A Retrospective Evaluation of the Intraocular Pressure Spikes With Loteprednol and Loteprednol/Tobramycin: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Retrospective Evaluation of the Intraocular Pressure Spikes With Loteprednol and Loteprednol/Tobramycin (NCT00834171), a registered phase not stated study. The primary outcome is Mean Elevated Intraocular Pressure (IOP) During Treatment.

2. Clinical Question

3. Trial Design

This completed study is described as a retrospective evaluation based on patient-chart review. Its stated purpose was to evaluate the incidence of IOP spikes while patients were using loteprednol etabonate ophthalmic suspension 0.5% or loteprednol etabonate (0.5%) and tobramycin (0.3%). The registry lists enrollment of 50 and gives a start date of 2009-01 and a primary completion date of 2009-12. It names Allergan as the sponsor.

The registry identifies two treatment groups, but does not state how treatment was allocated. It explicitly cautions that this was not a randomized comparison and that the arms are not a like-for-like control comparison. As a retrospective chart review, the record describes evaluation of existing patient charts rather than a stated randomized assignment. It does not report further details about chart selection, treatment choice, follow-up procedures, or how patients came to receive either treatment.

The primary outcome is a continuous measure: mean elevated IOP during treatment, reported in millimeters of mercury. The description says that IOP was recorded when an elevation of 5 mmHg or more occurred while using study treatment; the median treatment duration at the time of observed elevation was 55 days. The primary population is described as intent to treat and includes all patients in the study. The registry lists one outcome measure and zero analyses. Numerical primary-outcome results and a statistical comparison 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

Analysis population for the primary outcome, as defined in the registry: Intent to treat, which included all patients 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 Mean Elevated Intraocular Pressure (IOP) During Treatment, measured over 55 days.

The primary endpoint is titled mean elevated intraocular pressure (IOP) during treatment, with units of millimeters of mercury (mmHg). The record defines an IOP elevation as an increase of 5 mmHg or more while a patient was using study treatment. It explains that IOP measures the fluid pressure inside the eye. The stated timeframe is 55 days, and the endpoint description says that the median treatment duration at the time of observed IOP elevation was 55 days.

A mean is a measure of the average value of a continuous outcome. It can describe the level of IOP elevation in a group, while a standard deviation describes how spread out individual measurements are around that group mean. In general, a continuous endpoint can retain information about the magnitude of measurements rather than reducing them to a simple event count. That is useful only when the measurement and the way it was collected are adequately described.

The registry identifies the endpoint as a mean and names standard deviation as its dispersion measure, but does not report either group’s mean or standard deviation. It also does not report a confidence interval or a statistical comparison. Consequently, the endpoint’s observed values and their precision cannot be evaluated from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry classifies the primary outcome as a continuous measure summarized by a mean, with standard deviation as the dispersion measure. In general, the mean is the arithmetic average of recorded values. The standard deviation describes the degree to which individual values vary around that average. A group mean alone does not show the variation among patients; the standard deviation adds information about that spread. Neither measure gives a complete description of a distribution, particularly if values are strongly skewed or include influential observations.

The endpoint’s units are millimeters of mercury. Its description defines an elevation as an increase of 5 mmHg or more while the patient was using study treatment, and states that the median treatment duration at the time of observed elevation was 55 days. A threshold can identify which measurements qualify as elevated, but the registry does not explain the reference measurement used to establish an increase, the timing or frequency of measurements, or the method for combining measurements into each patient’s value. Those operational details are not reported in the registry record. That limits the ability to reproduce or fully assess the endpoint from the description alone.

In general, an estimated difference between group means would describe how far apart the group averages are, and a confidence interval would show the uncertainty around that difference. A confidence interval depends on the observed data and the analysis model; it is not a guarantee that a particular value is correct. The registry does not report group means, a difference in means, standard errors, confidence intervals, or a p-value for the primary endpoint. No primary-outcome inferential analysis is reported, so there is no statistical estimate whose precision can be interpreted.

Because the study is described as retrospective and nonrandomized, a comparison of the two treatment groups would not by itself isolate a treatment effect. In general, treatment selection may be associated with patient features, clinical circumstances, or other factors that also relate to an outcome. Randomization is one design method for balancing such factors in expectation; the registry does not describe random allocation. It also does not report an adjusted analysis or the covariates that might have been considered. The record specifically warns that the arms are not a like-for-like control comparison.

