Jianling Bai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jianling Bai has a total of 185 co-authors. Here are the top co-authors by number of shared publications:
- Jinhui Liu
Department of Gynecology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China
17 shared publications Zihang Zhong
Department of Biostatistics, School of Public Health, Nanjing Medical University, Nanjing, Jiangsu 211166, China
13 shared publications- Hao Yu
Nanjing Medical University
12 shared publications Senmiao Ni
Department of Biostatistics, School of Public Health, Nanjing Medical University, Nanjing, China
12 shared publications- Rui Geng
Department of Biostatistics, School of Public Health, Nanjing Medical University, 101 Longmian Avenue, Jiangning District, Nanjing 211166, China
10 shared publications Yang Zhao
Department of Biostatistics, School of Public Health, Nanjing Medical University, Nanjing, Jiangsu 211166, China
9 shared publications
Latest Publications
- Open Access
- 29 April 2026
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 26 (1)
Choose-NI-metric: a shiny application for converting and comparing non-inferiority metrics in binary endpoint trials
In non-inferiority (NI) clinical trials with binary outcomes, treatment effects are commonly quantified using the risk difference (RD), relative risk (RR),
- 29 April 2026
- Published by Springer Nature in BMC Medical Research Methodology
- Vol. 26 (1)
Open AccessChoose-NI-metric: a shiny application for converting and comparing non-inferiority metrics in binary endpoint trials
In non-inferiority (NI) clinical trials with binary outcomes, treatment effects are commonly quantified using the risk difference (RD), relative risk (RR),
- Open Access
- 31 October 2025
- Published by Elsevier in Ecotoxicology and Environmental Safety
- Vol. 306, 119329
Dynamic associations between long-term exposure to ambient air pollution and respiratory-cardiovascular diseases: A trajectory analysis of a prospective study
- 31 October 2025
- Published by Elsevier in Ecotoxicology and Environmental Safety
- Vol. 306, 119329
Open AccessDynamic associations between long-term exposure to ambient air pollution and respiratory-cardiovascular diseases: A trajectory analysis of a prospective study
- 29 May 2025
- Published by SAGE Publications in Technology and Health Care
- Vol. 33 (5) , 2350-2370
Bayesian sequential decision-making for rare disease clinical trials
Background: Rare disease clinical trials face challenges due to limited sample sizes and ethical imperatives to minimize futile treatments. Bayesian sequential design dynamically optimizes decisions under uncertainty, offering efficiency gains over traditional fixed-sample approaches. Methods: Propose a framework integrating sequential Bayes factor and adaptive stopping rules for trials with binary endpoint. Bayesian posterior probabilities define early termination thresholds (superiority/futility), while Bayes Factor Design Analysis validates trial feasibility. Sequential Bayes factor updates iteratively guide interim decisions based on evidence strength. Results: The approach enables earlier trial termination (for superiority or futility), reducing sample size, time, and costs. Patients avoid unnecessary exposure to futility treatments, while results remain interpretable even if thresholds are unmet. Conclusion: The primary goal is to confirm treatment efficacy earlier, enabling trials to be stopped promptly for either superiority or futility treatments. This strategy reduces sample size, time, and financial costs, and prevents patient exposure to futile treatments. Moreover, the study aims to promote the adoption of Bayesian sequential decision-making, thereby accelerating rare disease clinical trial approvals and drug marketing.
