Carolin Herrmann
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Carolin Herrmann has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Geraldine Rauch
Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Institute of Biometry and Clinical Epidemiology, Berlin, Germany
11 shared publications - Maximilian Pilz
Fraunhofer Institute for Industrial Mathematics, Kaiserslautern, Germany
8 shared publications - Meinhard Kieser
Institute of Medical Biometry University of Heidelberg Heidelberg Germany
8 shared publications - Björn Bokelmann
Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Institute of Biometry and Clinical Epidemiology, Berlin, Germany
3 shared publications Geraldine Rauch
Institute of Biometry and Clinical Epidemiology Charité – Universitätsmedizin Berlin Berlin Germany
3 shared publications- Jan Meis
Institute of Medical Biometry, University Hospital of Heidelberg, University of Heidelberg, Heidelberg, Germany
3 shared publications
Latest Publications
- Open Access
- 18 March 2026
- Published by Wiley in Statistics in Medicine
- Vol. 45 (6-7) , e70490
Sample Size Recalculation in Adaptive Group Sequential Study Designs for Comparing Restricted Mean Survival Times
Non‐proportional hazards cases are frequently expected in clinical trials with time‐to‐event endpoints (e.g., cardiology, oncology). The relevance of hazard ratios to quantify the treatment effect is questionable and potentially misleading in this context. Hence, alternative methods comparing restricted mean survival times are increasingly promoted. Specific challenges arise when planning clinical trials for comparing restricted mean survival times, as several nuisance parameter estimates are needed for calculating the sample size. Precise estimates might be difficult to obtain at the planning stage and might lead to underpowered trials. One way of dealing with this insecurity is to apply adaptive group sequential study designs with the option to adapt the sample size during an ongoing trial. Within this work, we consider such sample size adaptations, with a specific focus on the context of delayed treatment effects. We compare the performance of an adaptive design with the restricted mean survival time as the primary endpoint with other commonly chosen endpoints in this scenario by means of an extensive simulation study. With our proposed method, adaptive designs with the restricted mean survival time as the primary endpoint are now thoroughly explained. The combination test that we describe can also be useful for other adaptations than sample sizes.
- 18 March 2026
- Published by Wiley in Statistics in Medicine
- Vol. 45 (6-7) , e70490
Open AccessSample Size Recalculation in Adaptive Group Sequential Study Designs for Comparing Restricted Mean Survival Times
Non‐proportional hazards cases are frequently expected in clinical trials with time‐to‐event endpoints (e.g., cardiology, oncology). The relevance of hazard ratios to quantify the treatment effect is questionable and potentially misleading in this context. Hence, alternative methods comparing restricted mean survival times are increasingly promoted. Specific challenges arise when planning clinical trials for comparing restricted mean survival times, as several nuisance parameter estimates are needed for calculating the sample size. Precise estimates might be difficult to obtain at the planning stage and might lead to underpowered trials. One way of dealing with this insecurity is to apply adaptive group sequential study designs with the option to adapt the sample size during an ongoing trial. Within this work, we consider such sample size adaptations, with a specific focus on the context of delayed treatment effects. We compare the performance of an adaptive design with the restricted mean survival time as the primary endpoint with other commonly chosen endpoints in this scenario by means of an extensive simulation study. With our proposed method, adaptive designs with the restricted mean survival time as the primary endpoint are now thoroughly explained. The combination test that we describe can also be useful for other adaptations than sample sizes.
- Open Access
- 25 September 2024
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 24 (1) , 1-15
Sample size recalculation in three-stage clinical trials and its evaluation
Background: In clinical trials, the determination of an adequate sample size is a challenging task, mainly due to the uncertainty about the value of the effect size and nuisance parameters. One method to deal with this uncertainty is a sample size recalculation. Thereby, an interim analysis is performed based on which the sample size for the remaining trial is adapted. With few exceptions, previous literature has only examined the potential of recalculation in two-stage trials. Methods: In our research, we address sample size recalculation in three-stage trials, i.e. trials with two pre-planned interim analyses. We show how recalculation rules from two-stage trials can be modified to be applicable to three-stage trials. We also illustrate how a performance measure, recently suggested for two-stage trial recalculation (the conditional performance score) can be applied to evaluate recalculation rules in three-stage trials, and we describe performance evaluation in those trials from the global point of view. To assess the potential of recalculation in three-stage trials, we compare, in a simulation study, two-stage group sequential designs with three-stage group sequential designs as well as multiple three-stage designs with recalculation. Results: While we observe a notable favorable effect in terms of power and expected sample size by using three-stage designs compared to two-stage designs, the benefits of recalculation rules appear less clear and are dependent on the performance measures applied. Conclusions: Sample size recalculation is also applicable in three-stage designs. However, the extent to which recalculation brings benefits depends on which trial characteristics are most important to the applicants.
