Characterization of Long COVID Definitions and Clinical Coding Practices (preprint)

Background: Long COVID characterized as post-acute sequelae of SARS-CoV-2 (PASC) has no universal clinical case definition. Recent efforts have focused on understanding long COVID symptoms and electronic health records (EHR) data provides a unique resource for understanding this condition. The introduction of the International Classification of Diseases (ICD)-10 code U09.9 for - Post COVID-19 condition, unspecified to identify patients with long COVID has provided a method of evaluating this condition in EHRs, however, the accuracy of this code is unclear. Objective: Our study aimed to characterize the utility and accuracy of the U09.9 code across three healthcare systems - The Veterans Health Administration (VHA), Beth Israel Deaconess Medical Center (BIDMC) and The University of Pittsburgh Medical Center (UPMC) against patients identified with long COVID via a chart review by operationalizing the World Health Organization (WHO) and Centers for Disease Control (CDC) definitions. Methods: COVID positive patients with either a U07.1 ICD code or positive polymerase chain reaction (PCR) test within these healthcare systems were identified for chart review. Among this cohort we sampled patients based on two approaches i) with a U09.9 code and ii) without a U09.9 code but with a new onset PASC related ICD code, which allows us to assess the sensitivity of the U09.9 code. To operationalize the long COVID definition based on health agency guidelines, we grouped symptoms into a core cluster of 11 commonly reported symptoms among long COVID patients and an extended cluster, that captured all other symptoms by disease domain. Patients having at least 2 symptoms persisting for >=60 days that were new onset after their COVID infection, with at least one symptom in the core cluster, were labeled as having long COVID per chart review. We compared the performance of the code across three health systems and across different time periods of the pandemic. Results: A total of 900 patient charts were reviewed across 3 healthcare systems. The prevalence of long COVID among the cohort with the U09.9 ICD code, based on the operationalized WHO definition was between 23.2%-62.4% across these healthcare systems. We also evaluated a less stringent version of the WHO definition and the Centers for Disease Control (CDC) definition and observed an increase in the prevalence of long COVID at all three healthcare systems. Conclusions: This is one of the first studies to evaluate the U09.9 code against a clinical case definition for long COVID, as well as the first to apply this definition to EHR data using a chart review approach on a nationwide cohort across multiple healthcare systems. This chart review approach can be implemented at other EHR systems to further evaluate the utility and performance of the U09.9 code.

Characterizing the use of the ICD-10 Code for Long COVID in 3 US Healthcare Systems (preprint)

The International Classification of Diseases (ICD)-10 code (U09.9) for post-acute sequelae of COVID-19 (PASC) was introduced in October of 2021. As researchers seek to leverage this billing code for research purposes in large scale real-world studies of PASC, it is of utmost importance to understand the functional use of the code by healthcare providers and the clinical characteristics of patients who have been assigned this code. To this end, we operationalized clinical case definitions of PASC using World Health Organization and Centers for Disease Control guidelines. We then chart reviewed 300 patients with COVID-19 from three participating healthcare systems of the 4CE Consortium who were assigned the U09.9 code. Chart review results showed the average positive predictive value (PPV) of the U09.9 code ranged from 40.2% to 65.4% depending on which definition of PASC was used in the evaluation. The PPV of the U09.9 code also fluctuated significantly between calendar time periods. We demonstrated the potential utility of textual data extracted from natural language processing techniques to more comprehensively capture symptoms associated with PASC from electronic health records data. Finally, we investigated the utilization of long COVID clinics in the cohort of patients. We observed that only an average of 24.0% of patients with the U09.9 code visited a long COVID clinic. Among patients who met the WHO PASC definition, only an average of 35.6% visited a long COVID clinic.

Acute respiratory distress syndrome after SARS-CoV-2 infection on young adult population: international observational federated study based on electronic health records through the 4CE consortium ARDS after SARS-CoV-2 infection on young adult (preprint)

Purpose : In young adults (18 to 49 years old), investigation of the acute respiratory distress syndrome (ARDS) after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has been limited. We evaluated the risk factors and outcomes of ARDS following infection with SARS-CoV-2 in a young adult population. Methods : A retrospective cohort study was conducted between January 1st, 2020 and February 28th, 2021 using patient-level electronic health records (EHR), across 241 United States hospitals and 43 European hospitals participating in the Consortium for Clinical Characterization of COVID-19 by EHR (4CE). To identify the risk factors associated with ARDS, we compared young patients with and without ARDS through a federated analysis. We further compared the outcomes between young and old patients with ARDS. Results : Among the 75,377 hospitalized patients with positive SARS-CoV-2 PCR, 1001 young adults presented with ARDS ( 7.8% of young hospitalized adults). Their mortality rate at 90 days was 16.2% and they presented with a similar complication rate for infection than older adults with ARDS. Peptic ulcer disease, paralysis, obesity, congestive heart failure, valvular disease, diabetes, chronic pulmonary disease and liver disease were associated with a higher risk of ARDS. We described a high prevalence of obesity (53%), hypertension (38%- although not significantly associated with ARDS), and diabetes (32%). Conclusion : Trough an innovative method, a large international cohort study of young adults developing ARDS after SARS-CoV-2 infection has been gather. It demonstrated the poor outcomes of this population and associated risk factor.

