Combination Antibiotic Treatment for Reactive Arthritis Caused by Chlamydia Bacteria: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Combination Antibiotic Treatment for Reactive Arthritis Caused by Chlamydia Bacteria (NCT00351273), a randomized phase 3 study. The primary outcome is Investigate Whether a 6 Month Course of Combined Antibiotics Was Effective Treatment..

2. Clinical Question

3. Trial Design

This was a randomized, triple-masked, parallel-group phase 3 trial sponsored by the University of South Florida. The registry lists the trial status as completed and enrollment as 42. Participants were assigned to azithromycin and rifampin, doxycycline and rifampin, or placebo. The registry does not describe the specific masking procedures, allocation ratio, or how randomization was implemented; those details are not reported in the registry record.

The condition of interest was reactive arthritis associated with chlamydia, described in the study summary as lasting more than 6 months. The stated treatment period was six months, and the primary outcome timeframe was Month 6. The primary measure was a composite clinical response rather than a single symptom or laboratory result. The registry says that participants had to meet 4/6 clinical criteria to count as responders.

The posted primary outcome counts are reported for 14 participants in the azithromycin-and-rifampin group, 10 in the doxycycline-and-rifampin group, and 10 in the placebo group. These arm counts account for the response figures described in the primary outcome record. The overall enrollment figure and arm counts serve different purposes in the record; the registry does not explain any difference between enrollment and the primary-outcome analysis counts.

Secondary outcomes listed include complete response, erythrocyte sedimentation rate, hsCRP, HAQ DI score, PhGA assessment, and swollen and tender joint counts. The record identifies seven posted secondary outcome measures. Further details about assessment schedules, measurement procedures, and the statistical handling of these 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:

Analysis population for the primary outcome, as defined in the registry: Efficacy and safety analyses were performed on an intent to treat(ITT) basis. Subjects who prematurely withdrew or who were lose to follow up for any reason were included in the ITT population and were considered nonresponders.

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 Investigate Whether a 6 Month Course of Combined Antibiotics Was Effective Treatment., measured over Month 6.

The primary endpoint was a composite measure of response assessed at Month 6. A participant counted as a responder only if they met 4/6 clinical criteria. The registry describes this as a complete response, but the criteria themselves and the way the criteria were combined are not reported in the registry record. The posted counts are 10/14 responders for azithromycin and rifampin, 7/10 for doxycycline and rifampin, and 3/10 for placebo.

A composite endpoint can represent several aspects of a condition in one outcome. In general, this can capture a broader clinical state than a single measure, but its meaning depends on the component criteria, their definitions, and the rule for combining them. Because the registry record does not identify the six criteria, readers cannot determine from this record which elements contributed to each participant’s response or whether all components had equal importance.

The endpoint is useful for comparing the proportion of participants who met a prespecified response definition by the stated assessment time. The registry’s intent-to-treat description says that participants who withdrew early or were lost to follow-up were included and considered nonresponders. This rule makes the treatment-group response counts reflect that stated classification, rather than excluding those participants from the efficacy and safety analyses. More detailed endpoint procedures and the handling of any other missing measurements are not reported in the registry record.

7. Results Posted in the Registry

• Azithromycin & Rifampin: 71.4% (10/14) versus 30.0% (3/10) on control; risk difference 41.4 percentage points (95% CI 4.5 to 78.4), risk ratio 2.38 (0.87–6.49); unadjusted two-proportion z-test p = 0.045.
• Doxycycline & Rifampin: 70.0% (7/10) versus 30.0% (3/10) on control; risk difference 40.0 percentage points (95% CI -0.2 to 80.2), risk ratio 2.33 (0.83–6.54); unadjusted two-proportion z-test p = 0.074.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the posted analysis compare?

The primary outcome is binary: participants either met the composite response definition or did not. For a binary outcome, a risk difference expresses the difference between the response proportions in two groups. The recalculated azithromycin-and-rifampin comparison has a risk difference of 0.414; the doxycycline-and-rifampin comparison has a risk difference of 0.4. These are unadjusted comparisons with placebo, calculated from the posted counts. They describe the observed differences in response proportions, rather than a change in an individual participant’s outcome.

The reported analysis is different in scope. The registry identifies a Fisher Exact analysis involving all three groups and reports p = 0.01. Fisher’s exact test is a method for testing association between group assignment and a categorical outcome, particularly useful when sample counts are limited. In general, an overall test asks whether the outcome distribution differs somewhere among the groups; it does not by itself specify which particular pair differs or estimate the size of each pairwise difference. The record’s reported value of 40.8 is labeled “Mean Difference (Final Values),” but the outcome is described as a responder count. The record does not explain the scale or interpretation of that value.

What does a p-value tell the reader?

