Compression and Cold Therapy on the Post-Operative Shoulder: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Compression and Cold Therapy on the Post-Operative Shoulder (NCT00703729), a randomized na study. The primary outcome is Patient Reported Pain on Day 0.

2. Clinical Question

3. Trial Design

The registry identifies this as a randomized, parallel-group controlled trial with no masking. It lists 58 participants enrolled and a completed status. The recorded study start is 2008-06 and primary completion is 2014-12. The registry lists the University of Colorado, Denver as sponsor. No phase is listed beyond the recorded designation NA.

The two randomized groups are named Cold Compression and Ice Wrap. Their arm descriptions include a one-week initial period using one method, followed by use of the other method for the remainder of the study period. The record describes the Cold Compression method as the Game Ready device and the Ice Wrap method as ice bags secured with an elastic wrap. Although both methods appear in each arm description at different times, the registry identifies the study model as parallel. Further details about the treatment schedule, adherence, or how the switch was incorporated into outcome analyses are not reported in the registry record.

The primary outcome summaries include 25 participants in Cold Compression and 21 in Ice Wrap. The posted summaries therefore do not include outcome values for all 58 enrolled participants. The reasons for this, the analysis population, participant flow, and handling of missing outcome measurements are not reported in the registry record. The record lists pain-medication use on Days 1 through 7 as secondary outcomes, while posted numerical results are available for Days 1 through 3.

Random allocation can help create comparable groups at the start of a trial, but it does not remove uncertainty from a small set of observed outcomes. Lack of masking is also relevant when interpreting a self-reported outcome: participants knew which method they were using, and the registry record does not describe masking of outcome assessment. These design features inform interpretation but do not establish the direction or size of any bias.

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 Patient Reported Pain on Day 0, measured over 1 day.

The primary endpoint is patient-reported pain on Day 0, measured on a visual analog scale. The registry describes a scale 10 cm in length, anchored by no pain at the left and extreme pain at the right. Scores were recorded to the closest millimeter and expressed on a 0-100 scale. The primary measure is a mean, with standard deviation used to describe the spread of individual scores within each group.

Pain is subjective, so asking participants to report it captures their own experience rather than relying only on an external observation. A continuous score preserves gradations across the scale and allows group means to be compared. In general, a mean describes the central level of scores, while a standard deviation describes how dispersed individual scores are around that mean. Neither statistic shows the full distribution of participants’ responses.

The posted mean was 30.2 in Cold Compression and 24.2 in Ice Wrap. The unadjusted recalculated mean difference, with Ice Wrap compared against Cold Compression, was -6.0. That direction indicates a lower observed group mean in Ice Wrap for this endpoint; it does not by itself establish a reliable difference between the regimens. The confidence interval and the registry’s lack of a reported sponsor analysis are important when judging how much can be concluded from these summaries.

7. Results Posted in the Registry

• Ice Wrap (IW): mean 24.20 versus 30.20 on control; difference -6.00 (95% CI -24.59 to 12.59), standardized effect d = -0.19; z-test p = 0.527.

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.

Use of Pain Medication on Day 1

• 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:
  • Ice Wrap (IW): mean 37.30 versus 38.60 on control; difference -1.30 (95% CI -14.72 to 12.12), standardized effect d = -0.06; z-test p = 0.849.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ice Wrap (IW) and the control. The interval gives the range of effects that remain compatible with the data.

Use of Pain Medication on Day 2

• 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:
  • Ice Wrap (IW): mean 33.20 versus 38.00 on control; difference -4.80 (95% CI -20.35 to 10.75), standardized effect d = -0.18; z-test p = 0.545.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ice Wrap (IW) and the control. The interval gives the range of effects that remain compatible with the data.

