Data Analysis for Drug Repurposing for Effective Alzheimer's Medicines (DREAM)- Propranolol/Carvedilol Versus Atenolol/Bisoprolol/Sotalol: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Data Analysis for Drug Repurposing for Effective Alzheimer's Medicines (DREAM)- Propranolol/Carvedilol Versus Atenolol/Bisoprolol/Sotalol (NCT05794997), a registered phase not stated study. The primary outcome is Time to Dementia Onset.

2. Clinical Question

The clinical question is comparative rather than a test of a treatment against no treatment: both groups received medications for the same indication, hypertension. The stated aim is to evaluate comparative risk of dementia or Alzheimer's disease onset. The endpoint description is broader than Alzheimer's disease alone, however, and includes several dementia categories.

The registry names the study cohort after 1:1 propensity score matching as the population for the primary outcome. Matching is a design approach used in nonrandomized data to make groups more similar on measured characteristics. It does not, by itself, establish that the groups are comparable on characteristics that were not measured or included in matching. The registry record does not identify the matched characteristics or provide a matched cohort size.

No outcome estimates or analysis results are registered. Therefore, the comparative question is defined, but the posted record does not show whether the observed rates differed, how large any difference may have been, or how precisely the comparison was estimated.

3. Trial Design

The registry describes a completed study sponsored by Brigham and Women's Hospital. Its title identifies a comparison of propranolol or carvedilol with atenolol, bisoprolol or sotalol, and the condition listed is hypertension. The study aim is to evaluate comparative dementia or Alzheimer's disease onset among patients treated with medications targeting specific metabolic pathways versus alternative medications for the same indication.

The arms are labelled exposure group and reference group. The record explicitly classifies the comparison as not randomized, states that allocation is not stated, and cautions that the arms are not a like-for-like control comparison. These details matter because a nonrandomized medication comparison can reflect differences between patients as well as differences in their medication exposures. Propensity score matching is named for the primary study cohort, but the registry does not describe its implementation, matching variables, or balance assessment.

The registry lists enrollment of 817337. That figure does not establish the size of the matched population used for the endpoint: the endpoint population is described as the cohort after 1:1 propensity score matching, and its size is not reported in the registry record. The registry lists a start date of 2022-11-30, a primary completion date of 2023-01-01, and a completed status.

The listed primary outcome is time to dementia onset, expressed as an incidence rate per 1000 person year, with a 95% Confidence Interval as the dispersion measure. The record gives median follow-up times for exposure and reference groups in four numbered pairs: 1) 298 days (exp), 249 days (ref) 2) 995 days (exp), 981 days (ref) 3) 515 days (exp), 448 days (ref) 4) 307 days (exp), 253 days (ref). It does not explain what distinguishes these pairs. The registry lists one outcome measure but no analyses.

The record does not report the statistical analysis plan, sample-size justification, interim analyses, or details of the data source. Those design features cannot be inferred from the enrollment figure or the completed status.

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: Study cohort after 1:1 propensity score matching

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.

Population: Study cohort after 1:1 propensity score matching

6. Primary Endpoint

The registry lists the primary outcome as Time to Dementia Onset, measured over Median follow up times: 1) 298 days (exp), 249 days (ref) 2) 995 days (exp), 981 days (ref) 3) 515 days (exp), 448 days (ref) 4) 307 days (exp), 253 days (ref).

The primary endpoint is titled “Time to Dementia Onset.” The registry describes the event as Alzheimer's disease, vascular dementia, senile, presenile or unspecified dementia, or dementia in other diseases classified elsewhere. It states that the full definition is in an uploaded protocol because of size limitations; the full operational definition is not reported in the registry record.

The outcome is expressed as an incidence rate per 1000 person year, with a 95% Confidence Interval. In general, an incidence rate relates the number of new events to the amount of time people are observed and at risk. This can be useful when participants have different follow-up durations: it accounts for the time contributed by each person rather than treating every participant as if observed for an identical period. The registry's stated median follow-up times also indicate that follow-up duration is part of the outcome context.