In general, a p-value measures how compatible the observed data are with a specified null hypothesis under the assumptions of the test. A small p-value is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry’s p = 0.01 belongs to its reported three-group analysis. The p-values 0.045 and 0.074 belong to the separate unadjusted pairwise recalculations. Because the analyses differ, the p-values should not be read as competing versions of one identical test.

Why distinguish recalculation from the sponsor analysis?

The recalculated comparisons use the posted arm-level counts to produce simple unadjusted estimates. They do not reproduce an unknown model or adjustment strategy. The registry flags a discrepancy, so the sponsor’s own analysis takes precedence. The record does not explain the discrepancy or provide enough detail to reconcile the sponsor’s reported value with the pairwise risk differences. In particular, no inference should be made about the sponsor’s calculation method beyond the reported Fisher Exact label and the listed result.

What do confidence intervals add?

In general, a confidence interval describes a range of values compatible with the data and the method used to calculate it. A wider interval indicates less precision: the plausible effect sizes span a broader range. The recalculated intervals are wide, reflecting the limited amount of information in the relatively small arm counts. The azithromycin interval runs from 0.045 to 0.784; the doxycycline interval runs from -0.002 to 0.802. The latter includes zero, while the former does not. This interval-based description applies to the recalculated pairwise comparisons and does not resolve the discrepancy with the sponsor’s analysis.

What does intent to treat mean in this record?

The registry says efficacy and safety analyses used an intent-to-treat basis. It specifically states that participants who withdrew early or were lost to follow-up for any reason remained in that population and were considered nonresponders. In general, intent-to-treat analysis aims to retain participants in their randomized groups, preserving the comparison created by randomization. The stated nonresponder rule is the record’s method for classifying those participants for the primary response outcome. Details about other missing measurements, deviations from assigned treatment, or additional analysis populations are not reported in the registry record.

What can be said about multiplicity?

The registry lists one primary analysis and seven posted secondary outcome measures. It does not describe whether the sponsor adjusted for testing multiple outcomes or contrasts, whether comparisons were prespecified in a hierarchy, or whether any other multiplicity strategy was used. In general, multiple statistical tests can increase the chance of at least one apparently unusual result when all tests are considered together. That general concern is relevant when reading multiple outcomes, but the record does not allow a conclusion about the actual impact of multiplicity in this trial.

The methods separate general statistical interpretation from the specific record: the primary outcome is a binary composite, the recalculations are unadjusted pairwise risk differences, and the sponsor reports a three-group Fisher Exact result. The registry does not report enough detail to reconstruct a fuller statistical analysis plan.

10. Confidence Intervals and Precision

The recalculated risk-difference confidence intervals show substantial uncertainty about the size of the pairwise response differences. For azithromycin and rifampin versus placebo, the estimate is 0.414, with a 95% confidence interval from 0.045 to 0.784. For doxycycline and rifampin versus placebo, the estimate is 0.4, with a 95% confidence interval from -0.002 to 0.802. Both intervals span a broad range of possible differences, so the point estimates should not be read as precise measures of the effect.

In general, a confidence interval gives a range of values compatible with the observed data under the calculation method. It does not guarantee that the true effect lies within the interval, and it does not capture every source of uncertainty, such as limitations in the endpoint definition or analysis plan. The interval for the doxycycline comparison includes zero, meaning that the recalculated data are compatible with no risk difference as well as with differences in either direction. The interval for azithromycin is above zero throughout, but it remains wide. These statements concern the recalculated comparisons, not the sponsor’s three-group analysis.

Precision is constrained by the small response-analysis groups: 14 participants in the azithromycin-and-rifampin group, 10 in the doxycycline-and-rifampin group, and 10 in the placebo group. With limited group counts, a small change in the number of responses can materially affect an estimated proportion. No additional precision claim can be drawn from the registry’s reported p-value of 0.01, since the record does not provide a confidence interval for that analysis or clarify the scale of its reported value. The sponsor’s own analysis takes precedence because the registry flags a discrepancy.

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 planning figures are illustrative calculations based on the observed effects, not 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 to detect is set to a stated fraction of the observed effect. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

For these binary outcomes, the planning figures are participants per arm. They are not total event counts. The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the selected effect under the stated assumptions. The 90% power column describes the corresponding number for a 90% chance. These are planned detection probabilities under a statistical model, not guarantees about what a future study would find. For example, planning at 100% of the observed azithromycin comparison gives 22 participants per arm for 80% power and 29 participants per arm for 90% power; the corresponding doxycycline figures are 24 and 31.

