Tolerability/Safety and Efficacy of Inhaled AZD4818 in Patients With Moderate to Severe Chronic Obstructive Pulmonary Disease (COPD): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Tolerability/Safety and Efficacy of Inhaled AZD4818 in Patients With Moderate to Severe Chronic Obstructive Pulmonary Disease (COPD) (NCT00629239), a randomized phase 2 study. The primary outcome is Number of Patients Experiencing Adverse Events.

2. Clinical Question

3. Trial Design

This was a Phase IIa, randomized, parallel-group study comparing inhaled AZD4818 with placebo in participants with moderate to severe COPD. The official study title describes it as double-blind and placebo-controlled, while the registry lists quadruple masking. The record does not explain which roles were masked or how masking was maintained.

The planned treatment period was 4 weeks. The primary outcome was assessed at all study visits, and the registry describes the outcome as the number of participants who had an adverse event. The record lists an enrollment of 65 and marks the study as completed. The sponsor was AstraZeneca. The recorded study start is 2008-01, and primary completion is recorded as 2008-06.

The two reported arms were AZD4818 and placebo. The registry does not report further details of the allocation procedure, treatment dose, adherence, withdrawals, or the number of participants included in each outcome analysis. Those details should not be inferred from the arm counts alone. The original sample-size justification and the trial's prespecified statistical plan are not reported in the registry record.

Secondary outcomes listed include lung-function measures such as FEV1, FVC, vital capacity, inspiratory capacity, forced expiratory flow, and morning and evening peak expiratory flow, as well as the Clinical COPD Questionnaire total score. Numerical results for these secondary 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

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients Experiencing Adverse Events, measured over At all study visits.

The primary endpoint is the number of participants who experienced an adverse event at any study visit. It is a participant-level event count: a participant is counted as having an event if an adverse event was recorded, rather than the result describing the total number of separate event episodes. The record gives the timeframe as all study visits.

Adverse-event outcomes are used in clinical trials to describe tolerability and safety during the period under study. Counting participants with an event provides a direct comparison of how often at least one event was recorded in each randomized group. It does not describe the full event burden: multiple events in one participant, event timing, duration, severity, or causal relationship to treatment may require separate information.

The posted results report 22 participants with an adverse event in the AZD4818 group and 27 in the placebo group. They also report group sizes of 33 and 32, respectively. The record does not give a more detailed definition of which events qualified for this primary endpoint, nor does it report an analysis population or rules for handling missing visit data. These limitations affect how narrowly the endpoint can be interpreted.

7. Results Posted in the Registry

• AZD4818: 66.7% (22/33) versus 84.4% (27/32) on control; risk difference -17.7 percentage points (95% CI -38.1 to 2.7), risk ratio 0.79 (0.60–1.05); unadjusted two-proportion z-test p = 0.098.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count of participants who experienced an adverse event, with each participant classified according to whether an event occurred. The comparison therefore concerns event proportions in the two randomized groups. The registry gives the counts and group sizes and reports an absolute risk difference, along with relative measures and an unadjusted p-value.

What does the risk difference express?

The risk difference is the event proportion in the AZD4818 group minus the event proportion in the placebo group. A negative estimate means the observed proportion was lower in the AZD4818 group; a positive estimate would mean it was higher. The reported estimate is -0.177. Its interpretation is an absolute contrast in the chance of having at least one recorded event during the study, not a comparison of the total number of events or their severity.

The risk difference is often useful because it stays on the proportion scale and makes the absolute size of a contrast visible. It is also sensitive to the study setting and the underlying event frequency. It should be read alongside its confidence interval rather than as a stand-alone fixed effect. The reported interval is -0.381 to 0.027, so it includes zero and covers differences in either direction.

What do the risk ratio and odds ratio add?

A risk ratio compares the event proportions as a ratio, while an odds ratio compares the odds of an event. The registry reports a risk ratio of 0.790, with an interval from 0.595 to 1.049, and an odds ratio of 0.370, with an interval from 0.112 to 1.227. Each measure presents the same group contrast on a different scale. They are not separate confirmations of the finding. When events are common, an odds ratio can look more extreme than a risk ratio, so these scales should not be treated as interchangeable.

How should the p-value be understood?

The unadjusted p-value is 0.098. In general, a p-value describes how unusual data at least as incompatible with a specified null model would be if that model were correct and the analysis assumptions held. It is not the probability that the null model is true, and it does not quantify the clinical importance of the observed difference. The record does not report an adjusted p-value or a multiplicity procedure. The p-value is therefore best read as a single unadjusted result, not as a complete account of uncertainty.

For a binary outcome, the precision of a group comparison depends on the number of participants and on how events are distributed across groups. Sparse counts can make estimates less stable, especially for serious-event comparisons. In general, confidence intervals communicate this uncertainty more directly than a threshold-based label. The registry's group counts are modest, and its interval for the primary difference is broad enough to include a range of plausible contrasts.

What is not documented about the analysis?

