Advancing Interpretable Regression Analysis for Binary Data: A Novel Distributed Algorithm Approach.

Sparse data bias, where there is a lack of sufficient cases, is a common problem in data analysis, particularly when studying rare binary outcomes. Although a two-step meta-analysis approach may be used to lessen the bias by combining the summary statistics to increase the number of cases from multiple studies, this method does not completely eliminate bias in effect estimation. In this paper, we propose a one-shot distributed algorithm for estimating relative risk using a modified Poisson regression for binary data, named ODAP-B. We evaluate the performance of our method through both simulation studies and real-world case analyses of postacute sequelae of SARS-CoV-2 infection in children using data from 184 501 children across eight national academic medical centers. Compared with the meta-analysis method, our method provides closer estimates of the relative risk for all outcomes considered including syndromic and systemic outcomes. Our method is communication-efficient and privacy-preserving, requiring only aggregated data to obtain relatively unbiased effect estimates compared with two-step meta-analysis methods. Overall, ODAP-B is an effective distributed learning algorithm for Poisson regression to study rare binary outcomes. The method provides inference on adjusted relative risk with a robust variance estimator.

One-shot distributed algorithms for addressing heterogeneity in competing risks data across clinical sites.

OBJECTIVE: To characterize the interplay between multiple medical conditions across sites and account for the heterogeneity in patient population characteristics across sites within a distributed research network, we develop a one-shot algorithm that can efficiently utilize summary-level data from various institutions. By applying our proposed algorithm to a large pediatric cohort across four national Children's hospitals, we replicated a recently published prospective cohort, the RISK study, and quantified the impact of the risk factors associated with the penetrating or stricturing behaviors of pediatric Crohn's disease (PCD). METHODS: In this study, we introduce the ODACoRH algorithm, a one-shot distributed algorithm designed for the competing risks model with heterogeneity. Our approach considers the variability in baseline hazard functions of multiple endpoints of interest across different sites. To accomplish this, we build a surrogate likelihood function by combining patient-level data from the local site with aggregated data from other external sites. We validated our method through extensive simulation studies and replication of the RISK study to investigate the impact of risk factors on the PCD for adolescents and children from four children's hospitals within the PEDSnet, A National Pediatric Learning Health System. To evaluate our ODACoRH algorithm, we compared results from the ODACoRH algorithms with those from meta-analysis as well as those derived from the pooled data. RESULTS: The ODACoRH algorithm had the smallest relative bias to the gold standard method (-0.2%), outperforming the meta-analysis method (-11.4%). In the PCD association study, the estimated subdistribution hazard ratios obtained through the ODACoRH algorithms are identical on par with the results derived from pooled data, which demonstrates the high reliability of our federated learning algorithms. From a clinical standpoint, the identified risk factors for PCD align well with the RISK study published in the Lancet in 2017 and other published studies, supporting the validity of our findings. CONCLUSION: With the ODACoRH algorithm, we demonstrate the capability of effectively integrating data from multiple sites in a decentralized data setting while accounting for between-site heterogeneity. Importantly, our study reveals several crucial clinical risk factors for PCD that merit further investigations.

Learning competing risks across multiple hospitals: one-shot distributed algorithms.

OBJECTIVES: To characterize the complex interplay between multiple clinical conditions in a time-to-event analysis framework using data from multiple hospitals, we developed two novel one-shot distributed algorithms for competing risk models (ODACoR). By applying our algorithms to the EHR data from eight national children's hospitals, we quantified the impacts of a wide range of risk factors on the risk of post-acute sequelae of SARS-COV-2 (PASC) among children and adolescents. MATERIALS AND METHODS: Our ODACoR algorithms are effectively executed due to their devised simplicity and communication efficiency. We evaluated our algorithms via extensive simulation studies as applications to quantification of the impacts of risk factors for PASC among children and adolescents using data from eight children's hospitals including the Children's Hospital of Philadelphia, Cincinnati Children's Hospital Medical Center, Children's Hospital of Colorado covering over 6.5 million pediatric patients. The accuracy of the estimation was assessed by comparing the results from our ODACoR algorithms with the estimators derived from the meta-analysis and the pooled data. RESULTS: The meta-analysis estimator showed a high relative bias (∼40%) when the clinical condition is relatively rare (∼0.5%), whereas ODACoR algorithms exhibited a substantially lower relative bias (∼0.2%). The estimated effects from our ODACoR algorithms were identical on par with the estimates from the pooled data, suggesting the high reliability of our federated learning algorithms. In contrast, the meta-analysis estimate failed to identify risk factors such as age, gender, chronic conditions history, and obesity, compared to the pooled data. DISCUSSION: Our proposed ODACoR algorithms are communication-efficient, highly accurate, and suitable to characterize the complex interplay between multiple clinical conditions. CONCLUSION: Our study demonstrates that our ODACoR algorithms are communication-efficient and can be widely applicable for analyzing multiple clinical conditions in a time-to-event analysis framework.