SurvMaximin: Robust Federated Approach to Transporting Survival Risk Prediction Models (preprint)

ObjectiveFor multi-center heterogeneous Real-World Data (RWD) with time-to-event outcomes and high-dimensional features, we propose the SurvMaximin algorithm to estimate Cox model feature coefficients for a target population by borrowing summary information from a set of health care centers without sharing patient-level information. Materials and MethodsFor each of the centers from which we want to borrow information to improve the prediction performance for the target population, a penalized Cox model is fitted to estimate feature coefficients for the center. Using estimated feature coefficients and the covariance matrix of the target population, we then obtain a SurvMaximin estimated set of feature coefficients for the target population. The target population can be an entire cohort comprised of all centers, corresponding to federated learning, or can be a single center, corresponding to transfer learning. ResultsSimulation studies and a real-world international electronic health records application study, with 15 participating health care centers across three countries (France, Germany, and the U.S.), show that the proposed SurvMaximin algorithm achieves comparable or higher accuracy compared with the estimator using only the information of the target site and other existing methods. The SurvMaximin estimator is robust to variations in sample sizes and estimated feature coefficients between centers, which amounts to significantly improved estimates for target sites with fewer observations. ConclusionsThe SurvMaximin method is well suited for both federated and transfer learning in the high-dimensional survival analysis setting. SurvMaximin only requires a one-time summary information exchange from participating centers. Estimated regression vectors can be very heterogeneous. SurvMaximin provides robust Cox feature coefficient estimates without outcome information in the target population and is privacy-preserving.

Polygenic predisposition to venous thromboembolism is associated with increased COVID-19 positive testing rates (preprint)

Genetic predisposition to venous thrombosis may impact COVID-19 infection and its sequelae. Participants in the ongoing prospective cohort study, Million Veteran Program (MVP), who were tested for COVID-19, with European ancestry, were evaluated for associations with polygenic venous thromboembolic risk, Factor V Leiden mutation (FVL) (rs6025) and prothrombin gene 3 -UTR mutation (F2 G20210A)(rs1799963), and their interactions. Logistic regression models assessed genetic associations with VTE diagnosis, COVID-19 (positive) testing rates and outcome severity (modified WHO criteria), and post-test conditions, adjusting for outpatient anticoagulation medication usage, age, sex, and genetic principal components. 108,437 out of 464,961 European American MVP participants were tested for COVID-19 with 9786 (9%) positive. PRS(VTE), FVL, F2 G20210A were not significantly associated with the propensity of being tested for COVID-19. PRS(VTE) was significantly associated with a positive COVID-19 test in F5 wild type (WT) individuals (OR 1.05; 95% CI [1.02-1.07]), but not in FVL carriers (0.97, [0.91-1.94]). There was no association with severe outcome for FVL, F2 G20210A or PRS(VTE). Outpatient anticoagulation usage in the two years prior to testing was associated with worse clinical outcomes. PRS(VTE) was associated with prevalent VTE diagnosis among both FVL carriers or F5 wild type individuals as well as incident VTE in the two years prior to testing. Increased genetic propensity for VTE in the MVP was associated with increased COVID-19 positive testing rates, suggesting a role of coagulation in the initial steps of COVID-19 infection. Key PointsO_LIIncreased genetic predisposition to venous thrombosis is associated with increased COVID-19 positive testing rates. C_LIO_LIPRS for VTE further risk stratifies factor V Leiden carriers regarding their VTE risk. C_LI

Federated Adaptive Causal Estimation (FACE) of Target Treatment Effects (preprint)

Federated learning of causal estimands may greatly improve estimation efficiency by leveraging data from multiple study sites, but robustness to heterogeneity and model misspecifications is vital for ensuring validity. We develop a Federated Adaptive Causal Estimation (FACE) framework to incorporate heterogeneous data from multiple sites to provide treatment effect estimation and inference for a flexibly specified target population of interest. FACE accounts for site-level heterogeneity in the distribution of covariates through density ratio weighting. To safely incorporate source sites and avoid negative transfer, we introduce an adaptive weighting procedure via a penalized regression, which achieves both consistency and optimal efficiency. Our strategy is communication-efficient and privacy-preserving, allowing participating sites to share summary statistics only once with other sites. We conduct both theoretical and numerical evaluations of FACE and apply it to conduct a comparative effectiveness study of BNT162b2 (Pfizer) and mRNA-1273 (Moderna) vaccines on COVID-19 outcomes in U.S. veterans using electronic health records from five VA regional sites. We show that compared to traditional methods, FACE meaningfully increases the precision of treatment effect estimates, with reductions in standard errors ranging from $26\%$ to $67\%$.

