This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Rosiglitazone (Extended Release Tablets) As Adjunctive Therapy For Subjects With Mild To Moderate Alzheimer's Disease (NCT00348309), a randomized na study. The primary outcome is Change From Baseline in Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS-Cog) Total Score at Week 48.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00348309 |
| Phase | NA |
| Condition | Alzheimer's Disease |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 1496 |
| Arms in the primary outcome | n/a |
| Lead sponsor | GlaxoSmithKline |
| Results posted | Yes |
2. Clinical Question
The study asked whether adding extended-release rosiglitazone to donepezil changed cognition and overall clinical response in people with mild to moderate Alzheimer’s disease, and whether response varied across APOE genetic subgroups. The primary outcome was change in ADAS-Cog total score from baseline to Week 48. The registry record defines the endpoint and analysis population, but numerical results for this primary outcome are not reported in the registry record.
Population
People with mild to moderate Alzheimer’s disease who were receiving donepezil. The primary analysis population included randomized participants who received at least one dose of study medication and had at least one post-baseline efficacy assessment. The record describes APOE4-negative, E4 heterozygote, and full-population analyses for the primary outcome.
Intervention
Rosiglitazone extended-release tablets, given once daily at the 2mg or 8mg dose, as adjunctive therapy to donepezil.
Comparator
Placebo, with donepezil listed as background therapy in the study title. The registry record lists placebo as an arm and describes the study as placebo-controlled.
Primary question
How did ADAS-Cog total score change from baseline to Week 48 in the rosiglitazone groups compared with placebo, and how did the comparison relate to the specified genetic subgroups? Numerical between-group results are not reported in the registry record.
The official title also identifies cognition and overall clinical response as study aims. The registry lists other outcomes covering function, neuropsychiatric symptoms, global cognition, resource use, quality of life, and additional ADAS-Cog and CDR-SB assessments. The record does not provide enough primary-outcome results to describe the direction, size, or precision of the treatment comparisons.
3. Trial Design
This was a randomized, double-blind, placebo-controlled, parallel-group study of extended-release rosiglitazone as adjunctive therapy to donepezil. The registry describes masking as quadruple. It does not identify which participant or study roles were masked, so those details are not reported in the registry record. The study title specifies a 54-week duration and APOE ε4 stratification.
The registry lists three arms: rosiglitazone extended release 2mg once daily, rosiglitazone extended release 8mg once daily, and placebo. Donepezil is listed among the interventions and is described in the study title as the background therapy. The registry record does not provide further details about the dosing or continuation of donepezil.
Enrollment was 1496. The registry records the study as completed, with a start date of 2006-07-06 and primary completion date of 2009-01-01. The phase is listed as NA. These administrative details describe the trial’s registration record; they do not establish how many participants contributed data at each assessment.
The primary endpoint was change from baseline in ADAS-Cog total score at Week 48, using the Week 0 baseline. The endpoint description states that change was calculated as the Week 48 value minus baseline. It also says the primary outcome was analyzed in APOE4-negative participants, E4 heterozygotes, and the full population. The analysis population included randomized participants who received at least one dose and had at least one post-baseline efficacy assessment, with only participants available at the specified time points analyzed.
The record does not report the planned sample-size justification, detailed allocation procedures, treatment adherence, or protocol deviations. It also does not report how the planned 54-week follow-up relates to the Week 48 primary endpoint and other assessment times beyond the listed outcome timeframes.
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.
Before assignment: A total of 1496 participants were randomized for the study, out of which seventeen participants did not receive study medication. A total of 1479 were included in the Safety population.