Development and validation of a federated learning framework for detection of subphenotypes of multisystem inflammatory syndrome in children.

Background: Multisystem inflammatory syndrome in children (MIS-C) is a severe post-acute sequela of SARS-CoV-2 infection. The highly diverse clinical features of MIS-C necessities characterizing its features by subphenotypes for improved recognition and treatment. However, jointly identifying subphenotypes in multi-site settings can be challenging. We propose a distributed multi-site latent class analysis (dMLCA) approach to jointly learn MIS-C subphenotypes using data across multiple institutions. Methods: We used data from the electronic health records (EHR) systems across nine U.S. children's hospitals. Among the 3,549,894 patients, we extracted 864 patients < 21 years of age who had received a diagnosis of MIS-C during an inpatient stay or up to one day before admission. Using MIS-C conditions, laboratory results, and procedure information as input features for the patients, we applied our dMLCA algorithm and identified three MIS-C subphenotypes. As validation, we characterized and compared more granular features across subphenotypes. To evaluate the specificity of the identified subphenotypes, we further compared them with the general subphenotypes identified in the COVID-19 infected patients. Findings: Subphenotype 1 (46.1%) represents patients with a mild manifestation of MIS-C not requiring intensive care, with minimal cardiac involvement. Subphenotype 2 (25.3%) is associated with a high risk of shock, cardiac and renal involvement, and an intermediate risk of respiratory symptoms. Subphenotype 3 (28.6%) represents patients requiring intensive care, with a high risk of shock and cardiac involvement, accompanied by a high risk of >4 organ system being impacted. Importantly, for hospital-specific clinical decision-making, our algorithm also revealed a substantial heterogeneity in relative proportions of these three subtypes across hospitals. Properly accounting for such heterogeneity can lead to accurate characterization of the subphenotypes at the patient-level. Interpretation: Our identified three MIS-C subphenotypes have profound implications for personalized treatment strategies, potentially influencing clinical outcomes. Further, the proposed algorithm facilitates federated subphenotyping while accounting for the heterogeneity across hospitals.

Spectrum of severity of multisystem inflammatory syndrome in children: an EHR-based cohort study from the RECOVER program.

Multi-system inflammatory syndrome in children (MIS-C) is a severe post-acute sequela of SARS-CoV-2 infection in children, and there is a critical need to unfold its highly heterogeneous disease patterns. Our objective was to characterize the illness spectrum of MIS-C for improved recognition and management. We conducted a retrospective cohort study using data from March 1, 2020-September 30, 2022, in 8 pediatric medical centers from PEDSnet. We included 1139 children hospitalized with MIS-C and used their demographics, symptoms, conditions, laboratory values, and medications for analyses. We applied heterogeneity-adaptive latent class analyses and identified three latent classes. We further characterized the sociodemographic and clinical characteristics of the latent classes and evaluated their temporal patterns. Class 1 (47.9%) represented children with the most severe presentation, with more admission to the ICU, higher inflammatory markers, hypotension/shock/dehydration, cardiac involvement, acute kidney injury and respiratory involvement. Class 2 (23.3%) represented a moderate presentation, with 4-6 organ systems involved, and some overlapping features with acute COVID-19. Class 3 (28.8%) represented a mild presentation. Our results indicated that MIS-C has a spectrum of clinical severity ranging from mild to severe and the proportion of severe or critical MIS-C decreased over time.

An optimal two-stage exploratory basket trial design with aggregated futility analysis.

A basket trial investigates the effects of one drug on multiple tumor indications. To discontinue potentially inactive indications early, an interim futility analysis is usually conducted for each indication individually once it reaches the pre-specified sample size. As enrollment rates vary among indications, the futility decisions for slow-enrolling indications could be made much later than other fast-enrolling indications, which could delay the overall decision for the trial significantly. To accelerate the futility decision in early-stage exploratory basket trials and potentially reallocate resources to other compounds earlier while still controlling the global type-I and type-II errors, we propose an optimal two-stage basket trial design with one aggregated futility analysis by aggregating (e.g., pooling) all indications together. The total sample size across all indications is pre-specified for the futility analysis, while the sample size per indication can be adapted to the enrollment rate. The final analysis is performed using the pruning and pooling approach (Chen et al. 2016). The design parameters are optimized by minimizing the expected total sample size under the null hypothesis, while explicitly controlling the global type-I and the type-II error rates. Simulation studies demonstrate that the proposed design has better operating characteristics than the designs with individual futility analysis (Zhou et al. 2019; Wu et al. 2021), while allowing for earlier futility decision.

