HIV and Kidney Function Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for HIV and Kidney Function Study (NCT00905151), a registered phase not stated study. The primary outcome is Performance of Glomerular Filtration Rate (GFR) Estimating Equations.

2. Clinical Question

3. Trial Design

The registry identifies NCT00905151 as the HIV and Kidney Function Study, with the official title “Evaluation of the Performance of GFR Estimating Equations in HIV Positive Patients.” Tufts Medical Center is listed as the sponsor. The record gives a start of 2009-01, primary completion of 2010-06, and status of COMPLETED. Enrollment is listed as 206.

The design is described as a cross-sectional analysis of HIV-positive patients with varying levels of kidney function. The arm description refers to 200 HIV-positive patients, while the enrollment field reports 206. The registry record does not explain this difference or state how many participants contributed usable measurements to the primary analysis. Allocation is not stated. The record explicitly characterizes the study as not being a randomized comparison, and its arm description does not establish a like-for-like control group.

The study’s purpose was to evaluate methods of assessing kidney function in people with HIV. The registry background notes concern that HIV and medications could affect the accuracy of existing assessment methods. It also describes the investigators’ hypothesis that tenofovir reduces creatinine secretion, raising blood creatinine without changing actual GFR. This is a stated rationale, not a reported finding. The registry does not describe an assigned tenofovir exposure, a comparison by medication use, or a prespecified analysis testing that hypothesis.

Blood samples for plasma iohexol clearance were taken at approximately 10, 30, 120, and 240 minutes after the iohexol dose, according to the primary-outcome timeframe. The registry does not specify the estimating equations, participant selection procedures, or detailed measurement and analysis protocols. It lists one outcome measure and no analysis entries.

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 Performance of Glomerular Filtration Rate (GFR) Estimating Equations, measured over Blood samples for plasma iohexol clearance were taken at approximately 10, 30, 120, and 240 minutes post-iohexol dose..

The primary outcome is titled “Performance of Glomerular Filtration Rate (GFR) Estimating Equations.” It is expressed in mL/min per 1.73 m^2 and uses the median as its summary measure, with a 95% confidence interval. The outcome description defines overall bias as the median difference between measured and estimated GFR across all estimated-GFR levels. Positive values indicate that estimated GFR is lower than measured GFR.

Measured GFR is based on plasma iohexol clearance. The record states that blood samples for this measurement were taken at approximately 10, 30, 120, and 240 minutes after an iohexol dose. The estimating equations are the methods under evaluation, but their names and details are not reported in the registry record.

Comparing estimates with a measured reference can reveal systematic differences that a measure of association alone may not capture. A method can track variation in measured GFR while still tending to produce estimates that are consistently higher or lower. Median bias describes the direction and typical size of the signed differences, while the confidence interval conveys uncertainty around the reported median. These are general interpretations of the statistical measures. The numerical median bias and confidence interval for this study are not reported in the registry record, so the direction, magnitude, and precision of the observed result cannot be assessed from the posted record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does median bias describe?

In general, bias compares an estimate with a reference measurement. The registry defines the primary outcome as the median difference between measured and estimated GFR across all estimated-GFR levels. Its sign convention matters: a positive value indicates that the estimate underestimates measured GFR. A value on the other side of zero would indicate overestimation under that same definition. The registry record does not report the observed median, so neither direction nor magnitude can be stated as a result.

The median summarizes the middle of a distribution of differences and is less sensitive to extreme individual differences than the arithmetic mean. That can be useful when differences are skewed or contain unusually large values. This general property does not show that the differences in this study were skewed, nor does it establish that a median was preferable for these data; the distribution and rationale for choosing the summary are not reported in the registry record.

What does the confidence interval contribute?

A confidence interval expresses uncertainty around an estimated quantity under a specified statistical procedure. In general, a narrower interval indicates greater precision than a wider interval, while the interval’s location indicates which values remain compatible with the analysis. The primary endpoint is specified with a 95% confidence interval, but its bounds are not reported in the registry record. Readers therefore cannot judge precision or determine whether the interval includes values near zero. A confidence interval is not a guarantee that the true value lies within the interval, and it does not by itself establish clinical importance.

How does the reference measurement fit the question?

The study uses plasma iohexol clearance as the basis for measured GFR and compares that measurement with estimates from GFR equations. The registry states that blood samples were taken at approximately 10, 30, 120, and 240 minutes after the iohexol dose. In general, a reference measurement gives an evaluation method a benchmark against which its estimates can be compared. The record does not describe the calculation procedure, the exact timing achieved for participants, or how missing or unusable measurements were handled.

Comparing estimated and measured values can address agreement and systematic error, rather than merely whether people with higher estimates also tend to have higher measured values. General statistical reasoning distinguishes association from agreement: strong association does not rule out consistent overestimation or underestimation. The registry identifies median bias as the primary summary but does not report additional agreement measures, equation-specific results, or the distribution of individual differences.