Overall Study
| Stage | Placebo | 2mg Rosiglitazone Extended Release | 8mg Rosiglitazone Extended Release |
|---|---|---|---|
| STARTED | 496 | 494 | 489 |
| COMPLETED | 362 | 396 | 346 |
| NOT COMPLETED | 134 | 98 | 143 |
Reasons for not completing this period:
| Reason | Placebo | 2mg Rosiglitazone Extended Release | 8mg Rosiglitazone Extended Release |
|---|---|---|---|
| Adverse Event | 56 | 28 | 63 |
| Lost to Follow-up | 6 | 7 | 4 |
| Protocol Violation | 7 | 12 | 7 |
| Withdrawal by Subject | 33 | 37 | 40 |
| Non-compliance | 8 | 6 | 8 |
| Death of caregiver | 1 | 0 | 1 |
| Lack of insight, dementia | 1 | 0 | 0 |
| Wrong bottle allocated | 1 | 0 | 0 |
| Caregiver sick | 1 | 0 | 0 |
| Use of prohibited medication | 5 | 2 | 2 |
| Caregiver withdrawal | 3 | 0 | 7 |
| Unable to complete ET visit | 1 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: ITT population included all the participants who were randomized to treatment, who had received at least one dose of study medication and who had at least one post baseline efficacy assessment. Only those participants available at the specified time points 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.
| Characteristic | Unit | Placebo | 2mg Rosiglitazone Extended Release | 8mg Rosiglitazone Extended Release | Total |
|---|---|---|---|---|---|
| Participants analysed | 496 | 494 | 489 | 1479 | |
| Age, Continuous | Years | 74.2 (SD 7.95) | 74.5 (SD 8.06) | 74.4 (SD 7.83) | 74.4 (SD 7.94) |
| Sex: Female, Male: Female | Participants | 304 | 286 | 305 | 895 |
| Sex: Female, Male: Male | Participants | 192 | 208 | 184 | 584 |
| Race/Ethnicity, Customized: Hispanic or Latino | Participants | 97 | 86 | 82 | 265 |
| Race/Ethnicity, Customized: Not Hispanic or Latino | Participants | 399 | 408 | 407 | 1214 |
6. Primary Endpoint
The registry lists the primary outcome as Change From Baseline in Alzheimer's Disease Assessment Scale - Cognitive Subscale (ADAS-Cog) Total Score at Week 48, measured over Baseline (Week 0) and Week 48.
| Element | Registry entry |
|---|---|
| Result type | Least Squares Mean |
| Measure of spread | Standard Error |
| Unit | Score on a scale |
| Endpoint classification | Continuous Endpoint |
The primary endpoint was change from baseline in the Alzheimer’s Disease Assessment Scale–Cognitive Subscale (ADAS-Cog) total score at Week 48. The record identifies the baseline as Week 0 and defines change as the Week 48 score minus the baseline score. ADAS is described as a performance-based assessment of specific cognitive and behavioral dysfunctions in participants with Alzheimer’s disease.
The ADAS-Cog contains 11 items summed to a total score ranging from 0 to 70. Lower scores indicate less severe impairment. As a general statistical point, a change score expresses how an assessment differs from its baseline value; interpreting a group comparison therefore requires the estimated changes and their uncertainty, not merely the scale definition. Those numerical primary-outcome results are not reported in the registry record.
The record describes the reported quantity as a least-squares mean, an adjusted mean, with standard error as the dispersion measure. In general, an adjusted mean is an estimate from a statistical model rather than a simple arithmetic average. The registry does not report the model specification or the numerical estimates and standard errors for this endpoint.
Cognition is the primary domain named by this endpoint. The registry also lists functional, behavioral, quality-of-life, resource-use, and other cognitive outcomes. These measures address different aspects of illness and should not be treated as interchangeable with the primary cognitive score.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record establishes the trial question and the primary endpoint, but it does not report numerical ADAS-Cog results. Consequently, the direction and magnitude of change in each arm, the between-group estimates, and the precision of the comparisons are not reported in the registry record. No conclusion about the primary comparison can be drawn from the endpoint description alone.
The primary outcome used least-squares means and standard errors. In general, the estimate describes the modeled mean change, while its standard error describes the estimated sampling uncertainty of that mean. A standard error is not itself a confidence interval and does not show the full range of values compatible with the data. Numerical estimates, standard errors, confidence intervals, and p-values for the primary outcome are not reported in the registry record.