- 25 September 2024
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 24 (1) , 1-15
Open AccessSample size recalculation in three-stage clinical trials and its evaluation
Background: In clinical trials, the determination of an adequate sample size is a challenging task, mainly due to the uncertainty about the value of the effect size and nuisance parameters. One method to deal with this uncertainty is a sample size recalculation. Thereby, an interim analysis is performed based on which the sample size for the remaining trial is adapted. With few exceptions, previous literature has only examined the potential of recalculation in two-stage trials. Methods: In our research, we address sample size recalculation in three-stage trials, i.e. trials with two pre-planned interim analyses. We show how recalculation rules from two-stage trials can be modified to be applicable to three-stage trials. We also illustrate how a performance measure, recently suggested for two-stage trial recalculation (the conditional performance score) can be applied to evaluate recalculation rules in three-stage trials, and we describe performance evaluation in those trials from the global point of view. To assess the potential of recalculation in three-stage trials, we compare, in a simulation study, two-stage group sequential designs with three-stage group sequential designs as well as multiple three-stage designs with recalculation. Results: While we observe a notable favorable effect in terms of power and expected sample size by using three-stage designs compared to two-stage designs, the benefits of recalculation rules appear less clear and are dependent on the performance measures applied. Conclusions: Sample size recalculation is also applicable in three-stage designs. However, the extent to which recalculation brings benefits depends on which trial characteristics are most important to the applicants.
- Open Access
- 12 August 2024
- Published by Wiley in Biometrical Journal
- Vol. 66 (6) , e202300271
Sample Size Calculation Under Nonproportional Hazards Using Average Hazard Ratios
Many clinical trials assess time‐to‐event endpoints. To describe the difference between groups in terms of time to event, we often employ hazard ratios. However, the hazard ratio is only informative in the case of proportional hazards (PHs) over time. There exist many other effect measures that do not require PHs. One of them is the average hazard ratio (AHR). Its core idea is to utilize a time‐dependent weighting function that accounts for time variation. Though propagated in methodological research papers, the AHR is rarely used in practice. To facilitate its application, we unfold approaches for sample size calculation of an AHR test. We assess the reliability of the sample size calculation by extensive simulation studies covering various survival and censoring distributions with proportional as well as nonproportional hazards (N‐PHs). The findings suggest that a simulation‐based sample size calculation approach can be useful for designing clinical trials with N‐PHs. Using the AHR can result in increased statistical power to detect differences between groups with more efficient sample sizes.
- 12 August 2024
- Published by Wiley in Biometrical Journal
- Vol. 66 (6) , e202300271
Open AccessSample Size Calculation Under Nonproportional Hazards Using Average Hazard Ratios
Many clinical trials assess time‐to‐event endpoints. To describe the difference between groups in terms of time to event, we often employ hazard ratios. However, the hazard ratio is only informative in the case of proportional hazards (PHs) over time. There exist many other effect measures that do not require PHs. One of them is the average hazard ratio (AHR). Its core idea is to utilize a time‐dependent weighting function that accounts for time variation. Though propagated in methodological research papers, the AHR is rarely used in practice. To facilitate its application, we unfold approaches for sample size calculation of an AHR test. We assess the reliability of the sample size calculation by extensive simulation studies covering various survival and censoring distributions with proportional as well as nonproportional hazards (N‐PHs). The findings suggest that a simulation‐based sample size calculation approach can be useful for designing clinical trials with N‐PHs. Using the AHR can result in increased statistical power to detect differences between groups with more efficient sample sizes.
