Summary We consider the problem of conditional independence testing: given a response Y and covariates (X, Z), we test the null hypothesis that Y ╨ X∣Z. The conditional randomization test was recently proposed as a way to use distributional information about X∣Z to exactly and non-asymptotically control Type-I error using any test statistic in any dimensionality without assuming anything about Y ∣ (X, Z). This flexibility, in principle, allows one to derive powerful test statistics from complex prediction algorithms while maintaining statistical validity. Yet the direct use of such advanced test statistics in the conditional randomization test is prohibitively computationally expensive, especially with multiple testing, due to the requirement to recompute the test statistic many times on resampled data. We propose the distilled conditional randomization test, a novel approach to using state-of-the-art machine learning algorithms in the conditional randomization test while drastically reducing the number of times those algorithms need to be run, thereby taking advantage of their power and the conditional randomization test’s statistical guarantees without suffering the usual computational expense. In addition to distillation, we propose a number of other tricks like screening and recycling computations to further speed up the conditional randomization test without sacrificing its high power and exact validity. Indeed, we show in simulations that all our proposals combined lead to a test that has similar power to the most powerful existing conditional randomization test implementations but requires orders of magnitude less computation, making it a practical tool even for large datasets. We demonstrate these benefits on a breast cancer dataset by identifying biomarkers related to cancer stage.

This cohort study aims to describe international hospitalization trends and key epidemiological and clinical features of children and youth with COVID-19.

In relating a response variable Y to covariates (Z, X), a key question is whether Y is independent of the covariate X given Z. This question can be answered through conditional independence testing, and the conditional randomization test (CRT) was recently proposed by Candès et al. (2018) as a way to use distributional information about X | Z to exactly (nonasymptotically) test for conditional independence between X and Y using any test statistic in any dimensionality without assuming anything about Y | (Z, X). This flexibility in principle allows one to derive powerful test statistics from complex state-of-the-art machine learning algorithms while maintaining exact statistical control of Type 1 errors. Yet the direct use of such advanced test statistics in the CRT is prohibitively computationally expensive, especially with multiple testing, due to the CRT's requirement to recompute the test statistic many times on resampled data. In this paper we propose a novel approach, called distillation, to using stateof-the-art machine learning algorithms in the CRT while drastically reducing the number of times those algorithms need to be run, thereby taking advantage of their power and the CRT's statistical guarantees without suffering the usual computational expense associated with their use in the CRT. In addition to distillation, we propose a number of other tricks to speed up the CRT without sacrificing its strong statistical guarantees, and show in simulation that all our proposals combined lead to a test that has the same power as the CRT but requires orders of magnitude less computation, making it a practical and powerful tool even for large data sets. We demonstrate our method's speed and power on a breast cancer dataset by identifying biomarkers related to cancer stage.

The increasing availability of electronic health record (EHR) systems has created enormous potential for translational research. However, it is difficult to know all the relevant codes related to a phenotype due to the large number of codes available. Traditional data mining approaches often require the use of patient-level data, which hinders the ability to share data across institutions. In this project, we demonstrate that multi-center large-scale code embeddings can be used to efficiently identify relevant features related to a disease of interest. We constructed large-scale code embeddings for a wide range of codified concepts from EHRs from two large medical centers. We developed knowledge extraction via sparse embedding regression (KESER) for feature selection and integrative network analysis. We evaluated the quality of the code embeddings and assessed the performance of KESER in feature selection for eight diseases. Besides, we developed an integrated clinical knowledge map combining embedding data from both institutions. The features selected by KESER were comprehensive compared to lists of codified data generated by domain experts. Features identified via KESER resulted in comparable performance to those built upon features selected manually or with patient-level data. The knowledge map created using an integrative analysis identified disease-disease and disease-drug pairs more accurately compared to those identified using single institution data. Analysis of code embeddings via KESER can effectively reveal clinical knowledge and infer relatedness among codified concepts. KESER bypasses the need for patient-level data in individual analyses providing a significant advance in enabling multi-center studies using EHR data.

Genome-wide association studies (GWAS) have underrepresented individuals from non-European populations, impeding progress in characterizing the genetic architecture and consequences of health and disease traits. To address this, we present a population-stratified phenome-wide GWAS followed by a multi-population meta-analysis for 2,068 traits derived from electronic health records of 635,969 participants in the Million Veteran Program (MVP), a longitudinal cohort study of diverse U.S. Veterans genetically similar to the respective African (121,177), Admixed American (59,048), East Asian (6,702), and European (449,042) superpopulations defined by the 1000 Genomes Project. We identified 38,270 independent variants associating with one or more traits at experiment-wide (P < 4.6x10-11) significance; fine-mapping 6,318 signals identified from 613 traits to single-variant resolution. Among these, a third (2,069) of the associations were found only among participants genetically similar to non-European reference populations, demonstrating the importance of expanding diversity in genetic studies. Our work provides a comprehensive atlas of phenome-wide genetic associations for future studies dissecting the architecture of complex traits in diverse populations.

No abstract

Objectives: To perform an international comparison of the trajectory of laboratory values among hospitalized patients with COVID-19 who develop severe disease and identify optimal timing of laboratory value collection to predict severity across hospitals and regions. Design: Retrospective cohort study. Setting: The Consortium for Clinical Characterization of COVID-19 by EHR (4CE), an international multi-site data-sharing collaborative of 342 hospitals in the US and in Europe. Participants: Patients hospitalized with COVID-19, admitted before or after PCR-confirmed result for SARS-CoV-2. Primary and secondary outcome measures: Patients were categorized as ″ever-severe″ or ″never-severe″ using the validated 4CE severity criteria. Eighteen laboratory tests associated with poor COVID-19-related outcomes were evaluated for predictive accuracy by area under the curve (AUC), compared between the severity categories. Subgroup analysis was performed to validate a subset of laboratory values as predictive of severity against a published algorithm. A subset of laboratory values (CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin) was compared between North American and European sites for severity prediction. Results: Of 36,447 patients with COVID-19, 19,953 (43.7%) were categorized as ever-severe. Most patients (78.7%) were 50 years of age or older and male (60.5%). Longitudinal trajectories of CRP, albumin, LDH, neutrophil count, D-dimer, and procalcitonin showed association with disease severity. Significant differences of laboratory values at admission were found between the two groups. With the exception of D-dimer, predictive discrimination of laboratory values did not improve after admission. Sub-group analysis using age, D-dimer, CRP, and lymphocyte count as predictive of severity at admission showed similar discrimination to a published algorithm (AUC=0.88 and 0.91, respectively). Both models deteriorated in predictive accuracy as the disease progressed. On average, no difference in severity prediction was found between North American and European sites. Conclusions: Laboratory test values at admission can be used to predict severity in patients with COVID-19. Prediction models show consistency across international sites highlighting the potential generalizability of these models.

No abstract