Use of Pain Medication on Day 3

• 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:
  • Ice Wrap (IW): mean 28.90 versus 30.50 on control; difference -1.60 (95% CI -17.61 to 14.41), standardized effect d = -0.06; z-test p = 0.845.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Ice Wrap (IW) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists medication use on Days 1 through 7 as secondary outcomes. Numerical summaries and unadjusted recalculated comparisons are posted for Days 1 through 3. Across those posted days, the Ice Wrap group’s mean medication-use values are numerically lower than the Cold Compression group’s values, but the comparisons are imprecise and do not establish a clear group difference. Results for Days 4 through 7 are not reported in the registry record.

These medication-use outcomes provide related context to the primary patient-reported pain score, but they measure a different aspect of postoperative experience. Their general direction is consistent with the primary point estimate in the sense that the Ice Wrap group has lower posted means, but this pattern does not resolve the primary interval’s uncertainty or establish a meaningful relationship between the outcomes. The record does not report an analysis connecting pain scores and medication use.

The secondary results are unadjusted, and the registry does not report a multiplicity strategy. In general, examining multiple outcomes creates more opportunities for an apparently notable result to occur by chance, even when no underlying effect exists. This makes isolated secondary findings less persuasive than a prespecified primary result and calls for caution when interpreting a pattern across endpoints. The posted secondary summaries are best treated as exploratory context rather than independent confirmation of the primary comparison.

The record also does not report a secondary-outcome analysis plan, missing-data methods, or analysis populations. The available summaries describe observed group means and variability, but they cannot establish how unreported days or missing measurements were handled. The pattern is descriptive and should not be extended to the days for which numerical results are not reported.

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

• Use of Pain Medication on Day 1
• Use of Pain Medication on Day 2
• Use of Pain Medication on Day 3
• Use of Pain Medication on Day 4
• Use of Pain Medication on Day 5
• Use of Pain Medication on Day 6
• Use of Pain Medication on Day 7

9. Statistical Interpretation

10. Statistical Methods Explained

The statistical quantities in the record answer different questions. A group mean describes the average observed score in that group. A standard deviation describes the spread of individual observations around the group mean. A mean difference compares those group averages, with its sign depending on which group is subtracted from which. A confidence interval expresses uncertainty around the estimated difference. The p-value assesses compatibility with a specified null model under the assumptions of the test. These measures complement one another; none is a substitute for the others.

What does the mean difference describe?

The primary comparison is expressed as Ice Wrap minus Cold Compression. With means of 24.2 and 30.2, respectively, the unadjusted recalculated difference is -6.0 points on the 0-100 scale. A negative value means the observed average was lower in Ice Wrap for this calculation. It does not indicate whether the difference is important to participants, because a threshold for clinical importance is not reported in the registry record. It also does not establish a causal effect by itself; interpretation depends on the trial design, outcome availability, and analysis assumptions.

For a continuous outcome, comparing means is a direct way to summarize how average scores differ. But two groups can have the same mean and different distributions, or different means with considerable overlap among individuals. The reported standard deviations, 34.3 in Cold Compression and 30.0 in Ice Wrap, describe considerable within-group spread relative to the difference in means. They do not reveal each participant’s score or the shape of the distributions.

What does the confidence interval add?

A confidence interval combines the point estimate with its estimated sampling uncertainty. In general, a narrower interval indicates greater precision under the calculation’s assumptions, while a wider interval leaves more possible values compatible with the data. The primary interval runs from -24.59 to 12.59. Because zero lies within it, the summaries do not distinguish the estimated difference from no difference at the conventional level associated with this interval. The interval also includes values in either direction, so it does not establish that one group has the lower underlying mean.

Confidence intervals are not guarantees that the true value lies inside a particular interval, and they do not account automatically for problems such as missing data, departures from model assumptions, selective reporting, or multiple comparisons. Those concerns require information about study conduct and analysis that is not reported in the registry record.

How should the p-value be read?

The primary unadjusted p-value is 0.527. In general, a p-value is calculated under a null hypothesis and a statistical model. It measures how compatible the observed data are with that null model, assuming the model and analysis are appropriate. It is not the chance that the null hypothesis is true, and it does not quantify the likelihood that a result will replicate. The unadjusted designation matters because the record does not report a multiplicity adjustment or a sponsor-reported inferential analysis.