The registry record lists no analysis entries. It does not report the exact statistical model, a prespecified analysis population, covariate adjustment, missing-data methods, or whether participants were analyzed according to randomized assignment. The posted group comparison can be described from the reported counts and estimates, but those omissions prevent a detailed audit of the analysis procedure. The trial's original statistical plan is not reported in the registry record, so no unlisted modeling or monitoring choices should be assumed.

Randomization supports a fair comparison in principle by assigning treatment independently of participant characteristics, while masking can reduce some forms of biased assessment or care. The record identifies the study as randomized and quadruple-masked and describes the official title as double-blind. It does not provide the mechanics or roles needed to judge how allocation and masking were implemented. These design labels are useful context but do not replace information about execution.

The primary endpoint is an event indicator over all study visits. If participants had different amounts of follow-up or missed visits, a simple participant count could be affected by exposure time or incomplete ascertainment. General statistical practice would examine the completeness of follow-up and define how incomplete records are treated. Those details are not reported in the registry record, so the posted comparison cannot establish whether such issues affected the estimate.

10. Confidence Intervals and Precision

The 95% confidence interval for the risk difference is -0.381 to 0.027, around a point estimate of -0.177. A confidence interval describes the uncertainty associated with an estimate under the chosen method and its assumptions. In repeated use of the same procedure, intervals constructed this way would contain the underlying value at the stated rate. It is not a probability statement that the particular underlying difference lies inside this particular interval.

The interval includes zero, the value representing no difference in event proportions. It also includes negative values, consistent with a lower event proportion in the AZD4818 group, and a positive value, consistent with a higher event proportion. This range shows why the point estimate alone is not enough to characterize the evidence. The data do not pin down the direction and magnitude with high precision.

The relative estimates also have intervals that include their no-difference value: the risk ratio is 0.790 with an interval of 0.595 to 1.049, and the odds ratio is 0.370 with an interval of 0.112 to 1.227. Different scales can make the same uncertainty look different, so the absolute risk-difference interval remains important for understanding the size of the group contrast.

In general, interval width reflects both the amount of information and the variability inherent in the outcome and analysis. A small trial and a short observation period can yield uncertain estimates, particularly for safety outcomes that are not frequent. The record does not report the precision of secondary outcomes or serious-event comparisons beyond the stated summaries. Confidence intervals describe statistical uncertainty, not the possibility of unreported biases or incomplete outcome recording.

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

The 80% column means participants per arm for 80% power under those planning assumptions; the 90% column means participants per arm for 90% power. Power is the probability, under a particular assumed effect and statistical model, of detecting an effect at the selected significance level. A higher target power generally requires more participants for the same assumed effect.

The rows vary the effect being targeted: planning for 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect. Smaller effects require more participants to detect, because they are harder to distinguish from random variation. These estimates are planning illustrations and are not observed power. They do not show what the completed trial was capable of detecting under its actual design, nor do they establish that a future study would attain the assumed effect or operating characteristics.

The record's planning note also distinguishes participant-based calculations from time-to-event calculations, which would be expressed as total events. This primary outcome is a participant event count, so the relevant planning quantities are per arm. The registry does not report the sample-size assumptions used when this trial was designed.

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 = 0.844; treatment proportion p₁ = 0.667; average p̄ = 0.755.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6081 = 1.1918.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5950 = 0.5008.
4. n = (1.1918 + 0.5008)² / (-0.1771)² = 91.4, rounded up to 92 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 <- 0.84375                          # control proportion
p1 <- 0.666666666666667                          # 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(629239)
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 = 92 (80% power), 122 (90% power)
 75% of observed effect: n per arm = 153 (80% power), 205 (90% power)
 50% of observed effect: n per arm = 320 (80% power), 428 (90% power)
power.prop.test: 91.4 
Simulated power at n = 92 per arm: 0.812 (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 13 secondary outcomes in addition to the primary outcome, including lung-function measures and the Clinical COPD Questionnaire total score. Numerical secondary results are not reported in the registry record. The record lists no analysis entries, and it does not describe a multiplicity strategy for assessing several outcomes or multiple time points.

In general, testing many outcomes can increase the chance that at least one result appears noteworthy by chance. Researchers may address this by prespecifying a hierarchy, adjusting significance thresholds, or treating secondary analyses as exploratory. Which approach, if any, was planned for this trial is not reported in the registry record. The primary outcome's unadjusted p-value should therefore not be interpreted as accounting for tests of other outcomes.

The registry does not report the analysis populations, whether the primary analysis followed randomized assignment, or whether any participants were excluded from an outcome analysis. It also does not describe missing-data handling, follow-up completeness, or sensitivity analyses. These details matter because event status may be unknown when participants miss visits or withdraw, and different analysis choices can affect estimates.

The record does not report a prespecified statistical analysis plan, interim analyses, or stopping rules. Their absence from the registry record does not establish that these procedures were or were not used; it means their details cannot be assessed from the posted information. No assumptions about adjustment, missing outcomes, or analysis populations should be added to the interpretation of the reported comparison.

14. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: deep venous thrombosis left lower leg (3.0%), copd exacerbation (3.0%).

The registry lists serious adverse events separately from the primary count of participants with any adverse event. It reports 2 serious adverse events in the AZD4818 group, for which 33 were at risk, and 0 in the placebo group, for which 32 were at risk. These are descriptive counts in a small study; the record does not provide a comparative estimate or confidence interval for serious events.

