Prospective Study of Methicillin-Resistant Staphylococcus Aureus (MRSA) Among HIV-Infected Persons: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Prospective Study of Methicillin-Resistant Staphylococcus Aureus (MRSA) Among HIV-Infected Persons (NCT00631566), a randomized na study. The primary outcome is The Presence of MRSA on Repeated Swabs to Assess the Efficacy of These Medications on Clearing MRSA Colonization.

2. Clinical Question

3. Trial Design

The registry describes a prospective study of MRSA among HIV-infected persons and a randomized, double-blind, parallel-group study of decolonization procedures. The title and summary identify several related aims: evaluating MRSA prevalence and incidence, investigating predictors of colonization and infection, examining decolonization for clearing carriage and preventing infection, and characterizing MRSA isolates and antimicrobial sensitivities. The primary endpoint posted for the randomized component is MRSA colonization at 6 months post-randomization.

The planned enrollment is 550, and the record lists the study as completed. The study began in 2007-05, with primary completion in 2017-05. Its phase is listed as NA. The listed interventions are mupirocin (Bactroban) 2% with hexachlorophene (pHisoHex) 3%, and placebo. The registry identifies randomized allocation, double masking, and a parallel model. Further details about the randomization procedure, masking roles, allocation concealment, visit schedule, and treatment administration are not reported in the registry record.

The posted primary outcome comparison includes 18 participants in the mupirocin/hexachlorophene group and 21 in the placebo group. The endpoint description says only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed. A separate “No MRSA Colonization” category is listed with 501 participants and no endpoint events. That category is not one of the two randomized treatment groups and should be kept distinct from the randomized comparison.

The posted outcome summaries contain a risk-difference comparison between the randomized groups. The registry record does not describe the planned statistical model or a sample-size justification. Although the study’s broad objectives include incidence, infection predictors, and isolate characteristics, those objectives should not be conflated with the particular 6-month colonization endpoint reported for the randomized comparison.

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

Analysis population for the primary outcome, as defined in the registry: Only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed.

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 The Presence of MRSA on Repeated Swabs to Assess the Efficacy of These Medications on Clearing MRSA Colonization, measured over 6 months.

The primary endpoint is the presence of MRSA on repeated swabs, with the posted outcome described as MRSA colonization at 6 months post-randomization. The registry expresses the measure as a count of participants. In the treatment comparison, the reported event counts are 12 among 18 participants in the mupirocin/hexachlorophene group and 14 among 21 participants in the placebo group. The observed proportions are reported as equal.

Colonization is carriage of MRSA, which is distinct from a clinical MRSA infection. The endpoint description frames the swabs as a way to assess whether the medications clear colonization. It therefore addresses carriage status at the stated follow-up rather than directly measuring all infections over the study, the incidence of infection, or the molecular and antimicrobial characteristics of isolates. Those broader topics appear among the study objectives or secondary outcomes, but they are not the primary endpoint being interpreted in the randomized comparison.

Repeated swabs can assess whether MRSA is detected at sampled body sites and follow-up. The registry’s secondary outcomes name the nares, throat, perirectal, axilla, and groin areas for colonization assessment over time. The exact swab protocol, definition of a positive result, and handling of discordant or missing swabs are not reported in the registry record. In general, a binary endpoint reduces a participant’s observations to whether the specified event is present or absent under the endpoint definition; that makes the result easier to interpret as a group proportion, while leaving details of timing and sampling important to its meaning.

7. Results Posted in the Registry

• Mupirocin/Hexachlorophene Group: 66.7% (12/18) versus 66.7% (14/21) on control; risk difference 0.0 percentage points (95% CI -29.7 to 29.7), risk ratio 1.00 (0.64–1.56); unadjusted two-proportion z-test p = 1.000.
• No MRSA Colonization: 0.0% (0/501) versus 66.7% (14/21) on control; risk difference -66.7 percentage points (95% CI -86.8 to -46.5); unadjusted two-proportion z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a risk difference express?

In general, a risk difference subtracts the event proportion in one group from the event proportion in another. It is expressed on the same scale as the proportions being compared. A value at the no-difference point indicates that the observed proportions are alike; a positive or negative value indicates the direction of the group contrast according to the order used in the analysis. For the randomized comparison, the posted risk difference is 0.0, and the registry reports equal proportions for the active and placebo groups.