A MUC5B gene polymorphism, rs35705950-T, confers protective effects in COVID-19 infection (preprint)

RationaleA common MUC5B gene polymorphism, rs35705950-T, is associated with idiopathic pulmonary fibrosis, but its role in the SARS-CoV-2 infection and disease severity is unclear. ObjectivesTo assess whether rs35705950-T confers differential risk for clinical outcomes associated with COVID-19 infection among participants in the Million Veteran Program (MVP) and COVID-19 Host Genetics Initiative (HGI). MethodsMVP participants were examined for an association between the incidence or severity of COVID-19 and the presence of a MUC5B rs35705950-T allele. Comorbidities and clinical events were extracted from the electronic health records (EHR). The analysis was performed within each ancestry group in the MVP, adjusting for sex, age, age2, and first twenty principal components followed by a trans-ethnic meta-analysis. We then pursued replication and performed a meta-analysis with the trans-ethnic summary statistics from the HGI. A phenome-wide association study (PheWAS) of the rs35705950-T was conducted to explore associated pathophysiologic conditions. Measurements and Main ResultsA COVID-19 severity scale was modified from the World Health Organization criteria, and phenotypes derived from the International Classification of Disease-9/10 were extracted from EHR. Presence of rs35705950-T was associated with fewer hospitalizations (Ncases=25353, Ncontrols=631,024; OR=0.86 [0.80-0.93], p=7.4 x 10-5) in trans-ethnic meta-analysis within MVP and joint meta-analyses with the HGI (N=1641311; OR=0.89 [0.85-0.93], p =1.9 x 10-6). Moreover, individuals of European Ancestry with at least one copy of rs35705950-T had fewer post-COVID-19 pneumonia events (OR=0.85 [0.76-0.96], p =0.008). PheWAS exclusively revealed pulmonary involvement. ConclusionsThe MUC5B variant rs35705950-T is protective in COVID-19 infection.

Proton pump inhibitor use is not associated with severe COVID-19 related outcomes: A propensity score weighted analysis of a national veteran cohort (preprint)

Background and Aims: Low pH deactivates most pathogens, including coronaviruses. Proton pump inhibitors (PPIs) are potent gastric acid suppressing medications. Whether PPI use vs non-use is associated with severe Coronavirus disease-2019 (COVID-19) outcomes remains uncertain. We aimed to compare severe COVID-19 outcomes between current outpatient PPI users and non-users. Methods: We conducted a retrospective propensity score-weighted analysis of a national cohort of US veterans with established care who tested positive for severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) through January 9, 2021, and who had 60 days of follow-up. The positive test date was the index date. Current outpatient PPI use up to and including the index date (primary exposure) was compared to non-use, defined as no PPI prescription fill in the 365 days prior to the index date. The primary outcome was a composite of use of mechanical ventilation or death within 60 days. Weighted logistic regression models evaluated severe COVID-19 outcomes between current PPI users vs non-users. Results: Of 97,674 Veterans with SARS-CoV-2 testing, 14,958 tested positive (6262 [41.9%] current PPI users, 8696 [58.1%] non-users) and comprised the analytic cohort. After weighting, all covariates were well-balanced. In the weighted cohort, there was no difference in the primary composite outcome (8.2% vs 8.0%; OR 1.03, 95% CI 0.91-1.16), secondary composite outcome, nor individual component outcomes between current PPI users and non-users. There was no significant interaction between age and PPI use on outcomes. Conclusion: Among patients with SARS-CoV-2 infection, current PPI use vs non-use is not associated with severe COVID-19 outcomes.