The registry’s sample-size assumptions describe a planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. This is a planning exercise, not evidence about how the study was designed, and it is not observed power. The registry does not report a sample-size justification for the retrospective study. More generally, power calculations rely on assumptions about an effect and variability; they do not establish that an observed estimate is reliable or that a study’s analysis was unbiased.

General statistical methods also need to match the structure of the data. If the outcome consists of repeated measurements from a patient, those measurements may be correlated rather than independent. Treating correlated observations as independent can understate uncertainty. The registry does not report whether repeated IOP measurements were present or how they were handled. Over-dispersion is a concern for count outcomes when variability exceeds what a simple count model assumes; it is not a substitute for describing the continuous primary endpoint, and no count-model analysis is reported.

The registry lists one outcome measure and zero analyses. It does not report a statistical analysis plan, the model used for the primary endpoint, assumptions checked, or methods for uncertainty estimation. No recalculated primary-outcome statistics are available because the necessary group summaries are not reported. Accordingly, the available record supports describing the intended endpoint and design limitations, but not reconstructing an effect estimate or a test result.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Loteprednol Etabonate Ophthalmic Suspension 0.5%", "Loteprednol Etabonate (0.5%) and Tobramycin (0.3%)")
n   <- c(43, 7)
m   <- c(24.8, 23.9)
sp  <- c(6.9, 3.5)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Loteprednol Etabonate Ophthalmic Suspension 0.5%  43  24.800  6.900  22.676  26.924
CARM  Loteprednol Etabonate (0.5%) and Tobramycin (0.3%)  7  23.900  3.500  20.663  27.137

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure, the mean elevated IOP during treatment, and zero analyses. This does not establish whether the study protocol considered additional outcomes or analyses; those details are not reported in the registry record. A multiplicity plan, including whether any adjustment was intended for multiple outcomes, time points, subgroups, or tests, is also not reported.

In general, when many statistical tests are conducted, the chance of at least one apparently notable finding can increase. Methods for controlling that issue depend on which outcomes and comparisons were planned and how they relate to one another. The registry does not provide enough information to determine whether multiplicity was relevant to the investigators’ analyses or whether any adjustment was used. Since no primary-outcome analysis is reported, no multiplicity-adjusted or unadjusted inference can be assessed.

The primary population is described as intent to treat, including all patients in the study. The record does not report how that population was applied to the retrospective chart data, whether each patient had a usable IOP measurement, or how missing or incomplete records were handled. It does not report imputation, exclusions from the primary endpoint, or a sensitivity analysis. The adverse-event counts have stated at-risk denominators, but the registry does not explain the relationship of those denominators to enrollment or describe missing adverse-event information.

These omissions matter because the analyzed population and handling of incomplete data can affect an estimate. The intent-to-treat label identifies the stated population, but does not by itself establish that outcome data were complete or that missingness was addressed in a particular way.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists serious adverse events separately for the two treatment groups. For loteprednol etabonate ophthalmic suspension 0.5%, it reports 0 serious events among 43 at risk. For loteprednol etabonate (0.5%) and tobramycin (0.3%), it reports 0 serious events among 7 at risk. The registry does not list serious-event terms. Death counts are not reported in the registry record.

For other adverse events, the record reports 4 affected among 43 at risk in the loteprednol-only group and 0 among 7 at risk in the combination group. These are counts, not percentages or comparative effect estimates. No statistical test or uncertainty interval for the adverse-event counts is reported. The absence of a listed event in these records should not be taken to establish that an event is impossible or that the groups have equivalent safety.

The registry cautions that allocation was not stated and that the arms are not a like-for-like control comparison. Differences in these descriptive counts therefore cannot be attributed to treatment on the basis of this record. The available group sizes and event counts also provide limited information for evaluating uncommon events. The registry does not report additional details about how adverse events were collected, their severity beyond the serious-event classification, or their timing.

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

The record establishes that the study was a retrospective chart review focused on IOP elevations during use of two named treatment regimens. It defines an elevation as an increase of 5 mmHg or more during study treatment and identifies mean elevated IOP as the primary endpoint. It also reports adverse-event counts for the two groups. These registry entries describe the study question and the information entered into the registry; they do not provide numerical primary-outcome results.

Without group means, standard deviations, an estimated contrast, and an uncertainty measure, the primary endpoint cannot be interpreted as indicating a higher or lower average elevation in either group. The registry does not report a statistical test of that endpoint. It is therefore not possible to characterize a primary-outcome difference as statistically significant or nonsignificant, or to judge how precisely such a difference was estimated.