The primary outcome description refers to a hierarchical testing procedure to control testing across the two rosiglitazone dose groups and genetic subgroups. This indicates that the analyses were not described as a set of unrelated tests. In general, a hierarchy specifies an order or structure for evaluating multiple hypotheses, helping control the chance of a false-positive finding across planned comparisons. The actual hierarchy, its decision rules, and the results at each step are not reported in the registry record. The record therefore does not permit an assessment of which comparisons, if any, met the prespecified criteria.
The primary analysis population was not every randomized participant without qualification. It included participants who were randomized, received at least one dose, and had at least one post-baseline efficacy assessment. The record further says only participants available at the specified time points were analyzed. As a general point, this means the analyzed population and the number contributing at a particular visit can differ from total enrollment. Visit-specific participant counts and the extent or handling of missing data are not reported in the registry record.
The safety information lists serious adverse events, other adverse events, and deaths by arm. These are descriptive counts in the registry record. They do not, on their own, establish that an event was caused by treatment or provide a complete comparison of risk over time. The record does not give exposure-adjusted rates, statistical comparisons, or event-level attribution details. Those safety figures should therefore be read as recorded event counts, not as proof of a causal effect.
Planning illustrations based on the observed effect are described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the detectable effect is a stated fraction of the observed effect. The numerical planning results are not reported in the registry record. Such calculations are planning illustrations, not observed power, and do not substitute for the trial’s actual estimates or their uncertainty. The trial’s original sample-size justification is not reported in the registry record.
Overall, the record supports a description of design, endpoint, analysis population, and recorded safety counts. It does not report the numerical primary efficacy results needed to quantify treatment differences or judge their precision. Any interpretation of efficacy beyond that boundary would go beyond the registry record.
9. Statistical Methods Explained
The registry describes a randomized, parallel-group study with placebo control and quadruple masking. Randomization is intended, in general, to make treatment assignment independent of participants’ prognosis and other characteristics on average. Masking is intended to reduce the influence of treatment knowledge on care, assessment, or reporting. The registry does not describe the allocation process or identify the roles covered by quadruple masking.
What does a least-squares mean represent?
The primary endpoint is reported in the form of a least-squares mean, described in the record as an adjusted mean. In general, least-squares means are model-based estimates of group means, often adjusted for variables included in the model. Their interpretation depends on the model, covariates, and assumptions. The registry does not provide the model specification, covariates, or numerical estimates, so the particular adjustment cannot be reconstructed from the record.
The endpoint’s dispersion measure is the standard error. In general, a standard error measures the estimated uncertainty in an estimate across repeated samples under the model. It is different from the spread of individual participant scores. A confidence interval combines an estimate with its uncertainty to express a range of values compatible with the data under specified assumptions. The primary numerical estimates, standard errors, and confidence intervals are not reported in the registry record.
Why does the endpoint use change from baseline?
Change from baseline is calculated as the follow-up score minus the baseline score. The registry explicitly defines the primary outcome this way for Week 48 and Week 0. A change score focuses on how a participant’s measured score differs from their own starting value. For a between-group comparison, the important quantity is the contrast between the groups’ changes, interpreted alongside its uncertainty. That contrast is not reported in the registry record.
The ADAS-Cog is a continuous measure: its total score can take values across a score scale rather than being limited to a category such as event or no event. As a general statistical point, a group mean summarizes the center of a distribution but does not describe every participant’s experience or the full shape of the score distribution. The record’s definition identifies lower scores as less severe impairment; it does not report the observed group distributions or changes.
What does the hierarchical testing statement mean?
The endpoint description states that a hierarchical testing procedure was used to control for the two rosiglitazone dose groups and genetic subgroups. In general, multiplicity arises when several hypotheses are tested: considering each test in isolation can make the overall chance of at least one false-positive result larger than the nominal level for an individual test. A hierarchical procedure organizes tests according to prespecified rules, so that interpretation of later tests can depend on the outcome of earlier tests.