In general, time-to-event outcomes address both whether an event occurs during observation and when it occurs. An incidence rate is a rate measure, however, and does not by itself convey the full distribution of time until an event or a person's absolute probability of developing dementia. Interpretation also depends on how events and person-time were defined, how censoring was handled, and which participants entered the analysis. Those details are not reported in the registry record.

The endpoint may capture a broader set of dementia diagnoses than Alzheimer's disease alone. That breadth is stated in the outcome description, but the registry does not provide separate results by diagnosis category. No incidence-rate estimate or confidence interval is posted, so the endpoint's observed magnitude and precision cannot be assessed from this record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does an incidence rate describe?

In general, an incidence rate relates event counts to accumulated time at risk. A person contributes time while under observation and eligible to experience the event under the study's definitions. Participants who have different observation periods can therefore contribute different amounts of person-time. A rate per 1000 person year expresses the event frequency relative to that exposure-time scale; it is not itself a probability that an individual will develop dementia.

For this study, the registry names time to dementia onset as the primary outcome and specifies an incidence rate per 1000 person year. It does not post the event counts or person-time totals used to obtain a rate. Nor does it describe the precise rules for starting and ending person-time, treatment exposure assignment, censoring, or competing events. Those operational details are not reported in the registry record, and they can affect the interpretation of a rate.

What does a confidence interval contribute?

In general, a confidence interval gives a range of values compatible with an estimate under the statistical model and its assumptions. It communicates sampling uncertainty and should be read together with the point estimate. The width reflects precision: greater uncertainty generally corresponds to a wider interval. A confidence interval does not account automatically for confounding, measurement error, or other systematic sources of bias.

The registry specifies a 95% Confidence Interval as the endpoint's dispersion measure. It does not post a point estimate or interval, and it does not identify the interval method. As a result, the stated interval type alone does not permit an assessment of the comparison's precision or statistical evidence.

Why might rate data require care?

In general, rate analyses often use models that account for event counts and exposure time. A simple count model may assume that variability is consistent with its mean. Over-dispersion occurs when the observed variability exceeds that assumption; ignoring it can make uncertainty appear smaller than it is. Methods that accommodate over-dispersion can adjust uncertainty estimates, but the appropriate approach depends on the data structure and model.

This is general statistical context, not a description of the analysis used in this study. The registry does not report a model, an over-dispersion assessment, a link function, covariate adjustment, or a method for comparing the groups. No particular count model or adjustment should be inferred from the incidence-rate outcome label.

What does propensity score matching do?

In general, a propensity score summarizes the probability of receiving one exposure rather than another conditional on measured characteristics. Matching people with similar scores is intended to create groups with improved balance on the characteristics used to estimate the score. The approach can reduce confounding from measured variables when it is appropriately designed and assessed.

Matching is not random assignment. It cannot balance variables that were not measured or included in the score, and it may exclude people who cannot be matched. The target population after matching may differ from the initial study population. For the registry's primary outcome, the population is described as the study cohort after 1:1 propensity score matching. The record does not identify the score model, matching algorithm, caliper, covariates, balance diagnostics, or number of participants retained. Thus, the matching label does not by itself establish the degree of comparability between the exposure and reference groups.

How should the follow-up information be read?

The registry reports median follow-up times for exposure and reference groups as 1) 298 days (exp), 249 days (ref) 2) 995 days (exp), 981 days (ref) 3) 515 days (exp), 448 days (ref) 4) 307 days (exp), 253 days (ref). A median describes the middle of a distribution, not the full range or the total person-time. It does not show whether follow-up distributions are otherwise similar. The record does not explain the relationship between the numbered pairs and the primary endpoint analysis.

In general, unequal follow-up does not automatically invalidate a rate comparison, because rate measures account for time at risk. But valid interpretation still depends on consistent event definitions, ascertainment, censoring, and analysis methods. The registry does not report these details. The follow-up medians alone cannot show whether differential observation affected the comparison.