The adverse-event counts are descriptive and should be read in light of the registry’s warning that this was not a randomized comparison and that allocation was not stated. The counts do not establish a causal relationship between a treatment and an event, and the record does not report an inferential safety comparison. The different at-risk counts are reported as such; the registry does not explain their relationship to the overall enrollment.

More generally, a completed study status does not mean that all results are available in the registry. Nor does a stated intent-to-treat population, on its own, describe how missing chart information was handled. The record does not provide an analysis plan, details of measurement collection, or an account of missing data. Those gaps constrain both reproducibility and interpretation.

The study can therefore be read as a registry record of a retrospective evaluation with a defined clinical measurement and limited descriptive safety counts. It cannot support a conclusion about comparative efficacy or safety from the primary outcome, because the primary results and comparative analysis are not reported in the registry record.

14. Why This Trial Matters Statistically

This record illustrates an important distinction between a defined research question and a reportable result. The registry describes a continuous primary endpoint, its measurement units, a threshold for an IOP elevation, and the stated treatment timeframe. Those details help identify what investigators intended to evaluate. But without group summary values or an analysis, readers cannot estimate the magnitude, direction, or precision of a treatment-group contrast.

It also illustrates why design information matters. The study is retrospective, and the registry explicitly states that allocation was not stated and that the groups are not a like-for-like control comparison. Descriptive differences between groups in a nonrandomized chart review can reflect differences in who received each treatment or how records were collected, as well as any treatment-related contribution. An outcome count or average alone cannot resolve those possibilities.

The entry also distinguishes safety counts from comparative safety evidence. Serious and other adverse events are recorded with group-specific at-risk counts, but no statistical comparison is reported. Those counts are useful as descriptive registry information, while their interpretation remains limited by the record’s design and reporting detail.

For clinical researchers and biostatisticians, the central lesson is that an endpoint definition, an enrollment figure, and a completed status do not replace outcome estimates, uncertainty measures, and transparent analytic methods. A useful interpretation depends on knowing what was measured, who contributed data, how missing records were handled, and how groups were compared. Several of those details are not reported in the registry record. Recognizing what is documented, and what remains unavailable, prevents descriptive information from being mistaken for a treatment-effect estimate.

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 main limitation for interpretation is that numerical results for the primary endpoint are not reported in the registry record. Although the endpoint is described as mean elevated IOP with standard deviation, neither group’s mean or standard deviation is available, and no estimated difference or uncertainty interval is reported. The registry also lists zero analyses. This prevents assessment of the primary outcome’s magnitude, variation, and statistical precision.

The design limits causal interpretation. The record describes a retrospective chart review, says allocation was not stated, and warns that the treatment arms are not a like-for-like control comparison. It does not describe treatment selection, patient characteristics, or an adjusted analysis. Potential differences between groups therefore cannot be separated from treatment effects using the information in the record.

The outcome description leaves measurement details unresolved. The record defines an elevation as an increase of 5 mmHg or more and gives the median treatment duration at the time of observed elevation as 55 days, but does not describe the reference value, measurement schedule, repeat readings, or how observations were combined into a patient-level mean. It does not report how many patients contributed a primary outcome measurement or how incomplete charts were handled.

The adverse-event entries are limited to counts and at-risk denominators. The registry does not report deaths, collection procedures, timing, or statistical comparisons; death counts are not reported in the registry record. It also does not explain the relation between enrollment and the adverse-event denominators. These limitations make the safety entries descriptive rather than a basis for a comparative conclusion.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Retrospective Evaluation of the Intraocular Pressure Spikes With Loteprednol and Loteprednol/Tobramycin — NCT00834171.
• Registry record last updated 2011-09-26; status last verified 2011-08.
• Results first posted 2011-09-26.
• 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 retrospective chart review sponsored by Allergan, with enrollment of 50, a start date of 2009-01, and primary completion in 2009-12. It concerns loteprednol etabonate ophthalmic suspension 0.5% and loteprednol etabonate (0.5%) with tobramycin (0.3%). The stated primary endpoint is mean elevated IOP during treatment, measured in mmHg, with an elevation defined as an increase of 5 mmHg or more. The outcome population is described as intent to treat.

The registry reports zero serious adverse events among 43 at risk in the loteprednol-only group and zero among 7 at risk in the combination group; other adverse-event counts are 4 among 43 and 0 among 7, respectively. It states that allocation was not reported and the groups are not a like-for-like control comparison. Numerical primary-outcome results, a statistical analysis, missing-data methods, and death counts are not reported in the registry record.