The p-value should not be used as a decision rule detached from the estimated difference and interval. A relatively large p-value does not prove that two regimens are equivalent. Equivalence or non-inferiority conclusions require a prespecified margin and an analysis designed for that question; such a margin or analysis is not reported in the registry record.

What is the status of the comparison?

The arm summaries are registry-posted means, standard deviations, and participant counts. The comparison statistics are simple unadjusted recalculations from those summaries. The registry lists zero reported analyses, and no sponsor analysis is reported in the registry record. The record does not describe a model, adjustment variables, handling of the schedules, or an analysis population. The mean-difference calculation should therefore be understood as a transparent summary-based comparison, not as a claim that it reproduces an unreported protocol analysis.

Because the outcome is continuous and the comparison uses group means, the calculation concerns an average difference rather than a difference in the proportion of participants meeting a threshold. No threshold-based responder result is reported. The score’s 0-100 range provides the measurement scale, but it does not specify a minimal important difference. That interpretive benchmark is not reported in the registry record.

Why do the assumptions matter?

Unadjusted calculations are useful for reading the posted summaries, but their validity depends on assumptions about the data and the comparison. In general, methods based on means often rely on the sampling distribution of the mean difference being adequately approximated by the method used. The record does not report diagnostic checks or alternative analyses. The posted standard deviations summarize observed variability, but they do not establish that every statistical assumption holds.

The random allocation and parallel-group designation are important design information, while the absence of masking and the sequential use described in the arm schedules add context to interpretation. The record does not explain whether the outcome measure was taken at a uniform point relative to treatment use or how the sequence was incorporated into analysis. These gaps do not change the arithmetic of the posted summary comparison, but they limit what can be inferred about the trial’s complete analysis and conduct.

11. Confidence Intervals and Precision

The primary estimate is a mean difference of -6.0 points, with a 95% confidence interval from -24.59 to 12.59. The interval includes zero, which represents no difference in group means under the scale of this comparison. It also includes negative and positive values. Thus, the data do not precisely locate the underlying mean difference or rule out either direction.

Precision is distinct from the point estimate. The estimate is a single summary of the observed difference, while the interval reflects sampling uncertainty under the calculation’s assumptions. The interval is broad relative to the estimate, so the observed means alone do not support a narrow conclusion about the size of any population difference. The interval should not be read as proof that every value within it is equally plausible, nor as a guarantee about the true value.

The outcome summaries include 25 participants in Cold Compression and 21 in Ice Wrap. Smaller outcome groups and substantial variability can limit precision, although the record does not provide enough information to isolate the contribution of each factor. The standard deviations are 34.3 and 30.0, indicating spread in individual scores within both groups. Group averages do not show how individual scores overlap, and the standard deviations do not describe the full distributions.

The confidence interval is a simple unadjusted recalculation from posted summaries. It does not include uncertainty about missing outcomes, the analysis population, possible protocol deviations, or unreported choices in the sponsor’s analysis. Those issues cannot be assessed from the registry record. The interval is therefore informative about the summary-based mean comparison, but it should not be treated as a complete account of all uncertainty in the study.

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 sample-size figures are planning illustrations based on the observed primary effect, not the trial’s original design calculation. They are participants per arm, not a combined total and not total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The target effect is the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column gives the planned participants per arm for a design that would have an 80% chance of detecting the specified target effect under the planning assumptions, if that effect were the true effect and the assumptions held. The 90% power column gives the corresponding per-arm count for a higher chance of detecting that target effect. Power is a property of a proposed design under specified assumptions, not a probability that the observed result is correct.

Rows planning for 100%, 75%, and 50% of the observed effect use progressively smaller target effects. Smaller effects are harder to distinguish from random variation and therefore require more participants. The curve expresses the same planning relationship: the participant requirement increases as the target effect gets smaller. These figures are illustrations tied to the observed effect and stated statistical assumptions, not evidence about what the original investigators planned. They are not observed power.

