High-dose vs. Standard-dose Cephalexin for Cellulitis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for High-dose vs. Standard-dose Cephalexin for Cellulitis (NCT04471246), a randomized phase 4 study. The primary outcome is Oral Antibiotic Treatment Failure.

2. Clinical Question

The trial addresses whether a higher cephalexin dose merits evaluation in a larger trial, while also assessing the feasibility and design of that future research. The registry describes this as a pilot randomized controlled trial. Its stated motivation is the need for evidence to guide antibiotic treatment for cellulitis and the concern that treatment failure may lead to changes in antibiotics, including escalation to intravenous treatment.

The primary endpoint is a patient-level failure outcome. It includes a change in the class of oral antibiotic or a step up to intravenous therapy within seven days due to worsening infection. Worsening is specified by new fever, or fever persisting at the day 3 follow-up; an increasing area of erythema relative to baseline; or increasing pain relative to baseline. The posted arm-specific percentages describe the assessed participants. They are not a direct estimate of the probability that a future patient will experience failure, and they do not by themselves establish a causal treatment effect.

3. Trial Design

The registry record describes a phase 4, randomized, triple-masked, parallel-group pilot trial. The listed enrollment is 69. The record identifies cellulitis as the condition and Ottawa Hospital Research Institute as the sponsor. The study status is completed, with a start date of 2021-08-16 and a primary completion date of 2022-02-23.

Participants were assigned to high-dose cephalexin or standard-dose cephalexin with oral placebo. The high-dose regimen was 1000mg orally four times daily for seven days. The standard-dose regimen was 500mg orally four times daily, with oral placebo, for seven days. The registry labels masking as triple, but which roles were masked and how masking was maintained are not reported in the registry record. The method used to generate or conceal the random allocation is also not reported in the registry record.

The enrollment field lists 69, whereas the primary-outcome description says that 66 patients were enrolled and confirmed to have cellulitis by the Principal Investigator. That description reports that 62 were assessed for the primary outcome and four were lost to follow-up. The treatment-failure results list 31 participants in each arm. These figures describe different registry entries; further explanation for the enrollment count difference is not reported in the registry record.

The trial was designed as a pilot to inform the feasibility and design of a larger clinical trial. Listed secondary outcomes include ability to approach eligible patients, assessment of blinding, protocol adherence, loss to follow-up, clinical cure, adverse events, and unplanned emergency-department visits or hospitalization. The registry record does not report further details of the procedures used to assess these outcomes.

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: Number of patients within the 6 months of the trial that were determined to be eligible of the trial. Of 134, 69 were successfully approached and recruited in the 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 Oral Antibiotic Treatment Failure, measured over Each patient was assessed for oral antibiotic treatment failure at the day 3 and day 7 follow-ups.

The primary endpoint is the percentage of enrolled participants in each arm with oral antibiotic treatment failure. Failure is defined as a change in antibiotic within seven days because the infection worsened. The change may be a change in the class of oral antibiotic or escalation to intravenous therapy.

The registry defines worsening using clinical criteria: new fever of at least 38.0°C or fever persisting at the day 3 follow-up; an erythema area increasing by at least 20% from baseline; or pain increasing by at least two points from baseline on a numeric rating scale. The registry says patients were assessed for failure at day 3 and day 7 follow-ups.

This endpoint is used because it captures a change in treatment prompted by worsening infection, rather than relying only on a symptom measurement. It combines the treatment decision with stated clinical criteria. That makes the definition consequential: events count when the specified clinical change leads to an antibiotic change within the stated period. The registry does not report how clinicians’ decisions were independently reviewed or adjudicated.

The primary outcome description says 62 participants were assessed, with 31 in each arm. Four patients were lost to follow-up and could not be assessed for failure. The posted percentages and confidence intervals therefore describe the assessed participants, not an outcome determination for every person listed in the enrollment field.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count-type outcome expressed as a percentage: whether a participant met the registry’s definition of oral antibiotic treatment failure. For each arm, the registry reports a percentage and a 95% confidence interval. The treatment-failure estimate is 3.2% for high-dose cephalexin, with an interval of 0.6% to 16.2%; the estimate is 12.9% for standard-dose cephalexin, with an interval of 5.1% to 28.9%. Each arm has 31 participants in the posted outcome results.

