difference is 0.076. It is an absolute-scale comparison, not a relative multiplier. It should not be read as a percentage unless converted, and no such conversion is needed to interpret the posted estimate.

The risk difference is often useful because it expresses the observed separation between groups on the outcome’s original probability scale. Its estimate depends on both group risks. In general, the same relative contrast can correspond to different absolute contrasts in settings with different baseline risks. The registry comparison also reports a control risk of 0.021 and a treatment risk of 0.097; the latter is not repeated as a rounded value because the registry’s precision exceeds the display limit used for this explanation.

How do the relative measures differ?

In general, a risk ratio divides the treatment risk by the control risk. A ratio above one indicates a higher observed risk in the treatment arm under the stated comparison direction. The reported risk ratio is 4.612, with a confidence interval from 1.037 to 20.51. An odds ratio compares the odds of an event rather than the risks themselves. The reported odds ratio is 5.0, with a confidence interval from 1.066 to 23.44. Odds and risks are related but are not identical measures, so the odds ratio should not be described as a risk ratio.

Both relative estimates are above one, consistent with the direction of the counts and risk difference. Their intervals are broad, indicating that the plausible relative effect sizes cover a wide range. In general, relative measures can appear large when the control risk is low; they should be considered alongside the absolute-scale estimate and the underlying event counts. These measures are descriptive comparisons in the posted results and do not by themselves establish causation.

What does the p-value say?

The reported p-value is unadjusted and equals 0.025. In general, a p-value is calculated under a null hypothesis and a statistical model, and it indicates how often data at least as incompatible with that null would arise if the model’s assumptions held. It does not describe the probability that a treatment effect is real, nor does it indicate the probability that the findings will replicate. Interpretation also depends on whether the test was chosen in advance and whether multiple outcomes or analyses were considered. Those aspects are not reported in the registry record.

The registry does not state which test or model generated the comparison, whether covariates were included, or whether the p-value was corrected for multiplicity. Accordingly, the result should be identified as the posted unadjusted p-value, not assigned to a specific test or adjusted analysis that the record does not describe.

Why do event counts and assumptions matter?

In general, estimates based on small numbers of observed events can be imprecise. Each participant with an event contributes to the numerator of a group risk, so a small number of outcomes can make estimates sensitive to the observed cases. Confidence intervals express sampling uncertainty under the method used to calculate them; they do not account automatically for missing data, measurement error, bias, or uncertainty about the endpoint definition.

Over-dispersion is a general concern when data vary more than a statistical model assumes. For binary outcomes, the registry record does not report an over-dispersion model or a need for one. The stated summary is a participant count for a binary endpoint, so there is no basis in the record to claim that a particular dispersion adjustment was used. Likewise, no information is given about interim analyses, stopping rules, or sample-size justification. The trial’s original statistical plan is not reported in the registry record.

Randomization generally aims to make assigned groups comparable in expectation, but it does not guarantee balance in every realized trial, particularly for factors that are not measured or reported. The record does not provide baseline characteristics, details of allocation implementation, adherence, or analysis populations. Those omissions make it impossible to assess these features from the posted summary alone.

10. Confidence Intervals and Precision

Confidence intervals describe uncertainty around an estimate under the statistical method used to construct them. In general, they show a range of values compatible with the observed data and model at the stated confidence level. They are not a guarantee that the true effect lies inside the interval, and they do not include every source of uncertainty, such as missing outcomes or bias in measurement.

The registry reports a risk difference of 0.076 and a confidence interval with a lower endpoint of 0.012; both recorded limits are positive. This is consistent with the reported direction of the group counts. The interval’s breadth matters: the point estimate is only one value among a range of plausible absolute differences under the calculation method. The record does not state the confidence level or interval-construction method, so those details should not be assumed.

The reported risk ratio is 4.612, with a confidence interval from 1.037 to 20.51. The reported odds ratio is 5.0, with a confidence interval from 1.066 to 23.44. These intervals extend over a wide range of relative effects. Their lower limits are close to one, while their upper limits are much larger, so the estimates do not pin down the magnitude precisely. This uncertainty is consistent with the relatively small number of participants recorded with the endpoint, especially in the control arm.

In general, an interval that excludes the null value can coexist with considerable uncertainty about effect size. For a risk difference, the null is zero; for a ratio measure, the null is one. Interval interpretation should therefore consider both whether the interval crosses the null and how wide it is. The registry’s reported comparison has positive limits for the risk difference and limits above one for the reported ratios, but the broad ratio intervals still leave substantial uncertainty about the size of the association.

These intervals do not resolve questions about endpoint adjudication, missing outcome data, multiplicity, or the analysis population. Those details are not reported in the registry record. Statistical precision is only one part of interpreting a trial result.