Why can over-dispersion matter?

In general, over-dispersion means that the observed variability is greater than a statistical model assumes. If a model understates variability, standard errors can be too small and confidence intervals can be too narrow; inferential tests may then appear more conclusive than warranted. This is a general concept, not a reported feature of this study. The registry does not identify a count model, a dispersion assessment, or an adjustment for over-dispersion. The primary endpoint is a median difference, and the analysis details needed to assess its variance estimation are not reported.

What can be inferred from the analysis record?

The registry lists one outcome measure and no analysis entries. The outcome definition identifies the median difference and 95% confidence interval, but it does not supply the numerical estimate or explain the analysis implementation. The estimating equations, inclusion criteria for the outcome analysis, handling of incomplete measurements, and any covariate adjustments are not reported in the registry record. It is therefore not possible to establish whether the posted endpoint represents an unadjusted or adjusted analysis, or to reproduce it from the registry description.

The record describes a cross-sectional analysis and does not describe randomized allocation. General statistical interpretation should therefore remain focused on measurement performance within the studied population, not causal effects of treatment. The stated tenofovir hypothesis does not by itself specify an exposure comparison or establish a causal analysis. Whether medication use was collected or analyzed is not reported in the registry record.

How should planning calculations be read?

In general, a sample-size calculation asks how many participants would be needed under stated assumptions to detect a specified effect with a chosen type-I error rate and power. The record describes planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. These assumptions describe illustrations, not the study’s documented design justification.

Such calculations do not establish the probability that the completed study detected an effect, and they are not “observed power.” They depend on their assumptions, including the effect used for planning. The record does not report a sample-size justification from the trial’s own design. Also, the described study is a cross-sectional evaluation without a stated two-arm randomized comparison, so the equal-allocation illustration should not be read as a description of actual allocation or enrollment.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no analysis entries. The primary outcome is identified, but the record does not state whether additional analyses were planned, whether there were multiple estimating equations or subgroups to compare, or how any multiplicity was addressed. It is therefore not possible to determine whether adjustments for multiple testing were used or whether the primary measure was selected from a broader set of analyses. Those details are not reported in the registry record.

The primary-outcome definition and measurement timeframe are present, but the record does not describe missing-data procedures. It does not state how participants with absent or incomplete blood samples, unavailable iohexol measurements, or missing equation inputs were handled. The number contributing to the primary analysis is not reported. Enrollment is 206, while the arm description refers to 200 HIV-positive patients; the record does not explain the difference or say whether either figure represents the analysis population.

Analysis-population definitions are also not reported. The registry does not specify whether the endpoint used all enrolled participants, a subset with complete measurements, or another population. It does not describe exclusions, protocol deviations, or sensitivity analyses. These omissions limit assessment of how representative the reported endpoint would be of enrolled participants, even if a numerical result were available.

In general, multiplicity concerns arise when multiple hypotheses or outcomes are tested, because the chance of a false-positive finding can increase across tests. Missing-data methods matter because conclusions can depend on why observations are absent and how the analysis handles them. These are general methodological considerations, not claims that multiplicity or missingness occurred in this study.

11. 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.

The registry’s safety entry lists 206 participants at risk for serious events and 206 at risk for other events in the HIV Positive group. It records 0 participants affected for serious events and 0 affected for other events. These figures describe what the registry reports for the listed categories; they do not establish that every possible event was captured or that no event occurred outside the reported definitions or collection period.

The deaths value is null, so the number of deaths is not reported in the registry record. No serious-event terms are listed among the top serious terms. The record does not provide event descriptions, severity grading beyond the serious/non-serious grouping, timing, relatedness assessments, or a detailed safety-collection method. It also does not describe a comparator group, consistent with the non-randomized, single-arm description.

These safety counts should be read separately from the primary GFR measurement question. The registry does not report numerical primary-outcome findings, and the safety entry cannot answer whether estimating equations corresponded closely to measured GFR. Likewise, the listed zero affected participants should be described as a registry entry, not expanded into a general claim of absence of harm.

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

The registry record establishes that the study was designed to evaluate GFR estimating equations in people with HIV, using plasma iohexol clearance as the measured-GFR reference. It defines the primary measure as median bias, with positive values indicating underestimation of measured GFR, and specifies a 95% confidence interval. The record also identifies a cross-sectional analysis and a sampling schedule for the reference measurement.

What the record does not establish is the numerical performance of any equation. The median difference, confidence interval bounds, equation-specific results, and analysis details are not reported in the registry record. It is therefore not possible to tell from the posted record whether the estimates were biased in either direction, how large any bias was, or how precise the estimate was. No conclusion about an equation’s accuracy in this study can be drawn from the outcome definition alone.