- Top Cited PapersOpen Access
- 10 June 2024
- Published by Springer Nature in Sports Medicine - Open
- Vol. 10 (1) , 1-16
Rating of Perceived Exertion: A Large Cross-Sectional Study Defining Intensity Levels for Individual Physical Activity Recommendations
Background: Physical inactivity is a growing risk factor worldwide, therefore getting people into sports is necessary. When prescribing physical activity, it is essential to recommend the correct training intensities. Cardiopulmonary exercise testing (CPX) enables precise determination of individuals’ training intensities but is unavailable for a broad population. Therefore, the Borg scale allows individuals to assess perceived exertion and set their intensity easily and cost-efficiently. In order to transfer CPX to rating of perceived exertion (RPE), previous studies investigated RPE on specific physiological anchors, e.g. blood lactate (bLa) concentrations, but representativeness for a broad population is questionable. Some contradictory findings regarding individual factors influencing RPE occur, whereas univariable analysis has been performed so far. Moreover, a multivariable understanding of individual factors influencing RPE is missing. This study aims to determine RPE values at the individual anaerobic threshold (LT2) and defined bLa concentrations in a large cohort and to evaluate individual factors influencing RPE with multivariable analysis.Methods: CPX with bicycle or treadmill ergometer of 6311 participants were analyzed in this cross-sectional study. RPE values at bLa concentrations 2 mmol/l, 3 mmol/l, 4 mmol/l, and LT2 (first rise in bLa over baseline + 1.5 mmol/l) were estimated by spline interpolation. Multivariable cumulative ordinal regression models were performed to assess the influence of sex, age, type of ergometry, VO2max, and duration of exercise testing on RPE.Results: Median values [interquartile range (IQR)] of the total population were RPE 13 [11; 14] at 2 mmol/l, RPE 15 [13; 16] at 3 mmol/l, RPE 16 [15; 17] at 4 mmol/l, and RPE 15 [14; 16] at LT2. Main influence of individual factors on RPE were seen especially at 2 mmol/l: male sex (odds ratio (OR) [95%-CI]: 0.65 [0.587; 0.719]), treadmill ergometry (OR 0.754 [0.641; 0.886]), number of stages (OR 1.345 [1.300; 1.394]), age (OR 1.015 [1.012; 1.018]), and VO2max (OR 1.023 [1.015; 1.030]). Number of stages was the only identified influencing factor on RPE at all lactate concentrations/LT2 (3 mmol/l: OR 1.290 [1.244; 1.336]; 4 mmol/l: OR 1.229 [1.187; 1.274]; LT2: OR 1.155 [1.115; 1.197]).Conclusion: Our results suggest RPE ≤ 11 for light intensity, RPE 12–14 for moderate intensity, and RPE 15–17 for vigorous intensity, which slightly differs from the current American College of Sports Medicine (ACSM) recommendations. Additionally, we propose an RPE of 15 delineating heavy and severe intensity domain. Age, sex, type of ergometry, duration of exercise, and cardiopulmonary fitness should be considered when recommending individualized intensities with RPE, primarily at lower intensities. Therefore, this study can be used as a new guideline for prescribing individual RPE values in the clinical practice, predominantly for endurance type exercise.
- 10 June 2024
- Published by Springer Nature in Sports Medicine - Open
- Vol. 10 (1) , 1-16
Top Cited PapersOpen AccessRating of Perceived Exertion: A Large Cross-Sectional Study Defining Intensity Levels for Individual Physical Activity Recommendations
Background: Physical inactivity is a growing risk factor worldwide, therefore getting people into sports is necessary. When prescribing physical activity, it is essential to recommend the correct training intensities. Cardiopulmonary exercise testing (CPX) enables precise determination of individuals’ training intensities but is unavailable for a broad population. Therefore, the Borg scale allows individuals to assess perceived exertion and set their intensity easily and cost-efficiently. In order to transfer CPX to rating of perceived exertion (RPE), previous studies investigated RPE on specific physiological anchors, e.g. blood lactate (bLa) concentrations, but representativeness for a broad population is questionable. Some contradictory findings regarding individual factors influencing RPE occur, whereas univariable analysis has been performed so far. Moreover, a multivariable understanding of individual factors influencing RPE is missing. This study aims to determine RPE values at the individual anaerobic threshold (LT2) and defined bLa concentrations in a large cohort and to evaluate individual factors influencing RPE with multivariable analysis.Methods: CPX with bicycle or treadmill ergometer of 6311 participants were analyzed in this cross-sectional study. RPE values at bLa concentrations 2 