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 sample-size illustrations are planning calculations based on the observed effect, not a record of the trial’s original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect targeted in each illustration is the stated fraction of the observed effect. These are participants per arm, not total event counts.

For planning based on 100% of the observed effect, the illustrations list 150 participants per arm for 80% power and 200 participants per arm for 90% power. For planning for 75% of the observed effect, they list 227 participants per arm for 80% power and 303 participants per arm for 90% power. For planning for 50% of the observed effect, they list 417 participants per arm for 80% power and 558 participants per arm for 90% power.

In general, power is the chance that a planned study will detect a specified effect under the assumptions used for its calculation. The 80% and 90% columns show different target chances for detecting the selected effect if the planning assumptions hold. A higher target power generally requires a larger planned sample. The rows also illustrate that smaller effects need more participants: detecting a smaller fraction of the observed effect requires larger planned group sizes in these calculations.

These figures are not observed power. They are planning illustrations based on the effect observed in this trial and may not predict the requirements of another study if its population, outcome frequency, design, or analysis differs. The original sample-size assumptions and justification are not reported in the registry record.

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 = 0.021; treatment proportion p₁ = 0.097; average p̄ = 0.059.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3334 = 0.6535.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3290 = 0.2769.
4. n = (0.6535 + 0.2769)² / (0.0760)² = 149.7, rounded up to 150 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 <- 0.0210526315789474                          # control proportion
p1 <- 0.0970873786407767                          # 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(1837069)
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 = 150 (80% power), 200 (90% power)
 75% of observed effect: n per arm = 227 (80% power), 303 (90% power)
 50% of observed effect: n per arm = 417 (80% power), 558 (90% power)
power.prop.test: 149.7 
Simulated power at n = 150 per arm: 0.837 (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 registry lists a primary composite endpoint and one secondary outcome, a modified composite of cardiovascular events. It also reports four outcome measures, but the posted analysis count is zero. The record does not explain how the primary and secondary outcomes relate to the four measures, whether additional analyses were planned, or which comparisons were prespecified.

In general, multiplicity arises when several outcomes, time points, subgroups, or statistical tests are examined. If many tests are conducted, the chance of at least one apparently unusual result can increase. Methods such as prespecifying a primary outcome or adjusting tests can address particular multiplicity concerns, but the registry record does not report whether any such strategy was used. The p-value of 0.025 is therefore described as unadjusted; it should not be interpreted as evidence that multiplicity was accounted for.

Missing-data handling is not reported in the registry record. The group sizes and endpoint counts are posted, but the record does not say whether every participant had endpoint information, how unavailable outcomes were treated, or whether any participants were excluded from the comparison. The analysis population is also not described. It is not possible to determine from the registry summary whether the comparison followed an intention-to-treat approach, a per-protocol approach, or another approach.

The record does not report a statistical analysis plan, covariate adjustment, subgroup analyses, interim analyses, stopping criteria, or sensitivity analyses. The trial is listed as terminated, but the reason for termination and its relationship, if any, to the analysis is not reported. These omissions do not establish that such methods were or were not used; they mean the registry record does not provide enough detail to evaluate them.

These issues matter because the reported group comparison is only one part of a full analysis. Prespecification, handling of missing outcomes, and clear analysis-population definitions help readers assess whether the reported estimate answers the intended question and how robust it may be to analytic choices.

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 0% reporting threshold.

The registry lists zero serious adverse events and zero other adverse events in the treatment arm, and zero serious adverse events and zero other adverse events in the control arm. It also lists zero deaths in each arm. The adverse-event record gives the arm-specific numbers at risk as 103 for treatment and 95 for control. No serious adverse-event terms are listed.

These entries describe what the registry reports; they do not establish that no adverse events occurred outside the recorded categories or reporting period. The adverse-event timeframe and ascertainment procedures are not reported in the registry record. The record also does not provide details about severity definitions, event monitoring, attribution, or completeness of reporting.

Zero recorded events in a group should not be interpreted as proof that the underlying risk is zero. In general, when events are uncommon, a study may record none even when some risk remains. The available safety entries also do not establish whether the groups were comparable in follow-up or reporting. They should be read as registry counts, not as a comprehensive safety assessment or a recommendation about treatment.

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

The posted results show that the primary composite endpoint was recorded for 10 participants in the treatment arm and 2 in the control arm during the 30-day timeframe. The reported risk difference is 0.076, with a risk ratio of 4.612 and an odds ratio of 5.0. The unadjusted p-value is 0.025. Together, these statistics describe a higher observed endpoint frequency in the treatment arm in this registry comparison.

The result does not establish why that difference occurred. Randomization is a feature of the design, but the registry summary does not report enough detail about allocation, adherence, missing outcomes, endpoint assessment, or analysis populations to evaluate all potential influences on the comparison. Nor does the composite result show which component events contributed to the counts. Component-specific results are not reported in the registry record.