The registry’s background describes a hypothesis that tenofovir could reduce creatinine secretion and raise blood creatinine without changing actual GFR. That statement is a proposed mechanism motivating the study, not a result. The record does not document an intervention assignment or a comparison that would establish an effect of tenofovir. The design description should not be interpreted as a randomized treatment evaluation.

More generally, evaluating agreement with a reference measurement is different from showing that one measure is associated with another, and neither alone establishes that a method is suitable for every population or setting. Such distinctions are general statistical cautions. The registry does not report enough outcome detail to assess the equations’ agreement, variation across GFR levels, or applicability beyond the described HIV-positive population.

The enrollment figure and arm description are not identical: enrollment is 206, while the arm description refers to 200 patients. Because the registry does not reconcile them or identify the endpoint analysis population, the number contributing to the primary analysis is uncertain. The completed status indicates that the record lists the study as completed; it does not substitute for missing numerical results.

13. Why This Trial Matters Statistically

This study matters statistically because it frames kidney-function assessment as a measurement problem. An estimating equation is useful only insofar as its values correspond to a relevant reference measurement for the population in which it is applied. Comparing estimated GFR with measured GFR can reveal systematic bias that might be obscured if analysis considered association alone. The registry’s choice of a signed median difference makes the direction of bias explicit: positive values are defined as underestimation of measured GFR.

The population context is central to the question. The registry background raises the possibility that HIV and medications could affect kidney-function assessment, and it describes a specific hypothesis involving tenofovir and creatinine secretion. This motivates evaluation rather than proving that an equation behaves differently in this population. The study’s cross-sectional description is suited to a measurement comparison at the observed assessment, but it is not described as a randomized test of a medication’s causal effect.

Precision is also part of a useful evaluation. A median difference without its uncertainty would provide limited information about how stable the estimated bias is. The endpoint specifies a 95% confidence interval, although its values are not reported in the registry record. The absence of numerical results means readers cannot assess the size or precision of any observed bias, despite the clear statistical question and outcome definition.

The record also illustrates why a design description and a result need to be distinguished. It identifies the reference method, sample timing, enrollment, and primary measure, but lists no analysis entries. Its arm description and enrollment figure differ, and the analysis population is not stated. These details affect how readers would assess representativeness and reproducibility if the numerical outcome were available. The statistical contribution is therefore clearest in the question it sets up; the registry record does not provide the result needed to quantify the answer.

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 most important limitation of the registry record for interpretation is that numerical primary-outcome results are not reported. There is no median bias or 95% confidence interval, so neither direction nor magnitude nor precision can be evaluated. The registry lists no analysis entries, and it does not give separate results for particular estimating equations. The equation names themselves are not reported in the registry record.

The record’s description of the study population is not fully reconciled. Enrollment is listed as 206, while the arm description refers to 200 HIV-positive patients. The number included in the primary analysis and reasons for any difference are not reported. Eligibility criteria, participant characteristics, and the distribution of kidney function are also not described in the information available in the registry record, limiting assessment of how broadly the results might apply.

Important analytic details are absent. The registry does not report how the median difference and confidence interval were calculated, whether any adjustments were made, how incomplete measurements were handled, or whether sensitivity analyses were conducted. It does not describe multiplicity procedures or define the analysis population. The stated sampling times are approximate, and details of the measurement procedure beyond those times are not reported.

The cross-sectional design does not establish the temporal or causal effect of a medication. Although the background states a tenofovir hypothesis, the record does not describe an exposure comparison or report an analysis of that hypothesis. Allocation is not stated, and no like-for-like control comparison is described. The record also does not provide a sample-size justification from the trial’s own design. These limitations concern what can be learned from the registry record; they do not determine what analyses may have been conducted outside it.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: HIV and Kidney Function Study — NCT00905151.
• Registry record last updated 2017-03-15; status last verified 2017-02.
• Results first posted 2017-02-06.
• 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

NCT00905151 is listed as the HIV and Kidney Function Study, sponsored by Tufts Medical Center, with status COMPLETED. The record gives a start of 2009-01, primary completion of 2010-06, and enrollment of 206. It describes a cross-sectional analysis of HIV-positive patients with varying levels of kidney function; the arm description refers to 200 patients.

The primary outcome evaluates GFR estimating equations using median difference between measured and estimated GFR, expressed in mL/min per 1.73 m^2, with a 95% confidence interval. Measured GFR is based on plasma iohexol clearance, with blood samples taken at approximately 10, 30, 120, and 240 minutes after the dose. Positive bias values indicate underestimation of measured GFR.

The registry lists one outcome measure and no analysis entries. Numerical primary-outcome results, equation-specific results, and analysis-population details are not reported in the registry record. The safety entry records 0 affected participants for serious events and 0 for other events among 206 at risk in each category; deaths are not reported.