mmol/l, 3 mmol/l, 4 mmol/l, and LT2 (first rise in bLa over baseline + 1.5 mmol/l) were estimated by spline interpolation. Multivariable cumulative ordinal regression models were performed to assess the influence of sex, age, type of ergometry, VO2max, and duration of exercise testing on RPE.Results: Median values [interquartile range (IQR)] of the total population were RPE 13 [11; 14] at 2 mmol/l, RPE 15 [13; 16] at 3 mmol/l, RPE 16 [15; 17] at 4 mmol/l, and RPE 15 [14; 16] at LT2. Main influence of individual factors on RPE were seen especially at 2 mmol/l: male sex (odds ratio (OR) [95%-CI]: 0.65 [0.587; 0.719]), treadmill ergometry (OR 0.754 [0.641; 0.886]), number of stages (OR 1.345 [1.300; 1.394]), age (OR 1.015 [1.012; 1.018]), and VO2max (OR 1.023 [1.015; 1.030]). Number of stages was the only identified influencing factor on RPE at all lactate concentrations/LT2 (3 mmol/l: OR 1.290 [1.244; 1.336]; 4 mmol/l: OR 1.229 [1.187; 1.274]; LT2: OR 1.155 [1.115; 1.197]).Conclusion: Our results suggest RPE ≤ 11 for light intensity, RPE 12–14 for moderate intensity, and RPE 15–17 for vigorous intensity, which slightly differs from the current American College of Sports Medicine (ACSM) recommendations. Additionally, we propose an RPE of 15 delineating heavy and severe intensity domain. Age, sex, type of ergometry, duration of exercise, and cardiopulmonary fitness should be considered when recommending individualized intensities with RPE, primarily at lower intensities. Therefore, this study can be used as a new guideline for prescribing individual RPE values in the clinical practice, predominantly for endurance type exercise.
- Open Access
- 2 May 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-10
Neurological affection and serum neurofilament light chain in wild type transthyretin amyloidosis
In contrast to inherited transthyretin amyloidosis (A-ATTRv), neuropathy is not a classic leading symptom of wild type transthyretin amyloidosis (A-ATTRwt). However, neurological symptoms are increasingly relevant in A-ATTRwt as well. To better understand the role of neurological symptoms in A-ATTRwt, A-ATTRwt patients were prospectively characterized at Amyloidosis Center Charité Berlin (ACCB) between 2018 and 2023 using detailed neurological examination, quality of life questionnaires, and analysis of age- and BMI-adapted serum neurofilament light chain (NFL) levels. 16 out of 73 (21.9%) patients presented with a severe neuropathy which we defined by a Neuropathy Impairment Score (NIS) of 20 or more. In this group, quality of life was reduced, peripheral neuropathy was more severe, and spinal stenosis and joint replacements were frequent. Age- and BMI matched serum NFL levels were markedly elevated in patients with a NIS ≥ 20. We therefore conclude that highly abnormal values in neuropathy scores such as the NIS occur in A-ATTRwt, and have an important impact on quality of life. Both peripheral neuropathy and spinal canal stenosis are likely contributors. Serum NFL may serve as a biomarker for neurological affection in patients with A-ATTRwt. It will be important to consider neurological aspects of A-ATTRwt for diagnosis, clinical follow-up, and future treatment development.
- 2 May 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-10
Open AccessNeurological affection and serum neurofilament light chain in wild type transthyretin amyloidosis
In contrast to inherited transthyretin amyloidosis (A-ATTRv), neuropathy is not a classic leading symptom of wild type transthyretin amyloidosis (A-ATTRwt). However, neurological symptoms are increasingly relevant in A-ATTRwt as well. To better understand the role of neurological symptoms in A-ATTRwt, A-ATTRwt patients were prospectively characterized at Amyloidosis Center Charité Berlin (ACCB) between 2018 and 2023 using detailed neurological examination, quality of life questionnaires, and analysis of age- and BMI-adapted serum neurofilament light chain (NFL) levels. 16 out of 73 (21.9%) patients presented with a severe neuropathy which we defined by a Neuropathy Impairment Score (NIS) of 20 or more. In this group, quality of life was reduced, peripheral neuropathy was more severe, and spinal stenosis and joint replacements were frequent. Age- and BMI matched serum NFL levels were markedly elevated in patients with a NIS ≥ 20. We therefore conclude that highly abnormal values in neuropathy scores such as the NIS occur in A-ATTRwt, and have an important impact on quality of life. Both peripheral neuropathy and spinal canal stenosis are likely contributors. Serum NFL may serve as a biomarker for neurological affection in patients with A-ATTRwt. It will be important to consider neurological aspects of A-ATTRwt for diagnosis, clinical follow-up, and future treatment development.