The risk difference is useful because it describes an absolute contrast in event proportions. It should be interpreted with its confidence interval, which conveys uncertainty about the estimate. Here, the posted interval spans negative and positive risk differences, so the data are compatible with differences in either direction. The interval’s width is important: an estimate at the no-difference point does not by itself imply high precision.

What does the p-value tell the reader?

In general, a p-value describes how unusual data at least as discrepant as those observed would be if a specified null model were true, under the assumptions of the test. It is not the probability that the null hypothesis is true, nor does it measure the probability that a treatment works. The randomized comparison has an unadjusted p-value of 1.0. This is consistent with the equal observed proportions, but it is not evidence of equivalence. Equivalence requires a defined margin and an analysis designed to assess whether the uncertainty lies within that margin; neither is reported in the registry record.

The record also reports p < 0.001 for a comparison of placebo with the separate “No MRSA Colonization” category. A small p-value cannot make a nonrandomized or categorically different group into a valid treatment comparator. The latter category is not an assigned decolonization arm, so this contrast does not answer whether mupirocin/hexachlorophene differs from placebo.

Why is the analysis population important?

The endpoint note says only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed. In general, missing outcome data can affect both precision and the validity of a comparison. If outcome availability is related to treatment assignment or to colonization status, an analysis restricted to participants with observed outcomes may not preserve the balance created by randomization. The registry record does not report why data were missing, how missing outcomes were handled, or whether sensitivity analyses were performed. It is therefore not possible to determine from the posted description how these issues were addressed.

The posted arm counts for the endpoint are 18 and 21, consistent with the outcome population described in the record. These are the analyzed group sizes for this endpoint, not the overall planned enrollment. The separate “No MRSA Colonization” category should not be combined with the randomized arms when interpreting the treatment comparison.

What can be inferred from a binary count?

In general, a binary endpoint records whether a participant meets an event definition. The event counts and group sizes can then be represented as proportions and compared. Such a summary is straightforward, but it does not show when colonization arose or cleared between assessments, how many sites were positive, or whether the participant developed an infection. Those distinctions matter because the registry’s broader objectives include incidence, change over time, and infection outcomes, while the primary comparison is a colonization status measure at 6 months.

The posted comparison is described as unadjusted. An unadjusted estimate compares the observed groups without accounting for additional covariates in a model. In general, adjusted analyses can account for measured characteristics when a suitable analysis plan supports them, but adjustment does not automatically remove bias from missing outcomes or unmeasured differences. No adjusted estimate or statistical model is reported in the registry record.

What does the design establish?

Randomized allocation and double masking are listed for the decolonization study. In general, randomization is intended to support a fair comparison by assigning treatment independently of participant characteristics, while masking can reduce some influences on care or outcome assessment. The registry record does not report the allocation mechanism, concealment details, masking roles, or whether masking was maintained. Nor does it provide enough information to assess adherence, crossover, or protocol deviations.

The registry lists parallel groups, with mupirocin/hexachlorophene and placebo as interventions. The record does not report the intended statistical plan, interim analyses, or sample-size justification. No trial-specific conclusions about those methods can be drawn from the design labels alone. A general sample-size illustration based on an observed effect, if considered, would be a planning exercise rather than observed power; no such numerical illustration is interpreted.

Finally, the sponsor’s own analysis would take precedence if a discrepancy with a recalculation were reported. The registry record indicates no discrepancy. The displayed risk difference and p-value are therefore discussed as posted, without presenting a separate recalculation.

10. Confidence Intervals and Precision

Precision is conveyed chiefly by the confidence interval around the randomized risk difference. The observed risk difference is 0.0, but its interval extends from a negative to a positive value. Thus the data do not isolate the estimate at no difference: they remain compatible with effects favoring either group. The interval is wide in relation to the estimate, so the apparent equality of the observed proportions should not be mistaken for a precise estimate of the underlying contrast.

In general, a confidence interval describes the range of values reasonably compatible with the data under the analysis assumptions. It is not a guarantee that the true effect lies in that particular interval, and it does not assign probabilities to individual values. Its interpretation also depends on the analysis being appropriate and on the outcome data representing the intended population. The posted endpoint population is restricted to participants with 6-month colonization data, and the record says only 80% of the 49 initial randomizations had such data and were analyzed.