The confidence intervals indicate that the effect size is uncertain. The relative-effect intervals are wide, and the risk-difference interval also spans a range of positive values. A p-value is not a measure of clinical importance, and statistical evidence in one trial does not by itself show that the same estimate would be observed in another population or setting.

The result also does not answer questions about longer-term outcomes, individual components, or the modified composite secondary outcome. A secondary outcome is listed, but the registry record does not include a posted analysis for it. The adverse-event entries report zero events in the recorded categories, but do not supply the monitoring and follow-up detail needed for a complete safety interpretation.

Accordingly, the most defensible reading is descriptive: the registry reports more primary endpoint participants in the treatment arm than in the control arm, with the stated unadjusted comparison and substantial uncertainty about the magnitude. The result alone is not a basis for medical advice or a conclusion that the treatment strategy does or does not work.

16. Why This Trial Matters Statistically

This trial illustrates why reporting an effect estimate alongside event counts is important. The counts make the observed data concrete, while the risk difference and relative measures express the comparison on different scales. Considering both absolute and relative measures helps avoid relying on a single statistic, especially when event counts are small.

It also demonstrates the distinction between a point estimate and its precision. The reported risk ratio and odds ratio point in the same direction as the risk difference, but their confidence intervals are broad. A single p-value cannot convey that uncertainty or describe the full range of effect sizes compatible with the data. Readers need the estimates, their intervals, and the underlying group results together to interpret the strength and limitations of a comparison.

The composite endpoint raises a further statistical issue: a single summary can combine events that may differ in frequency and clinical meaning. Without component-specific results, the composite count cannot show which events drove the observed difference. The registry’s listing of a modified composite as a secondary outcome also makes clear why prespecified outcome hierarchies and multiplicity handling matter; those details are not reported in the registry record.

Finally, the sample-size illustrations show how assumptions about a target effect and desired power translate into planning requirements. They are not evidence of the trial’s original design rationale and are not observed power. Separating planning calculations from observed results is essential when interpreting small-event trials and considering what future evidence would need to address.

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.

Several trial-specific limitations arise from what is and is not recorded. The study is listed as terminated, but the reason is not reported in the registry record. The registry also does not report the original sample-size justification, detailed eligibility criteria, or the full protocol. The listed enrollment and arm sizes describe the record, but do not by themselves reveal how many participants had complete endpoint follow-up or entered the analysis.

The primary outcome combines death, myocardial infarction, stroke, transient ischemic attack, myocardial necrosis, and venous thromboembolism. The registry reports the participant count for the composite but not component-specific counts, event adjudication procedures, or whether recurrent events were handled separately. The outcome timeframe is 30 days, so the posted primary result does not describe longer-term outcomes.

The intervention combines lifestyle counseling with medication components, including conditional use of metoprolol and lisinopril. The registry does not report treatment uptake, adherence, deviations, or the degree to which participants received each component. The comparator is labeled standard of care without further detail. This limits the ability to identify which aspects of the strategy are represented by the comparison.

The record lists no masking, but does not state whether outcome assessors were blinded. It does not describe missing-data methods, the analysis population, covariate adjustment, or multiplicity handling. The reported estimates are based on relatively few endpoint participants, and the relative-effect intervals are wide. These statistical and reporting limitations should be considered alongside the observed counts when assessing how precisely and broadly the results can be interpreted.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Risk Factor Control Before Orthopedic Surgery — NCT01837069.
• Registry record last updated 2020-07-27; status last verified 2020-07.
• Results first posted 2020-07-27.
• 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.

21. Record Summary

The registry describes a randomized, open-label, parallel-group phase 4 trial of multifactorial preoperative risk-factor control versus standard care for patients undergoing orthopedic surgery. NYU Langone Health is listed as sponsor. Enrollment is recorded as 198, the status is terminated, and the primary completion date is 2018-04-02. The reason for termination is not reported in the registry record.

The primary outcome is the number of participants experiencing a composite cardiovascular or thromboembolic event within 30 days. The registry records 10 participants with the endpoint in the treatment arm and 2 in the control arm. The reported risk difference is 0.076, the risk ratio is 4.612, the odds ratio is 5.0, and the unadjusted p-value is 0.025. The confidence intervals are broad, particularly for the relative measures.

A modified composite cardiovascular outcome is listed as secondary, but no analysis is posted in the registry’s analysis count. The record does not report a detailed statistical plan, missing-data handling, or analysis populations. Safety entries list zero serious adverse events, zero other adverse events, and zero deaths in each arm. These summaries describe the registry record and do not resolve the limitations in outcome detail or analysis reporting.