The top serious-event terms listed are Deep Venous Thrombosis Left Lower Leg and COPD Exacerbation, each reported as 3.0%. The record does not provide enough detail to assign those terms to a particular group from the listed summary, or to infer causality. Death counts are not reported in the registry record.

Serious adverse events and the primary endpoint are distinct summaries. The primary endpoint includes any adverse event, whereas the serious-event listing is limited to events classified as serious. Event counts alone do not describe severity beyond that classification, timing, duration, or whether an event was judged related to treatment. The short 4-week study period also limits what these data can say about longer-term safety.

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

The posted primary results show fewer participants with a recorded adverse event in the AZD4818 arm than in the placebo arm: 22 of 33 compared with 27 of 32. The estimated risk difference is -0.177, but its confidence interval, -0.381 to 0.027, includes no difference and values on either side. The unadjusted p-value is 0.098. Taken together, these results describe an uncertain estimate, not a definitive separation between the groups.

These findings do not establish that AZD4818 reduces adverse events, and they do not establish that the groups have equivalent safety. The interval allows for a range of possible group differences, while the modest group sizes and short observation period limit precision. The serious-event counts are also sparse and should be treated descriptively rather than as a reliable comparison of specific risks.

The endpoint captures whether a participant experienced an event during the study, not the number of event episodes or their overall clinical burden. The registry does not report details such as the timing and severity of each event, causal attribution, missing-data handling, or analysis populations. Those limits constrain interpretation of both the primary and serious-event summaries.

The study also listed lung-function and questionnaire outcomes, but their numerical results are not reported in the registry record. The primary adverse-event comparison therefore cannot answer whether lung function or symptoms changed. These results concern this randomized 4-week trial and should not be extended to longer treatment or other populations without further evidence.

16. Why This Trial Matters Statistically

This trial illustrates why the direction of a point estimate should not be separated from its uncertainty. The observed adverse-event count was lower in the AZD4818 group, and the absolute estimate favors that group. However, the confidence interval includes zero and reaches a positive value, so the posted result does not precisely determine the direction of the underlying difference.

It also shows why outcome scale matters. The risk difference expresses an absolute contrast, while the risk ratio and odds ratio express relative contrasts. These measures are related but not interchangeable, particularly when events are common. A clear statistical interpretation identifies the scale, direction, and uncertainty rather than treating a single estimate or p-value as a complete conclusion.

Safety data require attention to outcome definition and follow-up. The primary result counts participants with any adverse event over study visits, while serious-event listings provide a narrower summary. Neither alone captures every aspect of safety. The registry's short 4-week period, limited group sizes, and absence of detailed missing-data and analysis-population information are relevant when considering how much the results can support.

Finally, registry reporting is part of statistical interpretation. The record documents the randomized, masked, parallel-group design and the primary event counts, but it does not report the original statistical plan or numerical secondary results. Recognizing what is documented—and what is not—keeps conclusions aligned with the evidence available for this trial.

17. Limitations

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

The primary estimate is based on a relatively small study, with 33 participants in the AZD4818 arm and 32 in the placebo arm. The confidence interval is broad and includes zero, limiting precision about the difference in event proportions. The short 4-week treatment period also means the results do not describe longer-term event patterns.

The registry describes the primary outcome as the number of participants with an adverse event at all study visits but does not provide a detailed event definition or participant-level event information. It does not report analysis populations, withdrawals, visit completion, missing-data handling, or exposure time. As a result, the completeness and comparability of event ascertainment cannot be assessed from the registry record.

The record does not report the treatment dose or schedule, randomization implementation, masking roles, or the trial's original statistical plan. It also lists no analysis entries and does not provide numerical results for the secondary outcomes. The serious-event summary is limited, and deaths are not reported in the registry record. These omissions restrict assessment of the study's execution and the range of outcomes.

These limitations concern the information available in the registry record. They do not establish that a particular procedure was omitted during the trial; they mean that its use or details cannot be verified from the posted information.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Tolerability/Safety and Efficacy of Inhaled AZD4818 in Patients With Moderate to Severe Chronic Obstructive Pulmonary Disease (COPD) — NCT00629239.
• Registry record last updated 2010-12-06; status last verified 2010-11.
• Results first posted 2010-12-06.
• 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

NCT00629239 is a completed Phase IIa, randomized, parallel-group, placebo-controlled study of inhaled AZD4818 in participants with moderate to severe COPD. The registry lists quadruple masking; the official title describes the study as double-blind. The treatment period was 4 weeks, and the recorded enrollment was 65.

The primary outcome was the number of participants experiencing adverse events at all study visits. The registry reports 22 participants with events in the AZD4818 group and 27 in the placebo group. The risk difference was -0.177, with a 95% confidence interval from -0.381 to 0.027, and the unadjusted p-value was 0.098. The record also lists serious adverse events separately.

Secondary outcomes include lung-function measures and a COPD questionnaire, but their numerical results are not reported in the registry record. The original design assumptions, statistical plan, analysis populations, and missing-data methods are also not reported.