- Open Access
- 10 February 2024
- Published by Wiley in Statistics in Medicine
- Vol. 43 (8) , 1577-1603
Point estimation, confidence intervals, and P‐values for optimal adaptive two‐stage designs with normal endpoints
Due to the dependency structure in the sampling process, adaptive trial designs create challenges in point and interval estimation and in the calculation of P‐values. Optimal adaptive designs, which are designs where the parameters governing the adaptivity are chosen to maximize some performance criterion, suffer from the same problem. Various analysis methods which are able to handle this dependency structure have already been developed. In this work, we aim to give a comprehensive summary of these methods and show how they can be applied to the class of designs with planned adaptivity, of which optimal adaptive designs are an important member. The defining feature of these kinds of designs is that the adaptive elements are completely prespecified. This allows for explicit descriptions of the calculations involved, which makes it possible to evaluate different methods in a fast and accurate manner. We will explain how to do so, and present an extensive comparison of the performance characteristics of various estimators between an optimal adaptive design and its group‐sequential counterpart.
- 10 February 2024
- Published by Wiley in Statistics in Medicine
- Vol. 43 (8) , 1577-1603
Open AccessPoint estimation, confidence intervals, and P‐values for optimal adaptive two‐stage designs with normal endpoints
Due to the dependency structure in the sampling process, adaptive trial designs create challenges in point and interval estimation and in the calculation of P‐values. Optimal adaptive designs, which are designs where the parameters governing the adaptivity are chosen to maximize some performance criterion, suffer from the same problem. Various analysis methods which are able to handle this dependency structure have already been developed. In this work, we aim to give a comprehensive summary of these methods and show how they can be applied to the class of designs with planned adaptivity, of which optimal adaptive designs are an important member. The defining feature of these kinds of designs is that the adaptive elements are completely prespecified. This allows for explicit descriptions of the calculations involved, which makes it possible to evaluate different methods in a fast and accurate manner. We will explain how to do so, and present an extensive comparison of the performance characteristics of various estimators between an optimal adaptive design and its group‐sequential counterpart.
- Open Access
- 28 December 2022
- Published by Wiley in Statistics in Medicine
- Vol. 42 (4) , 536-558
Optimization of the two‐stage group sequential three‐arm gold‐standard design for non‐inferiority trials
If design parameters are chosen appropriately, group sequential trial designs are known to be able to reduce the expected sample size under the alternative hypothesis compared to single‐stage designs. The same holds true for the so‐called ‘gold‐standard’ design for non‐inferiority trials, a design involving an experimental group, an active control group, and a placebo group. However, choosing design parameters that maximize the advantages of a two‐stage approach for the three‐arm gold‐standard design for non‐inferiority trials is not a straightforward task. In particular, optimal choices of futility boundaries for this design have not been thoroughly discussed in existing literature. We present a variation of the hierarchical testing procedure, which allows for the incorporation of binding futility boundaries at interim analyses. We show that this procedure maintains strong control of the family‐wise type I error rate. Within this framework, we consider the futility and efficacy boundaries as well as the sample size allocation ratios as optimization parameters. This allows the investigation of the efficiency gain from including the option to stop for futility in addition to the ability to stop for efficacy. To analyze the extended designs, optimality criteria that include the design's performance under the alternative as well as the null hypothesis are introduced. On top of this, we discuss methods to limit the allocation of placebo patients in the trial while maintaining relatively good operating characteristics. The results of our numerical optimization procedure are discussed and a comparison of different approaches to designing a three‐arm gold‐standard non‐inferiority trial is provided.
- 28 December 2022
- Published by Wiley in Statistics in Medicine
- Vol. 42 (4) , 536-558
Open AccessOptimization of the two‐stage group sequential three‐arm gold‐standard design for non‐inferiority trials
If design parameters are chosen appropriately, group sequential trial designs are known to be able to reduce the expected sample size under the alternative hypothesis compared to single‐stage designs. The same holds true for the so‐called ‘gold‐standard’ design for non‐inferiority trials, a design involving an experimental group, an active control group, and a placebo group. However, choosing design parameters that maximize the advantages of a two‐stage approach for the three‐arm gold‐standard design for non‐inferiority trials is not a straightforward task. In particular, optimal choices of futility boundaries for this design have not been thoroughly discussed in existing literature. We present a variation of the hierarchical testing procedure, which allows for the incorporation of binding futility boundaries at interim analyses. We show that this procedure maintains strong control of the family‐wise type I error rate. Within this framework, we consider the futility and efficacy boundaries as well as the sample size allocation ratios as optimization parameters. This allows the investigation of the efficiency gain from including the option to stop for futility in addition to the ability to stop for efficacy. To analyze the extended designs, optimality criteria that include the design's performance under the alternative as well as the null hypothesis are introduced. On top of this, we discuss methods to limit the allocation of placebo patients in the trial while maintaining relatively good operating characteristics. The results of our numerical optimization procedure are discussed and a comparison of different approaches to designing a three‐arm gold‐standard non‐inferiority trial is provided.