- 29 May 2025
- Published by SAGE Publications in Technology and Health Care
- Vol. 33 (5) , 2350-2370
Bayesian sequential decision-making for rare disease clinical trials
Background: Rare disease clinical trials face challenges due to limited sample sizes and ethical imperatives to minimize futile treatments. Bayesian sequential design dynamically optimizes decisions under uncertainty, offering efficiency gains over traditional fixed-sample approaches. Methods: Propose a framework integrating sequential Bayes factor and adaptive stopping rules for trials with binary endpoint. Bayesian posterior probabilities define early termination thresholds (superiority/futility), while Bayes Factor Design Analysis validates trial feasibility. Sequential Bayes factor updates iteratively guide interim decisions based on evidence strength. Results: The approach enables earlier trial termination (for superiority or futility), reducing sample size, time, and costs. Patients avoid unnecessary exposure to futility treatments, while results remain interpretable even if thresholds are unmet. Conclusion: The primary goal is to confirm treatment efficacy earlier, enabling trials to be stopped promptly for either superiority or futility treatments. This strategy reduces sample size, time, and financial costs, and prevents patient exposure to futile treatments. Moreover, the study aims to promote the adoption of Bayesian sequential decision-making, thereby accelerating rare disease clinical trial approvals and drug marketing.
- Open Access
Association between elevated serum REG Iα levels and eGFR decline in patients with chronic kidney disease: a cross-sectional study in eastern China
Objectives This study aimed to investigate the relationship between serum regenerating protein Iα (REG Iα) levels and estimated glomerular filtration rate (eGFR) and to evaluate the diagnostic efficiency of REG Iα in chronic kidney disease (CKD). Design This is a cross-sectional study. Setting The study was conducted in eastern China between August 2022 and August 2023. Participants A total of 880 participants aged over 18 years were enrolled, with 220 non-CKD participants (111 males, 50.45%) and 660 patients with CKD (366 males, 55.45%). CKD was diagnosed based on the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 guidelines. Exclusion criteria included participation in other trials, acute kidney injury, end-stage kidney disease undergoing renal replacement therapy, pregnancy, active infections, gastrointestinal or pancreatic inflammation, history of gastrointestinal or pancreatic resections, cancer and mental disorders. Results Serum REG Iα was significantly higher in the CKD group (125.54 (60.28–303.39) ng/mL) compared with those in the non-CKD group (24.62 (14.09–37.32) ng/mL, pConclusions This study provided compelling evidence that serum REG Iα levels were notably elevated in patients with CKD and closely associated with kidney function. REG Iα may serve as a promising biomarker for CKD detection and risk stratification. Clinical trial registration The study was approved by the Ethics Committee of Zhongda Hospital (approval number: 2022ZDSYLL204-P01) and conducted in compliance with the Helsinki Declaration. The clinical trial was registered under ChiCTR2300072247.
Open AccessAssociation between elevated serum REG Iα levels and eGFR decline in patients with chronic kidney disease: a cross-sectional study in eastern China
Objectives This study aimed to investigate the relationship between serum regenerating protein Iα (REG Iα) levels and estimated glomerular filtration rate (eGFR) and to evaluate the diagnostic efficiency of REG Iα in chronic kidney disease (CKD). Design This is a cross-sectional study. Setting The study was conducted in eastern China between August 2022 and August 2023. Participants A total of 880 participants aged over 18 years were enrolled, with 220 non-CKD participants (111 males, 50.45%) and 660 patients with CKD (366 males, 55.45%). CKD was diagnosed based on the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 guidelines. Exclusion criteria included participation in other trials, acute kidney injury, end-stage kidney disease undergoing renal replacement therapy, pregnancy, active infections, gastrointestinal or pancreatic inflammation, history of gastrointestinal or pancreatic resections, cancer and mental disorders. Results Serum REG Iα was significantly higher in the CKD group (125.54 (60.28–303.39) ng/mL) compared with those in the non-CKD group (24.62 (14.09–37.32) ng/mL, pConclusions This study provided compelling evidence that serum REG Iα levels were notably elevated in patients with CKD and closely associated with kidney function. REG Iα may serve as a promising biomarker for CKD detection and risk stratification. Clinical trial registration The study was approved by the Ethics Committee of Zhongda Hospital (approval number: 2022ZDSYLL204-P01) and conducted in compliance with the Helsinki Declaration. The clinical trial was registered under ChiCTR2300072247.