Clinical Subphenotypes of Multisystem Inflammatory Syndrome in Children: An EHR-based cohort study from the RECOVER program.

Background: Multi-system inflammatory syndrome in children (MIS-C) represents one of the most severe post-acute sequelae of SARS-CoV-2 infection in children, and there is a critical need to characterize its disease patterns for improved recognition and management. Our objective was to characterize subphenotypes of MIS-C based on presentation, demographics and laboratory parameters. Methods: We conducted a retrospective cohort study of children with MIS-C from March 1, 2020 - April 30, 2022 and cared for in 8 pediatric medical centers that participate in PEDSnet. We included demographics, symptoms, conditions, laboratory values, medications and outcomes (ICU admission, death), and grouped variables into eight categories according to organ system involvement. We used a heterogeneity-adaptive latent class analysis model to identify three clinically-relevant subphenotypes. We further characterized the sociodemographic and clinical characteristics of each subphenotype, and evaluated their temporal patterns. Findings: We identified 1186 children hospitalized with MIS-C. The highest proportion of children (44·4%) were aged between 5-11 years, with a male predominance (61.0%), and non- Hispanic white ethnicity (40·2%). Most (67·8%) children did not have a chronic condition. Class 1 represented children with a severe clinical phenotype, with 72·5% admitted to the ICU, higher inflammatory markers, hypotension/shock/dehydration, cardiac involvement, acute kidney injury and respiratory involvement. Class 2 represented a moderate presentation, with 4-6 organ systems involved, and some overlapping features with acute COVID-19. Class 3 represented a mild presentation, with fewer organ systems involved, lower CRP, troponin values and less cardiac involvement. Class 1 initially represented 51·1% of children early in the pandemic, which decreased to 33·9% from the pre-delta period to the omicron period. Interpretation: MIS-C has a spectrum of clinical severity, with degree of laboratory abnormalities rather than the number of organ systems involved providing more useful indicators of severity. The proportion of severe/critical MIS-C decreased over time. Research in context: Evidence before this study: We searched PubMed and preprint articles from December 2019, to July 2022, for studies published in English that investigated the clinical subphenotypes of MIS-C using the terms "multi-system inflammatory syndrome in children" or "pediatric inflammatory multisystem syndrome" and "phenotypes". Most previous research described the symptoms, clinical characteristics and risk factors associated with MIS-C and how these differ from acute COVID-19, Kawasaki Disease and Toxic Shock Syndrome. One single-center study of 63 patients conducted in 2020 divided patients into Kawasaki and non-Kawasaki disease subphenotypes. Another CDC study evaluated 3 subclasses of MIS-C in 570 children, with one class representing the highest number of organ systems, a second class with predominant respiratory system involvement, and a third class with features overlapping with Kawasaki Disease. However, this study evaluated cases from March to July 2020, during the early phase of the pandemic when misclassification of cases as Kawasaki disease or acute COVID-19 may have occurred. Therefore, it is not known from the existing literature whether the presentation of MIS-C has changed with newer variants such as delta and omicron.Added value of this study: PEDSnet provides one of the largest MIS-C cohorts described so far, providing sufficient power for detailed analyses on MIS-C subphenotypes. Our analyses span the entire length of the pandemic, including the more recent omicron wave, and provide an update on the presentations of MIS-C and its temporal dynamics. We found that children have a spectrum of illness that can be characterized as mild (lower inflammatory markers, fewer organ systems involved), moderate (4-6 organ involvement with clinical overlap with acute COVID-19) and severe (higher inflammatory markers, critically ill, more likely to have cardiac involvement, with hypotension/shock and need for vasopressors).Implications of all the available evidence: These results provide an update to the subphenotypes of MIS-C including the more recent delta and omicron periods and aid in the understanding of the various presentations of MIS-C. These and other findings provide a useful framework for clinicians in the recognition of MIS-C, identify factors associated with children at risk for increased severity, including the importance of laboratory parameters, for risk stratification, and to facilitate early evaluation, diagnosis and treatment.