The registry does not report the hierarchy’s exact sequence, decision thresholds, or all test results. It is therefore not possible to identify from the record which comparisons were eligible for formal interpretation or whether any passed the stated testing procedure. The hierarchy statement should not be read as evidence that a result was statistically significant.
How should the analysis population be understood?
The primary outcome population is described as randomized participants who received at least one dose of study medication and had at least one post-baseline efficacy assessment. This is the registry’s stated definition; it is not an unqualified analysis of every randomized participant. The record also says that only participants available at the specified time points were analyzed. In general, an analysis restricted to participants with available measurements can be affected if missingness is related to health status, treatment, or the outcome. Whether that occurred cannot be assessed from the registry information.
Methods for handling missing measurements, participant withdrawals, and incomplete follow-up are not reported in the registry record. The record also does not report a sensitivity analysis for missing data. Those omissions limit assessment of how robust the primary result would be to different assumptions about unavailable measurements, quite apart from the fact that the numerical result itself is not reported.
What does over-dispersion mean in general?
For count outcomes, over-dispersion means that observed variability is greater than a simple count model expects. In general, it can affect standard errors and confidence intervals if the model does not accommodate the extra variation. The primary endpoint in this trial is a continuous score, not an event count, and the registry does not report an over-dispersion analysis. This concept is relevant as general statistical context, not as a stated feature of the primary analysis.
What do the planning assumptions describe?
The registry’s planning-assumption statement describes illustrations based on the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. In general, such calculations estimate a sample size under chosen assumptions; they are not measures of the probability that the completed study found a true effect. They are planning illustrations, not observed power. The actual numerical illustrations are not reported in the registry record, and the trial’s own sample-size justification is not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The primary endpoint description says a hierarchical testing procedure was used to control for the two rosiglitazone dose groups and genetic subgroups. This is relevant because several related comparisons can increase the chance of a false-positive finding if each is treated as independent. In general, a hierarchy controls interpretation through an ordered set of tests or rules. The precise order and decision criteria are not reported in the registry record, nor are the numerical results for the comparisons. The registry also lists 22 outcome measures and 74 analyses, but it does not describe how multiplicity was handled across all secondary outcomes and analyses.
The stated primary analysis population included randomized participants who had received at least one dose and had at least one post-baseline efficacy assessment. The record says that only participants available at the specified time points were analyzed. This is important because the number contributing to an endpoint may be smaller than enrollment, and the available participants may differ from those with missing measurements. Visit-specific analysis counts and the reasons for unavailable data are not reported in the registry record.
Methods for imputing or otherwise handling missing data, sensitivity analyses, and the assumptions underlying those methods are not reported in the registry record. The record also does not report whether an alternative analysis population was used or how protocol deviations were handled. These details matter because different approaches to incomplete follow-up can yield different estimates, especially when the likelihood of missingness is related to outcome or treatment.
The registry record does not report the planned sample-size justification, and numerical sample-size illustrations are not reported in the registry record. Its stated planning assumptions should not be interpreted as observed power or as a replacement for the study’s actual treatment estimates and uncertainty.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Placebo | 62 | 496 | 12.5% |
| 2mg Rosiglitazone Extended Release | 45 | 494 | 9.1% |
| 8mg Rosiglitazone Extended Release | 50 | 489 | 10.2% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Placebo | 12 of 496 |
| 2mg Rosiglitazone Extended Release | 6 of 494 |
| 8mg Rosiglitazone Extended Release | 8 of 489 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Placebo | 11 of 496 |
| 2mg Rosiglitazone Extended Release | 29 of 494 |
| 8mg Rosiglitazone Extended Release | 69 of 489 |