- Open Access
- 1 January 2025
- Published by Journal of Biomedical Research in Journal of Biomedical Research
- Vol. 39 (6) , 611-10
One-year outcomes of a single bolus r-SAK before primary PCI for STEMI: Follow-up of the OPTIMA-5 study
The OPTIMA-5 study demonstrated that a single bolus of half-dose recombinant staphylokinase (r-SAK) before primary percutaneous coronary intervention (PCI) significantly improved the patency of infarct related artery in patients with ST-elevation myocardial infarction (STEMI) expected to undergo PCI within 120 minutes. This study aimed to investigate the 1-year clinical outcome and the effect of the r-SAK antibody on a second r-SAK thrombolysis in OPTIMA-5 patients. The clinical outcome was major adverse cardiovascular events (MACE) within 360 days. Patients' r-SAK antibodies were determined on days 90 ± 7, 180 ± 7, and 360 ± 14 after thrombolysis, and in-vitro r-SAK antibody neutralization experiments were performed to explore an optimal interval for a second r-SAK thrombolysis. Results showed that the MACE incidence was numerically lower in r-SAK group compared with normal saline (NS) group (14.0% vs. 20.0%, HR 0.67, 95% CI: 0.34-1.32; log-rank P=0.245). The r-SAK antibody levels in r-SAK group decreased by time, but kept significantly higher than those in NS group on days 90 ± 7 (2.96 ± 0.68 vs. 0.22 ± 0.53, Pvs. 0.44 ± 0.65, Pvs. 0.37 ± 0.71, Pin-vitro r-SAK antibody neutralization experiments illustrated that the thrombolysis rate decreased exponentially as the antibody titer increased from 1.90 to 2.20 (67.80 ± 14.19% vs. 44.32 ± 21.54%, P< 0.0001). Therefore, for STEMI patients expected to undergo PCI within 120 minutes, a single bolus of half-dose r-SAK before primary PCI may reduce 1-year MACE risk. The r-SAK antibody lasts over 1 year, and a second r-SAK thrombolysis may not be indicated until 1 year after the first r-SAK thrombolysis if necessary.
- 1 January 2025
- Published by Journal of Biomedical Research in Journal of Biomedical Research
- Vol. 39 (6) , 611-10
Open AccessOne-year outcomes of a single bolus r-SAK before primary PCI for STEMI: Follow-up of the OPTIMA-5 study
The OPTIMA-5 study demonstrated that a single bolus of half-dose recombinant staphylokinase (r-SAK) before primary percutaneous coronary intervention (PCI) significantly improved the patency of infarct related artery in patients with ST-elevation myocardial infarction (STEMI) expected to undergo PCI within 120 minutes. This study aimed to investigate the 1-year clinical outcome and the effect of the r-SAK antibody on a second r-SAK thrombolysis in OPTIMA-5 patients. The clinical outcome was major adverse cardiovascular events (MACE) within 360 days. Patients' r-SAK antibodies were determined on days 90 ± 7, 180 ± 7, and 360 ± 14 after thrombolysis, and in-vitro r-SAK antibody neutralization experiments were performed to explore an optimal interval for a second r-SAK thrombolysis. Results showed that the MACE incidence was numerically lower in r-SAK group compared with normal saline (NS) group (14.0% vs. 20.0%, HR 0.67, 95% CI: 0.34-1.32; log-rank P=0.245). The r-SAK antibody levels in r-SAK group decreased by time, but kept significantly higher than those in NS group on days 90 ± 7 (2.96 ± 0.68 vs. 0.22 ± 0.53, Pvs. 0.44 ± 0.65, Pvs. 0.37 ± 0.71, Pin-vitro r-SAK antibody neutralization experiments illustrated that the thrombolysis rate decreased exponentially as the antibody titer increased from 1.90 to 2.20 (67.80 ± 14.19% vs. 44.32 ± 21.54%, P< 0.0001). Therefore, for STEMI patients expected to undergo PCI within 120 minutes, a single bolus of half-dose r-SAK before primary PCI may reduce 1-year MACE risk. The r-SAK antibody lasts over 1 year, and a second r-SAK thrombolysis may not be indicated until 1 year after the first r-SAK thrombolysis if necessary.