What do the percentages represent?

General explanation: a percentage for a binary outcome describes the share of assessed participants who met the event definition. Here the event is not simply persistent symptoms; it is an antibiotic change within seven days due to worsening infection, as defined in the registry. This distinction matters when interpreting the result: the measure incorporates both specified signs of worsening and a resulting treatment change.

The registry records the outcome unit as the percentage of enrolled participants in each arm, while also stating that four participants could not be assessed after loss to follow-up. The posted arm counts are 31 and 31. The registry record does not report a separate numerator for each percentage, and the analysis field does not post a comparative test. The rounded percentages should therefore be interpreted as the posted estimates, rather than used to reconstruct event counts or additional statistics.

How should the confidence intervals be read?

General explanation: a confidence interval represents uncertainty in an estimated quantity under the assumptions and method used to calculate it. A narrower interval indicates greater precision than a wider interval, all else being comparable. A broad interval means that the data do not pin down the population proportion closely. These intervals are broad relative to their point estimates, so the observed percentages should not be treated as precise predictions.

The confidence intervals are reported for each arm separately. General explanation: a comparison of treatments is ordinarily expressed using a contrast, such as a difference between event proportions, together with uncertainty for that contrast. The uncertainty of a contrast is not established merely by placing two arm-specific intervals side by side. The registry record does not report a contrast estimate, its confidence interval, a p-value, or a method for testing a difference.

The record does not specify the method used to calculate the confidence intervals. It is therefore not possible to identify from the registry whether a particular interval procedure, continuity adjustment, or other calculation was used. General statistical explanations of interval methods should not be mistaken for a description of the method used in this trial.

What does the randomized design contribute?

General explanation: random allocation is intended to create groups that are comparable, on average, with respect to factors that may affect outcomes. It supports a causal comparison when allocation, follow-up, outcome measurement, and analysis are appropriately conducted. Randomization does not remove uncertainty, especially when the number of assessed participants is limited, and it does not ensure that every relevant feature will be balanced in a particular trial.

The registry labels the design randomized, triple-masked, and parallel. It does not report the allocation-generation procedure, concealment process, masking roles, or details for maintaining the blind. The record also does not identify the analysis population or give a statistical analysis plan. Those details should not be inferred from the design labels alone.

What is the role of the pilot and planning information?

The stated purpose is to determine feasibility and inform the design of a future trial. General explanation: pilot studies can help researchers assess whether recruitment, follow-up, adherence, and outcome measurement are workable, while offering preliminary estimates that may help with later planning. An estimate from a pilot can be unstable, particularly when it is based on a limited number of assessed participants.

The registry describes planning assumptions as a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the effect to detect described as a fraction of the observed effect. It also distinguishes participant counts per arm for non-time-to-event outcomes from total event counts for time-to-event outcomes. A calculated planning sample size is not reported in the registry record. Any planning illustration based on an observed effect is a design aid, not observed power and not evidence that a particular effect is established.

General explanation: over-dispersion matters when variation in count data exceeds what a chosen model expects; ignoring it can make uncertainty appear too small. The posted primary outcome is a binary participant-level failure measure, and the registry does not report a count model or an over-dispersion adjustment. There is no basis in the record to say that such an adjustment was used or needed.

The analysis listing reports no posted analyses. The registry also indicates that values are missing or not estimable for a two-arm comparison of this count-type outcome. Consequently, the arm-specific estimates can be described, but a sponsor-reported inferential comparison is unavailable. No recalculated comparison is posted that could supplement or be weighed against a sponsor analysis.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists seven secondary outcomes: ability to approach eligible patients, assessment of blinding, protocol adherence, loss to follow-up, clinical cure, adverse events, and unplanned emergency-department visits or hospitalization. The record reports nine outcome measures overall. It does not post analyses for these outcomes in the analysis listing, or identify which outcome measures correspond to each secondary outcome.