Most frequent serious adverse event terms
| Event term | Organ system | Placebo | 2mg Rosiglitazone Extended Release | 8mg Rosiglitazone Extended Release |
|---|---|---|---|---|
| Pneumonia | Infections and infestations | 5 of 496 | 1 of 494 | 6 of 489 |
| Hip fracture | Injury, poisoning and procedural complications | 3 of 496 | 4 of 494 | 1 of 489 |
| Myocardial infarction | Cardiac disorders | 5 of 496 | 1 of 494 | 0 of 489 |
| Syncope | Nervous system disorders | 1 of 496 | 3 of 494 | 2 of 489 |
| Cardiac arrest | Cardiac disorders | 2 of 496 | 2 of 494 | 1 of 489 |
| Femoral neck fracture | Injury, poisoning and procedural complications | 0 of 496 | 1 of 494 | 3 of 489 |
| Fall | Injury, poisoning and procedural complications | 1 of 496 | 1 of 494 | 1 of 489 |
| Humerus fracture | Injury, poisoning and procedural complications | 1 of 496 | 1 of 494 | 1 of 489 |
| Urinary tract infection | Infections and infestations | 1 of 496 | 0 of 494 | 2 of 489 |
| Confusional state | Psychiatric disorders | 2 of 496 | 0 of 494 | 1 of 489 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Placebo | 2mg Rosiglitazone Extended Release | 8mg Rosiglitazone Extended Release |
|---|---|---|---|---|
| Oedema peripheral | General disorders | 11 of 496 | 29 of 494 | 69 of 489 |
Serious events with the highest arm-level rate: pneumonia (1.2%), myocardial infarction (1.0%), hip fracture (0.8%), femoral neck fracture (0.6%), syncope (0.6%).
The registry lists serious adverse events, other adverse events, and deaths separately for placebo, 2mg rosiglitazone extended release, and 8mg rosiglitazone extended release. The serious adverse event counts are 62 among 496 participants at risk in the placebo group, 45 among 494 in the 2mg group, and 50 among 489 in the 8mg group. The listed death counts are 12, 6, and 8, respectively. These are registry counts; the record does not provide statistical comparisons or exposure-adjusted rates.
For other adverse events, the counts are 11 among 496 participants at risk for placebo, 29 among 494 for 2mg, and 69 among 489 for 8mg. These figures should be read as counts classified in the registry, not as estimates of treatment-caused events. The record does not provide event-level causal assessments or explain how duration of exposure was incorporated into these counts.
The registry lists pneumonia at 1.2%, myocardial infarction at 1.0%, hip fracture at 0.8%, femoral neck fracture at 0.6%, syncope at 0.6%, and urinary tract infection at 0.4% among the leading serious terms. The record does not specify in this listing how those percentages are distributed across treatment groups. These figures alone do not establish a treatment effect or a comparison between groups.
In general, safety counts describe events recorded during a study, whereas causal interpretation requires context about timing, exposure, attribution, and comparison methods. Additional safety details are not reported in the registry record.
12. What the Result Does, and Does Not, Mean
The record identifies a randomized comparison of two rosiglitazone doses with placebo as adjunctive therapy to donepezil in people with mild to moderate Alzheimer’s disease. It specifies a primary cognitive endpoint, its scale, timing, and the analysis population. It also records adverse-event and death counts by arm. These details make the research question and some elements of the analysis interpretable.
However, numerical results for the primary ADAS-Cog endpoint are not reported in the registry record. The record therefore does not establish how much scores changed in any group, whether groups differed, or how precise any treatment comparison was. It also does not report the numerical results needed to assess the secondary outcomes. Descriptions of the endpoint and planned testing procedure cannot substitute for observed estimates.
The safety counts show how many participants in each arm were listed with serious adverse events, other adverse events, and deaths. Counts alone do not indicate whether an event was caused by the study treatment. Nor do they give a complete comparison of risk over time: exposure-adjusted rates, statistical comparisons, and event-attribution details are not reported in the registry record.
The trial’s design and enrollment do not, by themselves, determine the clinical meaning of its results. Statistical interpretation depends on estimated group differences and their uncertainty, the analysis population, the handling of missing observations, and the prespecified multiplicity procedure. Several of these details, including the numerical primary results and missing-data methods, are not reported in the registry record.
Accordingly, the record supports description of what was studied and how the primary endpoint was defined, but it does not support a conclusion about the size, direction, or precision of the primary treatment comparison. It also does not support a causal conclusion from the adverse-event counts alone. Those limits apply to interpretation of the registry record and are not statements about the treatment’s effects outside the information recorded.