- 21 November 2024
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 107 (2) , 196-207
Impact of Cement Distribution on the Efficacy of Percutaneous Vertebral Augmentation for Osteoporotic Fractures
Background: No studies have evaluated the impact of the cement distribution as classified on the basis of the fracture bone marrow edema area (FBMEA) in magnetic resonance imaging (MRI) on the efficacy of percutaneous vertebral augmentation (PVA) for acute osteoporotic vertebral fractures. Methods: The clinical data of patients with acute, painful, single-level thoracolumbar osteoporotic fractures were retrospectively analyzed. The bone cement distribution on the postoperative radiograph was divided into 4 types according to the distribution of the FBMEA on the preoperative MRI. The primary outcomes were the postoperative visual analog scale (VAS) for pain and Oswestry Disability Index (ODI) scores. Cement leakage, adjacent vertebral fractures (an important concern in complications after vertebroplasty and a subset of new fractures), and recollapse of the treated vertebra were also evaluated. Results: A total of 128 patients, 80.5% of whom were female, were included and had follow-up for 24 months. The mean patient age (and standard deviation) was 74.2 ± 8.6 years. The cement distribution was classified as Type I in 18 patients, Type II in 26, Type III in 46, and Type IV in 38. At the primary time point (6 months), there was a significant difference in the ODI score favoring the Type-III and Type-IV groups compared with the Type-I and Type-II groups (adjusted 95% confidence interval [CI]: Type I versus Type II, −2.40 to 4.50; Type I versus Type III, 1.35 to 7.63; Type I versus Type IV, 1.27 to 7.92; Type II versus Type III, 0.67 to 6.21; Type II versus Type IV, 0.63 to 6.46; adjusted p < 0.0083), whereas no significant differences were found between the 4 groups in the VAS pain score. The Type-II and Type-IV groups had a higher incidence of cement leakage, and the Type-III and Type-IV groups had a lower incidence of vertebral recollapse. Conclusions: An adequate distribution of bone cement is advantageous for functional improvement, short-term pain relief, and a lower rate of vertebral recollapse. The FBMEA appears to be a feasible reference marker for evaluating the performance of the PVA procedure. Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.
- 21 November 2024
- Published by Wolters Kluwer Health in Journal of Bone and Joint Surgery
- Vol. 107 (2) , 196-207
Impact of Cement Distribution on the Efficacy of Percutaneous Vertebral Augmentation for Osteoporotic Fractures
Background: No studies have evaluated the impact of the cement distribution as classified on the basis of the fracture bone marrow edema area (FBMEA) in magnetic resonance imaging (MRI) on the efficacy of percutaneous vertebral augmentation (PVA) for acute osteoporotic vertebral fractures. Methods: The clinical data of patients with acute, painful, single-level thoracolumbar osteoporotic fractures were retrospectively analyzed. The bone cement distribution on the postoperative radiograph was divided into 4 types according to the distribution of the FBMEA on the preoperative MRI. The primary outcomes were the postoperative visual analog scale (VAS) for pain and Oswestry Disability Index (ODI) scores. Cement leakage, adjacent vertebral fractures (an important concern in complications after vertebroplasty and a subset of new fractures), and recollapse of the treated vertebra were also evaluated. Results: A total of 128 patients, 80.5% of whom were female, were included and had follow-up for 24 months. The mean patient age (and standard deviation) was 74.2 ± 8.6 years. The cement distribution was classified as Type I in 18 patients, Type II in 26, Type III in 46, and Type IV in 38. At the primary time point (6 months), there was a significant difference in the ODI score favoring the Type-III and Type-IV groups compared with the Type-I and Type-II groups (adjusted 95% confidence interval [CI]: Type I versus Type II, −2.40 to 4.50; Type I versus Type III, 1.35 to 7.63; Type I versus Type IV, 1.27 to 7.92; Type II versus Type III, 0.67 to 6.21; Type II versus Type IV, 0.63 to 6.46; adjusted p < 0.0083), whereas no significant differences were found between the 4 groups in the VAS pain score. The Type-II and Type-IV groups had a higher incidence of cement leakage, and the Type-III and Type-IV groups had a lower incidence of vertebral recollapse. Conclusions: An adequate distribution of bone cement is advantageous for functional improvement, short-term pain relief, and a lower rate of vertebral recollapse. The FBMEA appears to be a feasible reference marker for evaluating the performance of the PVA procedure. Level of Evidence: Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.