What do the planning assumptions mean?

The registry describes sample-size assumptions as a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. It also distinguishes time-to-event calculations, which give total events, from other calculations, which give participants per arm.

In general, a planning calculation asks how much information would be needed to detect a specified effect under chosen assumptions. Such a calculation is not a measurement of how likely the completed study was to find a result, and it is not observed power. The registry does not report the observed effect or the numerical planning results in its analysis record. The assumptions also do not establish that the original study was designed using those calculations.

Overall, the statistical labels identify the intended outcome scale, interval type, matching population, and planning framework. The registry does not provide the analysis details or outcome estimates needed to reproduce or evaluate the primary comparison.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analyses. The single listed primary outcome is time to dementia onset, with the dementia categories described in the endpoint definition. The record does not report whether additional outcomes, subgroup analyses, sensitivity analyses, or repeated comparisons were conducted outside the registered analysis entries. A multiplicity strategy, such as an adjustment across several statistical tests, is not reported in the registry record.

In general, multiplicity matters when many hypotheses are tested: the chance of at least one apparently unusual result can increase as the number of tests grows. Prespecifying a primary outcome and analysis can help clarify which result is confirmatory, but the registry's outcome listing does not provide a complete statistical analysis plan. No interim analyses or stopping rules are reported.

Missing-data methods are also not described. The registry does not report the extent of missing exposure, covariate, follow-up, or outcome information, nor does it state whether records were excluded, imputed, or handled using another approach. It is therefore not possible to assess whether missingness could have affected group comparability or the rate estimates.

The analysis population is described as a study cohort after 1:1 propensity score matching. The record does not specify the number retained after matching, how unmatched participants were treated, or whether participants with incomplete covariates could be matched. The enrollment figure of 817337 should not be treated as the matched endpoint population size.

These omissions do not show that a multiplicity or missing-data procedure was absent; they mean that such procedures are not reported in the registry record. Without the analysis methods and results, readers cannot determine how many comparisons informed the conclusions or how incomplete data were addressed.

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

The registry record defines a comparison between an exposure group receiving propranolol or carvedilol and a reference group receiving atenolol, bisoprolol or sotalol. It identifies a matched study cohort and a primary endpoint concerning dementia onset. Those elements describe the intended question and broad outcome framework.

The record does not provide a primary outcome estimate, an event count, a group-specific incidence rate, a between-group comparison, or a confidence interval. It therefore does not show whether dementia-onset rates were higher, lower, or similar between the groups, and it does not show how precisely any comparison was estimated. A completed status and a large registry enrollment do not fill this gap.

Even if an association estimate were available, a nonrandomized comparison would not by itself establish that medication exposure caused a difference in dementia onset. In general, treatment selection can be related to patient characteristics that also relate to outcomes. Propensity score matching can address measured factors included in its design, but it does not eliminate all possible confounding. The record's caution that the groups are not a like-for-like control comparison reinforces the need for restraint in causal interpretation.

The endpoint combines Alzheimer's disease with other listed dementia categories. It should not be interpreted as an Alzheimer's-only outcome. The record points to an uploaded protocol for the full endpoint definition, but the complete operational definition is not reported in the registry record. The listed follow-up medians also lack an explanation of how their numbered pairs correspond to analyses.

The adverse-event entries report zero affected participants and zero deaths in the displayed categories. These entries do not establish the absence of all harms or provide a comprehensive assessment of safety. They are distinct from the missing primary dementia outcome results.

In short, the registry identifies a nonrandomized comparative research question and a rate-based endpoint, but it does not provide the results required to draw a conclusion about the observed comparison.

12. Why This Trial Matters Statistically

This study is statistically relevant because it frames dementia onset as an outcome of interest in a comparison between medication groups used for the same indication. The registry's stated aim is comparative, and its primary endpoint uses an incidence rate per 1000 person year. This scale can account for different observed time at risk, which is important in studies where participants may have different follow-up durations.