The counts are 12 among 18 in the mupirocin/hexachlorophene group and 14 among 21 in the placebo group. The relatively limited analyzed group sizes contribute to uncertainty in proportions and their contrast. The record does not report a precision target or a confidence-interval planning criterion. Nor does it report an equivalence margin. Consequently, a result centered on no observed difference cannot establish that any possible difference is clinically or scientifically negligible.

The unadjusted p-value of 1.0 is consistent with the observed equality, but the interval is more informative about the range of effects compatible with the data. The separate comparison with the “No MRSA Colonization” category has its own interval and addresses a different contrast; it should not be used to increase confidence in the randomized treatment estimate.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures, three secondary outcomes, and zero analysis entries. The three named secondary outcomes concern MRSA colonization prevalence and incidence at specified body areas and changes over time; CD4 counts and HIV viral loads around MRSA or soft tissue infection; and molecular and antimicrobial-sensitivity characteristics of MRSA isolates. The registry record does not report a multiplicity adjustment, a hierarchy of hypotheses, or a prespecified distinction between confirmatory and exploratory analyses. Without those details, the effect of testing multiple outcomes on interpretation is not reported in the registry record.

The primary endpoint note states that only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed. The record does not explain why the remaining participants lacked these data, whether missingness differed by group, or how missing outcomes were treated. It does not report an intention-to-treat analysis, an imputation method, or sensitivity analyses. The posted treatment comparison should therefore be understood as based on the participants with observed endpoint data described by the registry, rather than assumed to represent a complete analysis of all randomized participants.

In general, missing data can reduce precision and can introduce bias if the likelihood of an observed outcome is associated with treatment or the outcome itself. Randomization does not automatically resolve this problem when the analysis excludes participants with missing outcomes. The record does not provide enough information to assess whether such bias is likely or how robust the result is to alternative assumptions.

The posted counts for the two randomized groups are the relevant endpoint counts for interpreting the comparison. The “No MRSA Colonization” group is a separate category and is not part of the randomized active-versus-placebo analysis. The record does not report further analysis-population definitions, protocol deviations, or participant flow details needed to reconcile all reported groups with the broader planned enrollment.

12. 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 serious adverse event terms

Serious events with the highest arm-level rate: hospitalization for gastroparesis (4.8%).

The registry reports serious adverse events for the listed groups. For the mupirocin/hexachlorophene group, 1 serious event is reported among 21 participants at risk. For the placebo group, 0 serious events are reported among 28 at risk. The separate “No MRSA Colonization” category reports 0 serious events among 501 at risk. The record reports no deaths in any of these groups.

The registry lists hospitalization for gastroparesis as the top serious event term, with 4.8%. The record does not provide enough detail to attribute that term to a particular treatment group or to establish a causal relationship with study treatment. The serious-event counts are descriptive safety summaries; by themselves, they do not determine whether an event was caused by an intervention.

The numbers at risk in the safety summaries differ from the endpoint group counts reported for the 6-month colonization analysis. These summaries refer to different reported populations or outcome records, and the registry does not explain the denominator difference. They should not be substituted for one another or combined. The registry also reports no other adverse-event counts in the registry record described for this analysis. Absence of a reported event in these summaries is not proof that no event occurred outside the reporting scope.

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

The randomized comparison reports equal observed colonization proportions at 6 months in the mupirocin/hexachlorophene and placebo groups. The event counts are 12 among 18 and 14 among 21, respectively; the posted risk difference is 0.0, and the unadjusted p-value is 1.0. These statistics describe participants with available endpoint data, as the registry notes that only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed.

The result does not establish that the intervention and placebo are equivalent. The confidence interval spans risk differences in both directions, and no equivalence margin or equivalence analysis is reported. The result also does not show that the true treatment difference is exactly zero. It is an estimate with uncertainty, based on the analyzed population described in the record.

The primary endpoint concerns MRSA colonization, not all MRSA infections. It does not by itself establish an effect on infection prevention, incidence, CD4 counts, HIV viral loads, or isolate characteristics. Those matters are included among the study’s broader aims or secondary outcomes, but corresponding analysis results are not reported in the registry record in the registry record described for this page.