- Open Access
- 2 September 2024
- Published by Springer Nature in BMC Public Health
- Vol. 24 (1) , 1-12
Joint and interactive associations of body mass index and genetic factors with cardiovascular disease: a prospective study in UK Biobank
Both body mass index (BMI) and genetic factors independently contribute to cardiovascular disease (CVD). However, it is unclear whether genetic risk modifies the association between BMI and the risk of incident CVD. This study aimed to investigate whether BMI categories and genetic risk jointly and interactively contribute to incident CVD events, including hypertension (HTN), atrial fibrillation (AF), coronary heart disease (CHD), stroke, and heart failure (HF). A total of 496,851 participants from the UK Biobank with one or more new-onset CVD events were included in the analyses. BMI was categorized as normal weight (< 25.0 kg/m2), overweight (25.0–29.9 kg/m2), and obesity (≥ 30.0 kg/m2). Genetic risk for each outcome was defined as low (lowest tertile), intermediate (second tertile), and high (highest tertile) using polygenic risk score. The joint associations of BMI categories and genetic risk with incident CVD were investigated using Cox proportional hazard models. Additionally, additive interactions were evaluated. Among the 496,851 participants, 270,726 (54.5%) were female, with a mean (SD) age was 56.5 (8.1) years. Over a median follow-up (IQR) of 12.4 (11.5–13.1) years, 102,131 (22.9%) participants developed HTN, 26,301 (5.4%) developed AF, 32,222 (6.9%) developed CHD, 10,684 (2.2%) developed stroke, and 13,304 (2.7%) developed HF. Compared with the normal weight with low genetic risk, the obesity with high genetic risk had the highest risk of CVD: HTN (HR: 3.96; 95%CI: 3.84–4.09), AF (HR: 3.60; 95%CI: 3.38–3.83), CHD (HR: 2.76; 95%CI: 2.61–2.91), stroke (HR: 1.44; 95%CI: 1.31–1.57), and HF (HR: 2.47; 95%CI: 2.27–2.69). There were significant additive interactions between BMI categories and genetic risk for HTN, AF, and CHD, with relative excess risk of 0.53 (95%CI: 0.43–0.62), 0.67 (95%CI: 0.51–0.83), and 0.37 (95%CI: 0.25–0.49), respectively. BMI and genetic factors jointly and interactively contribute to incident CVD, especially among participants with high genetic risk. These findings have public health implications for identifying populations more likely to have cardiovascular benefit from weight loss interventions.
