Studies of the Ocular Complications of AIDS (SOCA)--Foscarnet-Ganciclovir CMV Retinitis Trial (FGCRT): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Studies of the Ocular Complications of AIDS (SOCA)--Foscarnet-Ganciclovir CMV Retinitis Trial (FGCRT) (NCT00000136), a randomized phase 3 study. The primary outcome is Mortality.

2. Clinical Question

In statistical terms, the question is whether the observed mortality outcome differed between participants assigned to foscarnet and those assigned to ganciclovir. Random assignment supports a between-group comparison, but the usefulness of that comparison depends on the outcome being recorded and analysed in a way that represents the intended clinical event. The registry lists 107 participants in the foscarnet arm and 127 in the ganciclovir arm. It does not report participant-level outcome details, a time-to-event analysis, or an explanation of how the mortality counts were obtained.

The registry states that all enrolled patients would be followed until a common study closing date chosen to provide a minimum of 1 year of follow-up for all enrolled patients. That statement describes the intended follow-up framework; it does not establish how complete follow-up was, how deaths were defined or ascertained, or whether the posted counts reflect observed deaths. Those details are not reported in the registry record.

3. Trial Design

This was a phase 3 randomized, single-masked, parallel-group trial. The registry lists 234 participants as enrolled and identifies Johns Hopkins Bloomberg School of Public Health as the sponsor. The study is marked completed. The listed start was 1990-03 and the primary completion was 1991-10. The two parallel arms were foscarnet and ganciclovir.

The registry gives arm sizes of 107 for foscarnet and 127 for ganciclovir. Random allocation means treatment assignment was intended to be determined by chance, rather than selected by the participant or investigator. In general, this can reduce systematic differences between groups at the start of a trial. It does not guarantee that every measured or unmeasured characteristic will be balanced, and interpretation still depends on how assignment, follow-up, and outcomes were handled. The allocation method and any stratification details are not reported in the registry record.

Single masking indicates that the record describes the trial as single-masked, but it does not identify which people were masked or explain how masking was maintained. Procedures for allocation concealment, treatment administration, and outcome assessment are not reported in the registry record. These details matter because knowledge of assigned treatment can influence care, reporting, or assessment, depending on who knows the assignment and which outcome is being measured.

The stated objective was to evaluate the relative safety and efficacy of foscarnet and ganciclovir as initial treatment for CMV retinitis. The primary outcome listed in the results is mortality, with a follow-up description referring to a common closing date and a minimum of 1 year of follow-up for all enrolled patients. A fuller protocol, including the rationale for the planned enrollment, the statistical analysis plan, and how treatment changes were handled, is not reported in the registry record.

Only one outcome measure is listed, and the record lists no posted analyses and no posted secondary outcomes. This limits the information available for assessing the analysis population, outcome definitions, and any adjustment for other factors. The available design labels establish the broad structure of the trial, but do not provide enough detail to reconstruct all design and analysis decisions.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Mortality, measured over All patients enrolled will be followed until a common study closing date, which was chosen to provide a minimum of 1 year of follow-up for all patients enrolled in the trial..

The primary endpoint is mortality, recorded in the registry as a number of participants. Mortality is an important clinical outcome because it records whether a participant died during the relevant observation period, rather than measuring a laboratory marker or a change in symptoms. For a comparison of randomized groups, the number of deaths can be expressed as a risk or proportion within each group, provided the event counts, denominators, and follow-up are defined consistently.

The registry describes follow-up until a common study closing date, selected to provide a minimum of 1 year of follow-up for all enrolled patients. A common closing date can provide a defined observation framework, but a single count does not by itself show when events occurred or how long each participant was observed. The record does not report a time-to-event endpoint, censoring details, or a fuller definition of the mortality outcome.

The posted result lists 107 mortality events among 107 participants in the foscarnet arm and 127 among 127 participants in the ganciclovir arm. These counts match the arm sizes recorded in the registry. The posted comparison gives a risk difference of 0.0 and p = 1.0. Because the record does not explain the count coding or provide further outcome detail, it is not possible to determine from the registry entry whether these figures represent deaths as ordinarily understood, or how the endpoint was derived. This uncertainty is central to interpreting the result.