The separate “No MRSA Colonization” category is not a randomized comparator and cannot answer the active-treatment question. The comparison of that category with placebo is therefore not evidence about the effect of mupirocin/hexachlorophene relative to placebo.

Overall, the posted randomized result is a null-centered estimate with substantial uncertainty, not proof of no effect and not evidence of benefit. Missing endpoint data, limited analyzed group sizes, and the absence of reported analysis details further constrain interpretation. No medical recommendation follows from these registry summaries.

14. Why This Trial Matters Statistically

This study illustrates why the comparison group and analysis population must be identified before interpreting a statistical contrast. The registry lists both randomized treatment groups and a separate “No MRSA Colonization” category. Only the mupirocin/hexachlorophene and placebo groups form the randomized comparison relevant to the decolonization question. A contrast with the separate category answers a different question and should not be treated as a treatment effect.

It also illustrates the distinction between an observed estimate and its precision. The treatment groups have equal posted proportions and a risk difference of 0.0, but the confidence interval spans effects in either direction. A null-centered estimate is not the same as a precise estimate of no effect, and a non-significant result does not establish equivalence.

The missing follow-up information is consequential for interpretation. The record says only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed, but does not describe the reasons for missingness or its handling. A reader therefore needs to consider both the uncertainty in the estimate and the limited information about how the analysis population was formed.

Finally, the trial’s broader prospective aims extend beyond the primary colonization comparison. Its objectives include incidence, predictors, infection, and characteristics of MRSA isolates. Keeping those aims separate from the posted 6-month endpoint prevents a narrow result from being overgeneralized. The record is a useful reminder that a trial’s title and objectives, endpoint definition, analyzed sample, and statistical contrast each contribute different pieces of evidence.

15. 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 endpoint analysis is based on participants with 6-month colonization data, and the registry states that only 80% of the 49 initial randomizations had those data and were analyzed. The reasons for missing outcomes and the method used to handle them are not reported in the registry record. This limits assessment of whether the analyzed participants remain representative of the randomized groups and how missingness may affect the estimate.

The randomized groups included in the endpoint summary are limited in size: 18 participants in the mupirocin/hexachlorophene group and 21 in the placebo group. The confidence interval spans effects in both directions, reflecting substantial uncertainty. The registry does not report a sample-size justification or a precision target, so the adequacy of the planned enrollment for this particular comparison cannot be assessed.

The record does not provide detailed information about swab collection, the definition of a positive result, treatment adherence, or the timing and handling of repeat assessments. It also does not report the full statistical plan, adjusted analyses, multiplicity procedures, interim analyses, or sensitivity analyses. The registry’s broad aims include infection and incidence questions, but the primary result discussed is colonization at 6 months and should not be generalized to those other outcomes.

The safety summaries use different reported at-risk counts from the endpoint summaries, and the record does not explain the difference. The separate “No MRSA Colonization” category is not randomized and is not an appropriate treatment comparator. Together, these limitations support a cautious, outcome-specific reading of the posted randomized result.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Prospective Study of Methicillin-Resistant Staphylococcus Aureus (MRSA) Among HIV-Infected Persons — NCT00631566.
• Registry record last updated 2023-03-03; status last verified 2023-03.
• Results first posted 2018-08-16.
• 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.

19. Record Summary

The registry record describes a completed, randomized, double-blind, parallel-group study of decolonization procedures within a prospective study of MRSA among HIV-infected persons. Planned enrollment is 550; the listed study dates are 2007-05 to 2017-05, and phase is recorded as NA. The randomized interventions are mupirocin (Bactroban) 2% with hexachlorophene (pHisoHex) 3%, and placebo.

The primary outcome is MRSA colonization at 6 months post-randomization. The posted randomized groups include 18 participants in the active group and 21 in the placebo group, with event counts of 12 and 14. The observed proportions are equal, the risk difference is 0.0, and the unadjusted p-value is 1.0. The confidence interval spans both negative and positive risk differences. The record notes that only 80% of the 49 initial randomizations had 6-month colonization data and were analyzed.

The separate “No MRSA Colonization” category is not a randomized treatment group. The registry does not report the detailed statistical plan, handling of missing outcomes, or a sample-size justification. The posted randomized result should be interpreted as an uncertain comparison of colonization status, not as a conclusion about the study’s broader infection or incidence aims.