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 = -6.000; pooled standard deviation σ = 32.416.
2. n = 2 × (1.96 + 0.842)² × 32.416² / -6.000² = 458.2, rounded up to 459 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 <- -6                      # observed difference in means
sigma <- 32.4162417203587                     # 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(703729)
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 = 459 (80% power), 614 (90% power)
 75% of observed effect: n per arm = 815 (80% power), 1,091 (90% power)
 50% of observed effect: n per arm = 1,833 (80% power), 2,454 (90% power)
power.t.test: 459.2 
Simulated power at n = 459 per arm: 0.798 (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 identifies one primary outcome and lists pain-medication use on Days 1 through 7 among the secondary outcomes. It also reports 15 outcome measures overall, but no reported analyses are listed. Numerical secondary summaries are available for Days 1 through 3; results for the other listed medication-use days are not reported in the registry record. A full analysis plan, including which outcomes were confirmatory and how they were ordered or tested, is not reported in the registry record.

Multiplicity arises when many outcomes or comparisons are examined. In general, the more unadjusted tests conducted, the greater the opportunity for at least one result to appear statistically notable by chance. The posted comparison p-values are unadjusted recalculations, and no adjustment procedure is reported in the registry record. This does not invalidate the descriptive estimates, but it means secondary p-values should not be read as if they were protected from the consequences of multiple testing.

Missing-data methods are not reported in the registry record. The study enrollment is 58, while the primary outcome summaries list 25 participants for Cold Compression and 21 for Ice Wrap. Safety denominators are also reported separately. The record does not explain why the primary endpoint counts differ from enrollment, how many measurements were unavailable, whether participants were excluded, or whether any imputation or sensitivity analysis was used. The effect of missing outcomes on the comparison therefore cannot be assessed.

The analysis population is not reported in the registry record. It is not stated whether the primary summaries include all randomized participants, participants with available Day 0 scores, or another defined group. The record also does not report whether participants were analyzed according to their randomized assignment when treatment use changed over time. These details matter for interpreting a randomized comparison and cannot be reconstructed from the group means alone.

15. Safety Results

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

Other (non-serious) adverse events

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

The posted serious adverse event figures are zero affected participants among 25 at risk in Cold Compression and zero among 24 at risk in Ice Wrap. The registry also records zero participants affected by other adverse events among 25 at risk in Cold Compression and among 24 at risk in Ice Wrap. These are counts for the denominators recorded in the safety data.

Zero recorded events in these groups means no affected participants are listed in the posted figures. It does not establish that an event is impossible or that the regimens have identical safety profiles. With limited numbers of participants, uncommon events may not be observed. The record does not provide further details about event ascertainment, follow-up completeness, or how events were attributed.

Deaths are not reported in the registry record. The safety denominators differ from the posted primary-outcome counts, and the record does not explain the difference. These safety counts are descriptive and should not be interpreted as a complete assessment beyond the recorded event categories and denominators.

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

The posted primary result is an unadjusted mean difference of -6.0 points on the 0-100 pain scale for Ice Wrap compared with Cold Compression on Day 0. The 95% confidence interval, -24.59 to 12.59, includes no difference and values in both directions. The data therefore do not establish a clear difference between the groups, but they also do not demonstrate equivalence. A conclusion of equivalence would require an appropriate prespecified margin and analysis, neither of which is reported in the registry record.

The point estimate describes the observed average difference in the available summaries. It does not show that every participant experienced a similar score or change. The group standard deviations indicate substantial variation in scores, and the registry does not provide individual-level distributions or a threshold-based result. The registry also does not report a minimal important difference for interpreting the score scale, so the clinical importance of the point estimate cannot be determined from the record alone.

Random allocation is a design feature intended to support a fair comparison between groups. However, the outcome counts are lower than enrollment, and the record does not describe the analysis population or missing-data handling. No masking was used, and the arm descriptions include sequential use of both methods. These details limit the extent to which the summary comparison can be interpreted as a complete account of treatment assignment and exposure.