7. Results Posted in the Registry

• Ganciclovir: 100.0% (127/127) versus 100.0% (107/107) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0), risk ratio 1.00 (1.00–1.00); unadjusted two-proportion z-test p = n/a.

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods help translate observed outcomes into comparisons while making assumptions and uncertainty visible. The registry record provides a risk difference, a confidence interval, and a p-value for mortality, but does not describe the statistical analysis plan or the procedures used to produce these results. The concepts below explain how such summaries are generally interpreted; they are not a claim about methods used in this trial.

What does a risk difference represent?

In general, a risk is the proportion of participants in a group who experience an event during a defined period. A risk difference subtracts the risk in one group from the risk in the other. It is expressed on the same scale as the risks, and a value of zero means the recorded risks are equal. The registry’s posted risk difference is 0.0, and the posted group proportions are 1.0 in both arms. This arithmetic summary does not establish that the underlying outcome data were correctly coded or that follow-up was complete.

For a risk comparison to be meaningful, the event definition, observation period, and denominators need to be clear. Counts must represent the intended event, and participants who could not be assessed need to be handled in a stated way. The registry records the endpoint as mortality and gives arm-level counts equal to arm sizes, but does not explain the coding or report missing-outcome handling. That makes it difficult to determine whether the posted proportions correspond to the clinical outcome as intended.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with an estimate under a specified statistical procedure. It reflects sampling uncertainty under that procedure and its assumptions. It is not a range containing a guaranteed proportion of individual outcomes, and it does not account automatically for errors in outcome definition, data capture, or analysis.

The posted interval for the risk difference runs from 0.0 to 0.0. The registry does not identify the interval method or explain the identical limits. With event counts matching each arm size, the usual statistical behavior can depend strongly on the chosen method and on how the observations are represented. Without the method and underlying data, the displayed interval cannot be independently interpreted as a complete measure of uncertainty.

What does the p-value mean?

In general, a p-value measures how unusual data at least as extreme as those observed would be under a specified null model and test procedure. It depends on the hypothesis, the assumed probability model, and the analysis choices. It does not measure the size or clinical importance of an effect, and it is not the probability that a treatment has no effect.

The registry reports p = 1.0. The test and its assumptions are not reported in the registry record, so the p-value should be described as a posted result rather than as a fully documented analysis. A p-value cannot by itself demonstrate equivalence. Equivalence or non-inferiority requires a prespecified margin and an analysis designed to evaluate that question; such a margin and analysis are not reported in the registry record.

How does randomization relate to analysis?

In general, random assignment supports a fair comparison by making treatment assignment independent of participants’ characteristics in expectation. An analysis based on assigned groups can preserve that advantage, but the registry does not report whether the mortality comparison followed an intention-to-treat approach or another analysis population. It does not specify how treatment discontinuation, switching, or deviations were handled. Those choices can affect what the comparison estimates.

For event outcomes, a simple comparison of proportions is one possible approach, while a time-to-event analysis can use information about the timing of events and censoring. The registry describes follow-up to a common study closing date, but does not report a time-to-event analysis, censoring rules, or a model. The record therefore does not show whether the posted comparison accounts for timing or differing observed follow-up.

What does the planning illustration mean?

Planning calculations use assumptions to illustrate the amount of information a future study might need to detect a chosen effect under a specified design. The registry-related planning assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with effects set as fractions of the observed effect. They are explicitly planning illustrations computed from the observed effect, not the trial’s own design justification.

Such illustrations are not observed power and do not establish that the trial had adequate power. Their interpretation depends on the effect used as a planning target and on the assumptions being reasonable for the intended study. Because the posted observed effect is 0.0 and the outcome counts are difficult to interpret, planning illustrations based on that effect should not be mistaken for evidence about the trial’s ability to detect a clinically meaningful difference. The trial’s own sample-size justification is not reported in the registry record.