General explanation: when researchers test multiple outcomes or make repeated comparisons, the chance of at least one apparently notable result arising by chance can increase. Statistical plans may address this through a prespecified primary outcome, a hierarchy of tests, or an adjustment for multiple comparisons. The registry record does not report a multiplicity strategy or enough analysis results to assess how any such strategy was applied. The listed outcomes should not be treated as independent confirmatory findings.

For missing primary-outcome data, the record says four of the 66 patients confirmed as having cellulitis were lost to follow-up and could not be assessed for treatment failure. It does not report the reasons for loss, allocation of those patients by arm, or a missing-data method. No imputation, sensitivity analysis, or assumption about the unobserved outcomes is reported in the registry record.

The registry does not specify whether the primary analysis followed intention-to-treat, per-protocol, or another population definition. The arm-level counts of 31 in each group describe the participants with posted primary-outcome results, but do not establish the analysis population for all trial outcomes. The broader enrollment field lists 69, while the primary-outcome description refers to 66 enrolled and confirmed; the record does not explain the relationship between these entries.

11. 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.

Most frequent non-serious adverse event terms

The posted serious adverse event counts are zero in the high-dose arm and zero in the standard-dose arm. Each arm lists 31 participants at risk for serious adverse events. The record lists no serious adverse event terms. Deaths are reported as zero in each arm.

The entries for other adverse events show 12 affected participants among 31 at risk in the high-dose arm and 8 affected participants among 31 at risk in the standard-dose arm. These are participant counts as posted, not percentages. The registry record does not provide event descriptions, severity grading, timing, or details about whether participants could have experienced more than one event.

General explanation: an observed count of zero serious events does not establish that the risk is zero. It means no serious events were recorded in the listed groups during the period represented by the safety data. With the limited group sizes in this pilot, the absence of recorded events should be interpreted as a description of these observations, not as a precise estimate of rare risks or a comparison establishing equivalent safety.

The registry does not report a statistical comparison of adverse events between arms. Nor does it provide enough detail to interpret the other-event counts by type. The counts should be read as descriptive safety information, not as proof for or against a treatment-related difference.

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

The registry record provides preliminary, arm-specific estimates of oral antibiotic treatment failure in a pilot trial. The posted percentages are 3.2% for high-dose cephalexin and 12.9% for standard-dose cephalexin. The confidence intervals are wide, and the record does not post an inferential comparison between the treatments. These features limit how precisely the observed pattern can be interpreted.

The trial’s stated purpose is to assess feasibility and inform the design of a larger trial. It is therefore important to distinguish a pilot estimate from a definitive answer to a comparative research question. The point estimates describe the assessed participants under the registry’s outcome definition. They do not establish what would happen in a larger, different population or under different care conditions.

The results do not establish that high-dose cephalexin reduces treatment failure. A lower posted percentage in one arm is descriptive; without a reported between-arm estimate and its uncertainty, the size and precision of a comparative effect cannot be assessed from the registry analysis. The record also does not report a p-value, a prespecified statistical plan, or an adjusted analysis that would support a claim of statistical significance.

The results also do not establish that the treatments have the same failure rate. Wide arm-specific confidence intervals and a lack of a comparative analysis leave uncertainty. Absence of a reported difference is not evidence of equivalence or noninferiority; the registry does not report an equivalence or noninferiority framework.

Four participants in the primary-outcome population were lost to follow-up and could not be assessed. Their outcomes are unknown in the registry record. This missingness adds uncertainty, and its potential effect cannot be evaluated without further information about the participants and the handling of missing outcomes.

The safety data are similarly descriptive: no serious adverse events or deaths are listed in either arm, while other adverse events are recorded in both. These observations do not quantify rare risks or establish a comparative safety conclusion. Taken together, the posted data are best understood as preliminary information from a pilot study, with uncertainty that a larger, fully reported comparative analysis would need to address.

13. Why This Trial Matters Statistically

This trial matters statistically because it illustrates the difference between estimating each arm’s outcome and estimating a treatment contrast. The registry reports an outcome percentage and a confidence interval for each regimen, but does not post a direct comparison. A reader can describe the two observed estimates, yet cannot infer the precision of their difference from the arm-specific summaries alone.