- Open Access
- 1 November 2022
- Published by Wiley in Pharmaceutical Statistics
- Vol. 21 (6) , 1121-1137
Optimal unplanned design modification in adaptive two-stage trials
Adaptive planning of clinical trials allows modifying the entire trial design at any time point mid-course. In this paper, we consider the case when a trial-external update of the planning assumptions during the ongoing trial makes an unforeseen design adaptation necessary. We take up the idea to construct adaptive designs with defined features by solving an optimization problem and apply it to the situation of unplanned design reassessment. By using the conditional error principle, we present an approach on how to optimally modify the trial design at an unplanned interim analysis while at the same time strictly protecting the type I error rate. This linking of optimal design planning and the conditional error principle allows sound reactions to unforeseen events that make a design reassessment necessary.
- 1 November 2022
- Published by Wiley in Pharmaceutical Statistics
- Vol. 21 (6) , 1121-1137
Open AccessOptimal unplanned design modification in adaptive two-stage trials
Adaptive planning of clinical trials allows modifying the entire trial design at any time point mid-course. In this paper, we consider the case when a trial-external update of the planning assumptions during the ongoing trial makes an unforeseen design adaptation necessary. We take up the idea to construct adaptive designs with defined features by solving an optimization problem and apply it to the situation of unplanned design reassessment. By using the conditional error principle, we present an approach on how to optimally modify the trial design at an unplanned interim analysis while at the same time strictly protecting the type I error rate. This linking of optimal design planning and the conditional error principle allows sound reactions to unforeseen events that make a design reassessment necessary.
- Open Access
- 15 April 2022
- Published by Wiley in Biometrical Journal
- Vol. 64 (6) , 989-1006
Optimization of adaptive designs with respect to a performance score
Adaptive designs are an increasingly popular method for the adaptation of design aspects in clinical trials, such as the sample size. Scoring different adaptive designs helps to make an appropriate choice among the numerous existing adaptive design methods. Several scores have been proposed to evaluate adaptive designs. Moreover, it is possible to determine optimal two-stage adaptive designs with respect to a customized objective score by solving a constrained optimization problem. In this paper, we use the conditional performance score by Herrmann et al. (2020) as the optimization criterion to derive optimal adaptive two-stage designs. We investigate variations of the original performance score, for example, by assigning different weights to the score components and by incorporating prior assumptions on the effect size. We further investigate a setting where the optimization framework is extended by a global power constraint, and additional optimization of the critical value function next to the stage-two sample size is performed. Those evaluations with respect to the sample size curves and the resulting design's performance can contribute to facilitate the score's usage in practice.
- 15 April 2022
- Published by Wiley in Biometrical Journal
- Vol. 64 (6) , 989-1006
Open AccessOptimization of adaptive designs with respect to a performance score
Adaptive designs are an increasingly popular method for the adaptation of design aspects in clinical trials, such as the sample size. Scoring different adaptive designs helps to make an appropriate choice among the numerous existing adaptive design methods. Several scores have been proposed to evaluate adaptive designs. Moreover, it is possible to determine optimal two-stage adaptive designs with respect to a customized objective score by solving a constrained optimization problem. In this paper, we use the conditional performance score by Herrmann et al. (2020) as the optimization criterion to derive optimal adaptive two-stage designs. We investigate variations of the original performance score, for example, by assigning different weights to the score components and by incorporating prior assumptions on the effect size. We further investigate a setting where the optimization framework is extended by a global power constraint, and additional optimization of the critical value function next to the stage-two sample size is performed. Those evaluations with respect to the sample size curves and the resulting design's performance can contribute to facilitate the score's usage in practice.
- 1 December 2021
- Published by Elsevier in Theoretical Population Biology
- Vol. 142, 46-56
Selection, recombination, and the ancestral initiation graph
- 1 December 2021
- Published by Elsevier in Theoretical Population Biology
- Vol. 142, 46-56
Selection, recombination, and the ancestral initiation graph