13. Why This Trial Matters Statistically
This trial illustrates why an endpoint definition and an observed result are different kinds of information. The registry specifies ADAS-Cog change from baseline to Week 48, describes the scale, and identifies the primary analysis population. Those details clarify what the study intended to measure. Without numerical estimates and uncertainty, however, the central treatment comparison cannot be quantified or interpreted statistically.
The record also highlights multiplicity in trials that compare more than one dose and consider genetic subgroups. It states that a hierarchical procedure was used for the primary outcome. Such procedures can organize interpretation across related tests, but the hierarchy and its results need to be available to evaluate what was concluded. Those details are not reported in the registry record.
Several kinds of outcomes appear in the record, including cognition, function, neuropsychiatric symptoms, quality of life, resource use, serious adverse events, and deaths. These measures represent distinct domains. A statistical reader should not treat evidence from one domain as a substitute for another, or assume that a count establishes a causal relationship.
The study is also a reminder that analysis populations and missing observations affect interpretation. The primary population excludes randomized participants who did not meet the stated dose and post-baseline assessment criteria, and the record says only participants available at each time point were analyzed. Without visit-specific counts and missing-data methods, the possible influence of unavailable observations cannot be assessed.
Finally, planning calculations and trial findings answer different questions. The stated planning assumptions describe illustrations based on an observed effect, not observed power. The numerical planning results and the original sample-size justification are not reported in the registry record. Keeping those distinctions clear is essential when reading clinical-trial statistics.
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 largest limitation for statistical interpretation is that numerical results for the primary endpoint are not reported in the registry record. Without group estimates, treatment contrasts, and uncertainty measures, the direction, magnitude, and precision of the primary comparison cannot be assessed. Numerical results for the listed secondary outcomes are also not reported in the registry record.
The endpoint description gives the analysis population and says only participants available at specified time points were analyzed. It does not report how many participants contributed at each visit, reasons for missing assessments, or methods for handling missing data. No sensitivity-analysis details are reported in the registry record. This limits assessment of whether estimates might depend on assumptions about incomplete follow-up.
The record states that hierarchical testing was used for the two dose groups and genetic subgroups, but does not describe the sequence or decision rules. It also does not state how multiplicity was addressed across the wider set of outcomes and analyses. The analysis model and its covariates are not reported, despite the primary result being described as a least-squares mean.
The safety record gives event counts and at-risk counts but does not provide exposure-adjusted rates, statistical comparisons, or attribution details. The listing of leading serious adverse-event terms does not specify their distribution by treatment arm. Those omissions restrict causal and comparative interpretation of safety information.
The planned sample-size justification is not reported in the registry record. The stated planning assumptions describe illustrations, but their numerical results are not reported in the registry record. These are trial-specific reporting limitations; they do not establish whether additional information exists outside this record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Rosiglitazone (Extended Release Tablets) As Adjunctive Therapy For Subjects With Mild To Moderate Alzheimer's Disease — NCT00348309.
- Registry record last updated 2017-11-28; status last verified 2017-09.
- Results first posted 2017-11-28.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT00348309 was a completed randomized, double-blind, placebo-controlled, parallel-group study of rosiglitazone extended release as adjunctive therapy to donepezil in mild to moderate Alzheimer’s disease. The registry lists 1496 enrolled participants, three arms—2mg rosiglitazone, 8mg rosiglitazone, and placebo—and quadruple masking. The official title describes a 54-week study with APOE ε4 stratification.
The primary endpoint was change in ADAS-Cog total score from baseline at Week 0 to Week 48. The record describes least-squares means, standard errors, and a hierarchical testing procedure, but numerical primary-outcome results are not reported in the registry record. It also lists secondary outcomes and adverse-event counts by arm.
Details of the missing-data methods, full testing hierarchy, model specification, and original sample-size justification are not reported in the registry record. The posted information supports a description of the design and endpoint, but not a quantitative conclusion about the primary treatment comparison.