The width of the confidence intervals also matters. The estimates are based on 31 assessed participants per arm, and the intervals span broad ranges. Such imprecision is expected to be important when interpreting pilot results: a point estimate can look more definite than the underlying evidence warrants. General explanation: a pilot can provide preliminary information for future planning, but the observed estimate itself may be unstable and should not automatically become the assumed effect in a later trial.

The record’s stated purpose—assessing feasibility and informing a larger trial—puts emphasis on more than the treatment-failure percentages. Recruitment, blinding, adherence, and follow-up are among the listed secondary outcomes. Those outcomes can help determine whether a definitive study can be conducted and what procedures may need attention. However, the registry record does not post analyses for those outcomes, so their feasibility findings cannot be evaluated from the reported results.

Missing outcome data are also a statistical concern. Four participants in the described primary-outcome population could not be assessed. Without a reported missing-data approach or sensitivity analysis, it is not possible to determine how much the absent outcomes could influence the estimates. Reporting those details would help readers assess the robustness of the result.

Finally, sample-size planning should be separated from observed evidence. The registry describes planning assumptions based on the observed effect, but a calculated planning sample size is not reported. General explanation: planning estimates are conditional on assumptions about the effect, variability, design, and analysis; they do not show that a study had a given probability of detecting an effect after it was conducted. Pilot results are useful for planning questions, but are not a substitute for a powered comparative analysis.

14. 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 outcome is based on 62 assessed participants, with 31 in each arm. The primary-outcome description reports four losses to follow-up among 66 patients enrolled and confirmed to have cellulitis. The registry does not describe the reasons for loss, their arm allocation, or the missing-data method. The enrollment field separately lists 69, and the record does not explain the difference between that number and the primary-outcome population.

The treatment-failure percentages have broad 95% confidence intervals, limiting precision. No between-arm effect estimate, p-value, or inferential analysis is posted. The record also does not report the method used to construct the confidence intervals, the statistical analysis plan, or the analysis population. These omissions restrict evaluation of the statistical approach and of the comparative result.

Although the registry labels the trial randomized and triple-masked, details of allocation generation, concealment, masking roles, and maintenance of masking are not reported in the registry record. The record does not explain how the clinical criteria were assessed or whether treatment-failure decisions were independently adjudicated. Such details would help readers assess potential sources of bias.

The trial is described as a pilot intended to inform a larger trial. The registry lists several feasibility and clinical secondary outcomes, but does not post analyses for them. It also does not report a calculated planning sample size. Thus, the feasibility conclusions and design implications cannot be assessed in detail from the posted analyses.

The safety record lists no serious adverse events or deaths in either arm, but the group sizes are limited and other adverse events are reported without event descriptions. These data cannot provide a detailed characterization of safety or establish a comparative safety effect. The registry record also does not report further context for follow-up beyond the primary outcome’s day 3 and day 7 assessments.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: High-dose vs. Standard-dose Cephalexin for Cellulitis — NCT04471246.
• Registry record last updated 2025-01-17; status last verified 2024-12.
• Results first posted 2025-01-17.
• 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.

18. Record Summary

The registry record describes NCT04471246, a completed phase 4 randomized, triple-masked, parallel pilot trial of high-dose versus standard-dose cephalexin for cellulitis. Ottawa Hospital Research Institute is listed as the sponsor. The record lists enrollment of 69 and study dates from 2021-08-16 to primary completion on 2022-02-23.

The primary outcome is oral antibiotic treatment failure within seven days, defined by an antibiotic change due to worsening infection. Results are reported for 31 assessed participants in each arm: 3.2% for high-dose cephalexin, with a 95% confidence interval of 0.6% to 16.2%, and 12.9% for standard-dose cephalexin, with a 95% confidence interval of 5.1% to 28.9%. The record does not post a comparative statistical analysis.

The registry describes the study as a pilot intended to inform the feasibility and design of a larger trial. It reports four losses to follow-up in the primary-outcome population, and does not report a method for handling missing outcomes. Serious adverse events and deaths are listed as zero in both arms; other adverse events are listed for 12 participants in the high-dose arm and 8 in the standard-dose arm.