The medication-use summaries for the posted secondary days are descriptive and do not settle the uncertainty in the primary outcome. The registry record does not report numerical results for all listed secondary days or a multiplicity plan. Safety figures list no affected participants in the reported categories, but those counts do not establish the absence of risk.

Overall, the posted results are a limited set of group summaries and simple unadjusted comparisons. They describe the observed data and their statistical uncertainty; they do not warrant medical advice, a claim that one regimen works or fails, or a conclusion that the regimens are interchangeable.

17. Why This Trial Matters Statistically

This trial offers a useful statistical example of why an estimated difference, its uncertainty, and the available analysis details need to be read together. The primary point estimate has a direction, but the confidence interval is broad and includes zero. Focusing only on the point estimate would overstate what the data establish; focusing only on the p-value would omit both the magnitude and precision of the comparison.

The record also illustrates the distinction between enrollment and outcome-specific sample sizes. Although enrollment is 58, the primary summaries include 25 and 21 participants in the respective groups. The record does not explain why. That gap matters because a reader cannot assess the analysis population, missingness, or the effect of unavailable measurements from arm means alone.

There are several layers of uncertainty: the sampling variability reflected in the interval, unreported details about the analysis and missing data, and the context of multiple listed outcomes. The secondary results offer a descriptive pattern but are unadjusted, and numerical results are not reported for all the listed days. These features show why secondary findings should be interpreted cautiously rather than treated as independent confirmation.

The sample-size planning illustration also demonstrates that required sample size depends on the target effect and design assumptions. It is not a retrospective measure of whether this completed trial had adequate power, and the original design assumptions are not reported in the registry record. Keeping planning calculations separate from observed results is essential for sound interpretation.

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 primary endpoint summaries cover 25 Cold Compression participants and 21 Ice Wrap participants, while the registry reports enrollment of 58. The reason for the difference is not reported in the registry record. Without participant flow, missing-data information, or a defined analysis population, it is not possible to determine how representative the available outcome summaries are of all randomized participants.

The registry records no masking. This is relevant because the primary outcome is self-reported pain, and participants knew which method they were using. The registry does not report whether outcome assessors were masked or how that was handled. The arm descriptions also state that both methods were used in sequence, but the record does not explain how timing, adherence, or the transition between methods was assessed in relation to the Day 0 endpoint.

No sponsor-reported analyses are listed. The statistical comparisons are unadjusted recalculations from posted summary statistics, and the record does not report an analysis plan, covariate adjustment, sensitivity analyses, or multiplicity procedure. The primary confidence interval is informative for the simple mean comparison, but cannot address unreported design or analysis choices.

Secondary medication-use results are posted for Days 1 through 3, while the registry lists outcomes through Day 7. Results for the remaining listed days are not reported in the registry record. The record also does not provide a minimal important difference for the pain scale, so the practical importance of the mean difference cannot be judged using a stated clinical threshold. These limitations constrain interpretation without changing the numerical summaries that are posted.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Compression and Cold Therapy on the Post-Operative Shoulder — NCT00703729.
• Registry record last updated 2015-09-10; status last verified 2015-08.
• Results first posted 2015-09-10.
• 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

The registry record describes NCT00703729, titled Compression and Cold Therapy on the Post-Operative Shoulder. It is recorded as a completed randomized, parallel-group trial with no masking and 58 enrolled participants. The recorded study start is 2008-06, and primary completion is 2014-12. The sponsor is the University of Colorado, Denver.

The primary outcome is patient-reported Day 0 pain on a 0-100 visual analog scale. The posted group means are 30.2 for Cold Compression and 24.2 for Ice Wrap. The unadjusted recalculated mean difference is -6.0, with a 95% confidence interval from -24.59 to 12.59 and an unadjusted p-value of 0.527. The registry record does not report sponsor analyses or the original sample-size assumptions.

Medication-use outcomes are listed for Days 1 through 7, with numerical summaries posted for Days 1 through 3. The posted serious and other adverse event counts are zero in each reported group; deaths are not reported in the registry record. Missing-data handling, analysis populations, and a multiplicity plan are not reported in the registry record.