The effect-size scenarios use the full observed effect, 75% of the observed effect, and 50% of the observed effect. Smaller effects require more participants to detect under the same assumptions, which is why the planning counts rise as the assumed effect is reduced. The registry’s sample-size assumptions also note that time-to-event planning figures, when applicable, would be total events, but these comparisons are not time-to-event calculations. These figures are planning illustrations, not observed power.

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 = 0.300; treatment proportion p₁ = 0.714; average p̄ = 0.507.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7070 = 1.3858.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6435 = 0.5416.
4. n = (1.3858 + 0.5416)² / (0.4143)² = 21.6, rounded up to 22 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 <- 0.3                          # control proportion
p1 <- 0.714285714285714                          # 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(351273)
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 = 22 (80% power), 29 (90% power)
 75% of observed effect: n per arm = 40 (80% power), 52 (90% power)
 50% of observed effect: n per arm = 87 (80% power), 116 (90% power)
power.prop.test: 21.6 
Simulated power at n = 22 per arm: 0.809 (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. Analyses Reported by the Sponsor

The registry reports the sponsor’s analysis as a Fisher Exact test across the azithromycin-and-rifampin, doxycycline-and-rifampin, and placebo groups. It lists a p-value of 0.01 and a value of 40.8 under the label “Mean Difference (Final Values).” The registry does not report a confidence interval for this result or explain the scale of the 40.8 value in relation to the binary response endpoint.

The recalculated estimates are unadjusted pairwise risk differences: azithromycin and rifampin versus placebo, and doxycycline and rifampin versus placebo. Their p-values and confidence intervals describe those separate comparisons, not the sponsor’s overall three-group test. The registry flags a discrepancy between the sponsor result and recalculation. Under that circumstance, the sponsor’s own analysis takes precedence; the recalculation is best treated as a transparent, simple comparison of the posted counts rather than as a correction to the reported analysis.

The record does not explain the source of the discrepancy, the analysis specification behind the sponsor’s result, or how the reported value of 40.8 was derived. It is therefore not possible to determine whether the results differ because of their statistical methods, outcome scale, or another feature of the analysis. Readers should keep the analysis types distinct rather than treating their numerical results as directly interchangeable.

14. Multiplicity, Missing Data and Analysis Populations

The registry describes efficacy and safety analyses as intent to treat. It says that participants who prematurely withdrew or were lost to follow-up for any reason remained in the intent-to-treat population and were considered nonresponders. This gives a stated classification rule for missing primary response status: those participants were counted as not responding. The record does not report how missing values for the secondary measures were handled, whether any sensitivity analyses were performed, or whether the same rule was applied to every measure.

The record lists seven secondary outcome measures, including complete response, erythrocyte sedimentation rate, hsCRP, HAQ DI score, PhGA assessment, and swollen and tender joint counts. It also lists one primary analysis. These listings do not establish whether every outcome was analyzed, how the secondary outcomes were prioritized, or which comparisons were prespecified. Such details are not reported in the registry record.

Whether the sponsor adjusted for multiple outcomes or multiple group comparisons is not reported in the registry record. The registry does not describe a testing hierarchy, family-wise error procedure, or other multiplicity strategy. In general, testing several outcomes or contrasts without accounting for their number can increase the chance of at least one result appearing statistically unusual by chance. This is a general statistical issue, not evidence that a particular adjustment was or was not applied in this trial.

Additional details about the randomization procedure, allocation concealment, analysis populations beyond the stated intent-to-treat population, and handling of protocol deviations are not reported in the registry record. The registry’s triple-masking designation identifies the broad design but does not specify which roles were masked or how masking was maintained. Those omissions limit the ability to reconstruct the analysis plan or assess the effects of missing data and multiplicity from the registry information 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 1% reporting threshold.

The adverse-event record lists zero serious adverse events in each group: azithromycin and rifampin, doxycycline and rifampin, and placebo. It also lists zero other adverse events in each group. The corresponding numbers at risk are 15, 12, and 15, respectively. These denominators differ from the primary response analysis counts; the registry does not explain the difference.

These figures describe what the registry record reports for the listed risk groups. They do not establish that no adverse events occurred outside the registry’s reported categories or ascertainment period. The record does not provide a list of serious adverse-event terms, and it does not report death counts. More detailed information about adverse-event definitions, collection methods, duration of follow-up, or event severity is not reported in the registry record.

In general, safety counts need to be read with their denominators and reporting definitions. Zero reported events in a limited group does not by itself establish the absence of risk. The figures should be interpreted as registry-recorded counts, not as a complete characterization of safety.

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

The primary outcome results describe a greater observed proportion meeting the composite response definition in each combination-antibiotic group than in the placebo group. The posted counts are 10/14 for azithromycin and rifampin, 7/10 for doxycycline and rifampin, and 3/10 for placebo. The registry’s combined description gives 17/24 responders across the combination groups and 3/10 in placebo. Those results concern the specific Month 6 composite definition, which required participants to meet 4/6 clinical criteria.