The record also illustrates why design labels need careful interpretation. A 1:1 propensity score-matched cohort can make a nonrandomized comparison more structured, but matching does not have the same role as random assignment. The registry explicitly notes that the comparison is not randomized and that the arms are not a like-for-like control comparison. Statistical adjustment and matching should therefore be considered alongside possible residual confounding, the choice of comparator, and the way exposure and outcomes were defined.

Follow-up is another important part of the analysis. The registry lists several median follow-up pairs, but does not explain the groupings or connect them to endpoint estimates. A median cannot substitute for the event counts, person-time, or full follow-up distribution needed to understand a rate comparison.

Equally important is the distinction between a defined analysis and a reported result. The registry identifies one outcome measure and specifies a confidence-interval type, but lists no analyses. Without estimates and uncertainty intervals, neither the magnitude nor precision of the comparison can be evaluated. The sample-size assumptions described in the record are planning concepts, not evidence of an observed effect or proof that the original study had adequate information.

For biostatistical interpretation, the record emphasizes the value of transparent reporting: the matched population, analysis methods, outcome definition, follow-up rules, missing-data approach, and numerical results are all necessary to assess a comparative rate study. Their absence limits what can be concluded from the registry entry.

13. 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 principal limitation for interpreting this study record is that no primary outcome analysis is posted. The registry lists one outcome measure but zero analyses, and it gives no group-specific dementia incidence rates, event counts, comparative estimate, or confidence interval. The direction, magnitude, and precision of the outcome comparison therefore remain unknown.

The registry lists enrollment of 817337, but does not report the size of the matched population used for the primary outcome. The outcome population is described as a cohort after 1:1 propensity score matching, without the matching variables, procedure, balance diagnostics, or number of participants retained. This prevents assessment of how well the exposure and reference groups were made comparable.

The record states that the comparison is not randomized, allocation is not stated, and the arms are not a like-for-like control comparison. These features limit causal interpretation. The available record does not report the data source, medication exposure definitions, treatment duration, dose, covariate adjustment, or approach to confounding.

The endpoint description covers multiple dementia categories and refers readers to an uploaded protocol for the full definition. The complete operational definition is not reported in the registry record. The follow-up information lists several exposure and reference median durations but does not explain their corresponding groups or analyses. The record also does not describe person-time construction, censoring, competing events, or outcome ascertainment.

Finally, missing-data procedures, multiplicity handling, interim analyses, and the statistical analysis plan are not reported. The displayed adverse-event entries are limited to the listed categories and cannot establish comprehensive safety reporting. These are limitations of what the registry record documents; they do not establish what was or was not done outside that record.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Data Analysis for Drug Repurposing for Effective Alzheimer's Medicines (DREAM)- Propranolol/Carvedilol Versus Atenolol/Bisoprolol/Sotalol — NCT05794997.
• Registry record last updated 2025-11-10; status last verified 2025-10.
• Results first posted 2025-11-10.
• 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.

17. Record Summary

The registry identifies a completed, nonrandomized comparison sponsored by Brigham and Women's Hospital, with hypertension listed as the condition. The exposure group received propranolol or carvedilol; the reference group received atenolol, bisoprolol or sotalol. Enrollment is listed as 817337, and the primary population is described as a study cohort after 1:1 propensity score matching.

The primary outcome is time to dementia onset, expressed as an incidence rate per 1000 person year with a 95% Confidence Interval. The endpoint description includes Alzheimer's disease and several other dementia categories. The record lists median follow-up durations for exposure and reference groups, but does not explain the numbered pairings.

The registry reports one outcome measure and zero analyses. It does not post numerical primary outcome results or describe key analysis details such as the matching procedure, missing-data methods, or multiplicity strategy. The record therefore establishes the intended comparison and endpoint, but not the study's observed dementia-rate findings.