- 2 September 2024
- Published by Springer Nature in BMC Public Health
- Vol. 24 (1) , 1-12
Open AccessJoint and interactive associations of body mass index and genetic factors with cardiovascular disease: a prospective study in UK Biobank
Both body mass index (BMI) and genetic factors independently contribute to cardiovascular disease (CVD). However, it is unclear whether genetic risk modifies the association between BMI and the risk of incident CVD. This study aimed to investigate whether BMI categories and genetic risk jointly and interactively contribute to incident CVD events, including hypertension (HTN), atrial fibrillation (AF), coronary heart disease (CHD), stroke, and heart failure (HF). A total of 496,851 participants from the UK Biobank with one or more new-onset CVD events were included in the analyses. BMI was categorized as normal weight (< 25.0 kg/m2), overweight (25.0–29.9 kg/m2), and obesity (≥ 30.0 kg/m2). Genetic risk for each outcome was defined as low (lowest tertile), intermediate (second tertile), and high (highest tertile) using polygenic risk score. The joint associations of BMI categories and genetic risk with incident CVD were investigated using Cox proportional hazard models. Additionally, additive interactions were evaluated. Among the 496,851 participants, 270,726 (54.5%) were female, with a mean (SD) age was 56.5 (8.1) years. Over a median follow-up (IQR) of 12.4 (11.5–13.1) years, 102,131 (22.9%) participants developed HTN, 26,301 (5.4%) developed AF, 32,222 (6.9%) developed CHD, 10,684 (2.2%) developed stroke, and 13,304 (2.7%) developed HF. Compared with the normal weight with low genetic risk, the obesity with high genetic risk had the highest risk of CVD: HTN (HR: 3.96; 95%CI: 3.84–4.09), AF (HR: 3.60; 95%CI: 3.38–3.83), CHD (HR: 2.76; 95%CI: 2.61–2.91), stroke (HR: 1.44; 95%CI: 1.31–1.57), and HF (HR: 2.47; 95%CI: 2.27–2.69). There were significant additive interactions between BMI categories and genetic risk for HTN, AF, and CHD, with relative excess risk of 0.53 (95%CI: 0.43–0.62), 0.67 (95%CI: 0.51–0.83), and 0.37 (95%CI: 0.25–0.49), respectively. BMI and genetic factors jointly and interactively contribute to incident CVD, especially among participants with high genetic risk. These findings have public health implications for identifying populations more likely to have cardiovascular benefit from weight loss interventions.
- Open Access
- 29 August 2024
- Published by Springer Nature in BMC Public Health
- Vol. 24 (1) , 1-11
Associations of ambient air pollution and lifestyle with the risk of NAFLD: a population-based cohort study
Both ambient air pollution and lifestyle factors contribute to the incidence of non-alcoholic fatty liver disease (NAFLD), but previous studies usually focused on single-factor associations. We aimed to assess the joint associations of ambient air pollution and lifestyle with the NAFLD risk and investigate whether lifestyle modifies the association of air pollution with NAFLD risk. A total of 417,025 participants from the UK Biobank were included in this study. Annual average concentrations of NO2, NOx, PM2.5, PM10, and PM2.5−10 were estimated. A composite lifestyle score was determined based on physical activity, alcohol intake, smoking status, dietary patterns, sedentary time, and sleep duration. Cox proportional hazards regression models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs), as well as the population attributable fraction (PAF). Potential additive interactions of air pollution with lifestyle were also examined by the relative excess risk due to the interaction (RERI) and the attributable proportion due to the interaction (AP). 4752 (1.14%) incident NAFLD events were recorded. Long-term exposure to air pollutants and an unhealthy lifestyle were significantly associated with the increased risk of incident NAFLD. Lifestyle was the primary factor of incident NAFLD, with a PAF of 37.18% (95% CI: 29.67%, 44.69%). In addition, a significant additive interaction between air pollution and lifestyle for NAFLD risk was observed (RERI: 0.36, 95% CI: 0.09–0.63). Long-term exposure to ambient air pollutants and poor lifestyle were jointly associated with a higher risk of NAFLD.