The recalculated pairwise risk differences point toward more responses in the combination groups, but their confidence intervals are wide. The doxycycline interval includes zero, and the azithromycin interval ranges from 0.045 to 0.784. The sponsor’s reported analysis is a three-group Fisher Exact result with p = 0.01, and the registry flags a discrepancy. The sponsor’s own analysis takes precedence; its reported result and the pairwise recalculations address different comparisons and are not interchangeable.

These findings do not, on their own, show which individual clinical features changed, how meaningful a response was to participants, or how durable any response was after Month 6. The six components of the composite are not reported in the registry record. The available result also does not establish whether findings would be similar in populations or settings beyond those represented by this trial. No medical recommendation follows from these statistics.

The appropriate statistical reading is bounded: the registry reports a higher observed response count in the antibiotic groups than in placebo, while the pairwise recalculations have limited precision and the sponsor analysis is not fully interpretable from the available detail. Understanding the clinical meaning requires the endpoint definitions and analysis details that are not reported in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why an observed difference in response counts and the precision of that difference must be read together. The response counts favor the two combination groups descriptively, but the arm sizes are limited and the recalculated confidence intervals are wide. A point estimate alone would obscure how much uncertainty remains about the size of the group differences.

It also demonstrates how the analysis question changes the interpretation. The sponsor reports a Fisher Exact result across three groups, whereas the recalculated results compare each antibiotic group separately with placebo. An overall test and pairwise estimates do not answer precisely the same question. Because the registry flags a discrepancy, the sponsor’s analysis takes precedence, while the unadjusted calculations remain useful as a transparent account of the posted pairwise counts.

The composite endpoint raises another general statistical issue: a single response label can combine several clinical criteria. Without the definitions of the criteria, readers cannot assess how each component shapes the result. Likewise, the record’s classification of withdrawals and losses to follow-up as nonresponders is important to understanding the primary endpoint counts, while missing-data handling for other measures remains unspecified.

Finally, planning calculations based on observed effects are not evidence of the trial’s original design rationale or of its achieved power. They show how required group sizes change under stated planning assumptions and different target effects. Separating those illustrations from the trial’s observed results helps prevent retrospective sample-size calculations from being mistaken for a design justification.

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 details of the primary endpoint unspecified. It states that participants had to meet 4/6 clinical criteria, but does not identify the criteria, their measurement rules, or how they were combined. This limits interpretation of what a complete response represents and prevents assessment of the contribution of individual components.

The analysis information is also incomplete. The sponsor’s reported Fisher Exact result includes a p-value of 0.01 and a value of 40.8 labeled “Mean Difference (Final Values),” but the record does not explain the value’s scale or provide a confidence interval. A discrepancy is flagged between the sponsor analysis and the unadjusted recalculations, but its source is not reported in the registry record. The sponsor’s analysis takes precedence, yet the available details are insufficient to reconcile the results.

The response-analysis groups are limited in size, and the recalculated confidence intervals are wide. Further details about missing secondary-outcome data, multiplicity adjustment, randomization procedures, allocation concealment, and protocol deviations are not reported in the registry record. The registry also does not explain the difference between the primary-outcome analysis counts and the adverse-event risk counts.

Safety information is limited to the reported serious and other adverse-event counts and their listed denominators. Event definitions, collection procedures, and death counts are not reported in the registry record. These gaps constrain interpretation of both efficacy and safety findings and should be kept in view alongside the posted response results.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Combination Antibiotic Treatment for Reactive Arthritis Caused by Chlamydia Bacteria — NCT00351273.
• Registry record last updated 2018-11-13; status last verified 2016-06.
• Results first posted 2016-08-10.
• 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 a completed, randomized, triple-masked, parallel-group phase 3 trial sponsored by the University of South Florida, with enrollment of 42. The study evaluated azithromycin and rifampin, doxycycline and rifampin, and placebo for chlamydia-induced reactive arthritis. Its primary outcome was a Month 6 composite response requiring 4/6 clinical criteria.

Posted response counts were 10/14, 7/10, and 3/10 in the respective groups. The sponsor reported a three-group Fisher Exact analysis with p = 0.01 and a value of 40.8 labeled “Mean Difference (Final Values).” Unadjusted recalculated pairwise risk differences have wide confidence intervals, and the registry flags a discrepancy; the sponsor’s own analysis takes precedence.

The record states that early withdrawals and participants lost to follow-up were included in the intent-to-treat population and considered nonresponders. It lists seven secondary outcome measures and reports zero serious and zero other adverse events in the listed risk groups. Several endpoint, analysis, missing-data, and safety details are not reported in the registry record.