IL10RB as a key regulator of COVID-19 host susceptibility and severity (preprint)

Background Recent efforts have identified genetic loci that are associated with coronavirus disease 2019 (COVID-19) infection rates and disease outcome severity. Translating these genetic findings into druggable genes and readily available compounds that reduce COVID-19 host susceptibility is a critical next step. Methods We integrate COVID-19 genetic susceptibility variants, multi-tissue genetically regulated gene expression (GReX) and perturbargen signatures to identify candidate genes and compounds that reverse the predicted gene expression dysregulation associated with COVID-19 susceptibility. The top candidate gene is validated by testing both its GReX and observed blood transcriptome association with COVID-19 severity, as well as by in vitro perturbation to quantify effects on viral load and molecular pathway dysregulation. We validate the in silico drug repositioning analysis by examining whether the top candidate compounds decrease COVID-19 incidence based on epidemiological evidence. Results We identify IL10RB as the top key regulator of COVID-19 host susceptibility. Predicted GReX up-regulation of IL10RB and higher IL10RB expression in COVID-19 patient blood is associated with worse COVID-19 outcomes. In vitro IL10RB overexpression is associated with increased viral load and activation of immune-related molecular pathways. Azathioprine and retinol are prioritized as candidate compounds to reduce the likelihood of testing positive for COVID-19. Conclusions We establish an integrative data-driven approach for gene target prioritization. We identify and validate IL10RB as a suitable molecular target for modulation of COVID-19 host susceptibility. Finally, we provide evidence for a few readily available medications that would warrant further investigation as drug repositioning candidates.

A Phenome-Wide Association Study of genes associated with COVID-19 severity reveals shared genetics with complex diseases in the Million Veteran Program (preprint)

The study aims to determine the shared genetic architecture between COVID-19 severity with existing medical conditions using electronic health record (EHR) data. We conducted a Phenome-Wide Association Study (PheWAS) of genetic variants associated with critical illness (n=35) or hospitalization (n=42) due to severe COVID-19 using genome-wide association summary from the Host Genetics Initiative. PheWAS analysis was performed using genotype-phenotype data from the Veterans Affairs Million Veteran Program (MVP). Phenotypes were defined by International Classification of Diseases (ICD) codes mapped to clinically relevant groups using published PheWAS methods. Among 658,582 Veterans, variants associated with severe COVID-19 were tested for association across 1,559 phenotypes. Variants at the ABO locus (rs495828, rs505922) associated with the largest number of phenotypes (nrs495828= 53 and nrs505922=59); strongest association with venous embolism, odds ratio (ORrs495828 1.33 (p=1.32 x 10-199), and thrombosis ORrs505922 1.33, p=2.2 x10-265. Among 67 respiratory conditions tested, 11 had significant associations including MUC5B locus (rs35705950) with increased risk of idiopathic fibrosing alveolitis OR 2.83, p=4.12 x 10-191; CRHR1 (rs61667602) associated with reduced risk of pulmonary fibrosis, OR 0.84, p=2.26x 10-12. The TYK2 locus (rs11085727) associated with reduced risk for autoimmune conditions, e.g., psoriasis OR 0.88, p=6.48 x10-23, lupus OR 0.84, p=3.97 x 10-06. PheWAS stratified by genetic ancestry demonstrated differences in genotype-phenotype associations across ancestry. LMNA (rs581342) associated with neutropenia OR 1.29 p=4.1 x 10-13 among Veterans of African ancestry but not European. Overall, we observed a shared genetic architecture between COVID-19 severity and conditions related to underlying risk factors for severe and poor COVID-19 outcomes. Differing associations between genotype-phenotype across ancestries may inform heterogenous outcomes observed with COVID-19. Divergent associations between risk for severe COVID-19 with autoimmune inflammatory conditions both respiratory and non-respiratory highlights the shared pathways and fine balance of immune host response and autoimmunity and caution required when considering treatment targets.

Actionable druggable genome-wide Mendelian randomization identifies repurposingopportunities for COVID-19 (preprint)

Drug repurposing provides a rapid approach to meet the urgent need for therapeutics to address COVID-19. To identify therapeutic targets relevant to COVID-19, we conducted Mendelian randomization (MR) analyses, deriving genetic instruments based on transcriptomic and proteomic data for 1,263 actionable proteins that are targeted by approved drugs or in clinical phase of drug development. Using summary statistics from the Host Genetics Initiative and the Million Veteran Program, we studied 7,554 patients hospitalized with COVID-19 and >1 million controls. We found significant Mendelian randomization results for three proteins (ACE2: P=1.6x10^-6, IFNAR2: P=9.8x10^-11, and IL-10RB: P=1.9x10^-14) using cis-eQTL genetic instruments that also had strong evidence for colocalization with COVID-19 hospitalization. To disentangle the shared eQTL signal for IL10RB and IFNAR2, we conducted phenome-wide association scans and pathway enrichment analysis, which suggested that IFNAR2 is more likely to play a role in COVID-19 hospitalization. Our findings prioritize trials of drugs targeting IFNAR2 and ACE2 for early management of COVID-19.