- 29 August 2024
- Published by Springer Nature in BMC Public Health
- Vol. 24 (1) , 1-11
Open AccessAssociations of ambient air pollution and lifestyle with the risk of NAFLD: a population-based cohort study
Both ambient air pollution and lifestyle factors contribute to the incidence of non-alcoholic fatty liver disease (NAFLD), but previous studies usually focused on single-factor associations. We aimed to assess the joint associations of ambient air pollution and lifestyle with the NAFLD risk and investigate whether lifestyle modifies the association of air pollution with NAFLD risk. A total of 417,025 participants from the UK Biobank were included in this study. Annual average concentrations of NO2, NOx, PM2.5, PM10, and PM2.5−10 were estimated. A composite lifestyle score was determined based on physical activity, alcohol intake, smoking status, dietary patterns, sedentary time, and sleep duration. Cox proportional hazards regression models were used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs), as well as the population attributable fraction (PAF). Potential additive interactions of air pollution with lifestyle were also examined by the relative excess risk due to the interaction (RERI) and the attributable proportion due to the interaction (AP). 4752 (1.14%) incident NAFLD events were recorded. Long-term exposure to air pollutants and an unhealthy lifestyle were significantly associated with the increased risk of incident NAFLD. Lifestyle was the primary factor of incident NAFLD, with a PAF of 37.18% (95% CI: 29.67%, 44.69%). In addition, a significant additive interaction between air pollution and lifestyle for NAFLD risk was observed (RERI: 0.36, 95% CI: 0.09–0.63). Long-term exposure to ambient air pollutants and poor lifestyle were jointly associated with a higher risk of NAFLD.
- 5 April 2024
- Published by Wiley in Biometrical Journal
- Vol. 66 (3) , e2300094
Beta spending function based on conditional power in group sequential design
Conditional power (CP) serves as a widely utilized approach for futility monitoring in group sequential designs. However, adopting the CP methods may lead to inadequate control of the type II error rate at the desired level. In this study, we introduce a flexible beta spending function tailored to regulate the type II error rate while employing CP based on a predetermined standardized effect size for futility monitoring (a so‐called CP‐beta spending function). This function delineates the expenditure of type II error rate across the entirety of the trial. Unlike other existing beta spending functions, the CP‐beta spending function seamlessly incorporates beta spending concept into the CP framework, facilitating precise stagewise control of the type II error rate during futility monitoring. In addition, the stopping boundaries derived from the CP‐beta spending function can be calculated via integration akin to other traditional beta spending function methods. Furthermore, the proposed CP‐beta spending function accommodates various thresholds on the CP‐scale at different stages of the trial, ensuring its adaptability across different information time scenarios. These attributes render the CP‐beta spending function competitive among other forms of beta spending functions, making it applicable to any trials in group sequential designs with straightforward implementation. Both simulation study and example from an acute ischemic stroke trial demonstrate that the proposed method accurately captures expected power, even when the initially determined sample size does not consider futility stopping, and exhibits a good performance in maintaining overall type I error rates for evident futility.
- 5 April 2024
- Published by Wiley in Biometrical Journal
- Vol. 66 (3) , e2300094
Beta spending function based on conditional power in group sequential design
Conditional power (CP) serves as a widely utilized approach for futility monitoring in group sequential designs. However, adopting the CP methods may lead to inadequate control of the type II error rate at the desired level. In this study, we introduce a flexible beta spending function tailored to regulate the type II error rate while employing CP based on a predetermined standardized effect size for futility monitoring (a so‐called CP‐beta spending function). This function delineates the expenditure of type II error rate across the entirety of the trial. Unlike other existing beta spending functions, the CP‐beta spending function seamlessly incorporates beta spending concept into the CP framework, facilitating precise stagewise control of the type II error rate during futility monitoring. In addition, the stopping boundaries derived from the CP‐beta spending function can be calculated via integration akin to other traditional beta spending function methods. Furthermore, the proposed CP‐beta spending function accommodates various thresholds on the CP‐scale at different stages of the trial, ensuring its adaptability across different information time scenarios. These attributes render the CP‐beta spending function competitive among other forms of beta spending functions, making it applicable to any trials in group sequential designs with straightforward implementation. Both simulation study and example from an acute ischemic stroke trial demonstrate that the proposed method accurately captures expected power, even when the initially determined sample size does not consider futility stopping, and exhibits a good performance in maintaining overall type I error rates for evident futility.