10. Confidence Intervals and Precision

Precision describes how much uncertainty surrounds an estimated effect. A confidence interval is commonly used to convey that uncertainty: a narrower interval indicates less sampling uncertainty under the method used, while a wider interval indicates more. Precision is distinct from accuracy. A result can be reported with a narrow interval yet still be misleading if the outcome is misclassified, the analysis is inappropriate, or the underlying data are incomplete.

For mortality, the registry reports a risk difference of 0.0 and a confidence interval from 0.0 to 0.0. The two arm-level proportions are both 1.0. The interval’s identical limits imply no displayed range around the estimate, but the registry does not state the interval method or explain why it has this form. Given that the event counts equal the reported arm sizes, the displayed precision cannot be assessed independently from the registry entry.

A useful assessment of precision would ordinarily consider the number of events, the denominators, the amount of follow-up, and the statistical method. The registry provides arm sizes and event counts, and describes a planned common closing date intended to provide a minimum of 1 year of follow-up. It does not report actual follow-up completeness, the timing of events, censoring, or the handling of missing outcomes. These omissions limit what can be concluded about uncertainty.

The reported p-value of 1.0 does not substitute for a confidence interval, and neither statistic answers whether an observed difference would matter clinically. The record does not report a threshold for a clinically important mortality difference. The posted estimate and interval therefore describe the registry’s comparison, but do not establish precise knowledge of the underlying treatment difference.

11. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or analyses are tested. Repeated testing can increase the chance of obtaining a statistically notable result by chance, so a prespecified analysis plan may identify a primary outcome and explain whether or how other tests are handled. The registry lists one outcome measure, mortality, and no posted secondary outcomes. It lists no posted analyses. The record does not report whether additional outcomes or analyses were collected or considered outside the posted results.

Missing-data handling is also important. If outcome status is unknown for some participants, the analysis needs to state how those observations are treated and whether assumptions are made about their outcomes. The registry does not report missing mortality data, an imputation method, or a sensitivity analysis. The planned follow-up description does not establish that outcome status was available for every participant.

The analysis population is not described. In general, an intention-to-treat analysis compares participants according to their randomized assignment, while other analysis populations may exclude participants or use treatment received. These choices answer different questions and can affect the comparison. The registry does not report which population was used, how assignment deviations were handled, or whether participants who stopped or changed treatment remained in the analysis.

The registry’s mortality counts equal the listed arm sizes, but it does not explain how those counts were generated. This makes it especially important to know the event definition, data completeness, and analysis population. The record does not provide enough information to determine whether missingness or exclusions contributed to the posted result.

12. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: opportunistic infection (57.5%).

The registry lists serious adverse events for both arms. For foscarnet, 60 participants were affected among 107 at risk; for ganciclovir, 73 were affected among 127 at risk. The record also identifies Opportunistic Infection as the top serious adverse event term and lists 57.5%. It does not explain the denominator or calculation behind that percentage, so it should be reported as listed rather than interpreted as a group-specific rate.

These figures are descriptive registry counts. They do not, by themselves, establish whether the arms differ in safety. A comparative safety interpretation would require clear event definitions, observation periods, handling of repeated events, and an analysis that accounts for the groups being compared. Those details are not reported in the registry record. Serious adverse events can also differ in severity, timing, and relation to treatment; the posted summaries do not provide those details.

The record lists zero other adverse events in each group, but does not provide further explanation of what was included in that category or how reports were collected. Deaths are not reported in the safety-group entries. These omissions mean the adverse-event summary should not be treated as a complete account of all harms or as a causal comparison between treatments.

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

The registry result means that the posted mortality comparison reports equal event proportions in the two randomized arms: 1.0 in each group. The risk difference is 0.0, the confidence interval is from 0.0 to 0.0, and the reported p-value is 1.0. These are the values recorded for the comparison, not an independent reconstruction of the trial’s mortality analysis.

The result does not establish that foscarnet and ganciclovir are equivalent in mortality. A failure to report a difference is not the same as evidence that any difference is too small to matter. Demonstrating equivalence would require a prespecified margin and an appropriate analysis, neither of which is reported in the registry record. The result also does not show that either drug is effective or ineffective, or establish that one is safer than the other.

The principal reason for caution is that the posted mortality count equals the number of participants in each arm. The registry does not explain this correspondence, provide a detailed event definition, or give participant-level data. It is therefore not possible to confirm from the record alone how the mortality result should be understood. The displayed interval also has identical limits, and the method used to calculate it is not reported.

The serious adverse-event summaries provide additional descriptive information, including the counts affected in each arm and a listed leading term. They do not supply a documented comparative analysis. The record also lacks details about missing outcomes, analysis populations, and how safety events were assessed.

Overall, the posted data document a randomized comparison and a set of registry results, but leave important questions about outcome coding and analysis unresolved. The numbers support describing what the registry reports; they do not support a confident treatment conclusion. Additional protocol, analysis, and outcome information is not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because it illustrates the distinction between a numerical result and an interpretable result. The registry gives a risk difference, confidence interval, and p-value for mortality, but the reported event counts match the arm sizes. A comparison can be calculated and displayed while still leaving uncertainty about what the underlying counts mean. Statistical interpretation therefore depends on checking the outcome definition and data structure, not only reading the estimate and p-value.

The trial also shows why randomization is important but not sufficient on its own. Random assignment can support a valid comparison, yet readers still need to know how outcomes were measured, which participants were included in the analysis, and how follow-up and missing observations were handled. Those details are not reported in the registry record. Without them, the benefits of the randomized design cannot be fully assessed from the posted outcome summary.

Confidence intervals and p-values answer different questions, and neither should be interpreted in isolation. A p-value of 1.0 does not demonstrate equivalence, and a confidence interval from 0.0 to 0.0 does not remove concerns about data quality or method. Clear reporting of the statistical procedure and its assumptions is necessary to understand what the displayed values can support.

Finally, planning illustrations should be separated from evidence about the completed trial. A calculation based on an observed effect can be useful as a design exercise, but it is not observed power and does not replace a prospective sample-size rationale. The record does not report the trial’s own sample-size justification. This distinction matters whenever retrospective planning calculations accompany a trial result.

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 registry record leaves several trial-specific questions unanswered. It does not explain why the posted mortality counts equal the number of participants in each arm, and it provides no participant-level data or fuller definition of how mortality was recorded. This is a substantial limitation for interpreting the primary outcome and the reported risk difference.

The record describes a common closing date intended to provide a minimum of 1 year of follow-up, but does not report actual follow-up completeness, the timing of deaths, or censoring. It does not state whether the analysis treated mortality as a simple event count or used time-to-event methods. The method used for the reported confidence interval and p-value is also not reported.

Further details about the randomized comparison are absent, including the allocation procedure, who was masked, the analysis population, and handling of treatment changes or deviations. The record does not report missing-outcome procedures, an analysis plan, or a sample-size justification. It lists no posted analyses and no posted secondary outcomes.

The safety summary gives serious adverse-event counts and a leading event term, but does not fully define the reporting period, event attribution, or the denominator associated with the listed 57.5%. These limits prevent a detailed comparative assessment of safety from the registry summary alone. Together, the gaps mean that the posted results should be treated as registry-level information rather than a complete account of the trial’s methods and outcomes.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Studies of the Ocular Complications of AIDS (SOCA)--Foscarnet-Ganciclovir CMV Retinitis Trial (FGCRT) — NCT00000136.
• Registry record last updated 2015-10-22; status last verified 2015-07.
• Results first posted 2015-10-22.
• 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 describes a completed phase 3 randomized, single-masked, parallel-group trial comparing foscarnet with ganciclovir as initial treatment for CMV retinitis. It lists 107 participants in the foscarnet arm and 127 in the ganciclovir arm. The primary outcome is mortality, with follow-up described to a common closing date intended to provide a minimum of 1 year for all enrolled patients.

The posted comparison reports a risk difference of 0.0, a confidence interval from 0.0 to 0.0, and p = 1.0. The mortality counts equal the listed arm sizes, and the registry does not explain this. Serious adverse-event summaries are also listed, but methodological detail is limited. No analysis plan, analysis population, missing-data approach, or sample-size justification is reported in the registry record.