Yong Xiang

Background: While SARS-CoV-2 causes multi-organ complications, comprehensive assessments of long-term mortality across organ systems remain limited. This study systematically evaluated COVID-19’s impact on all-cause and cause-specific mortality. Methods: This cohort study followed 467,522 UK Biobank participants (Jan2020-Dec2022). COVID-19 exposure (representing clinically apparent infection) was classified as overall, hospitalized, and non-hospitalized, and compared with reference cohort without documented SARS-CoV-2 records. Post-acute mortality (>30 days post-infection) was assessed using landmark analyses; overall mortality (all mortalities post-infection) served as complementary analyses. Adjusted Cox models estimated risks for 12 organ systems and 47 diseases, with subgroup analyses by key comorbidities and demographics. Results: Post-acute all-cause mortality was elevated in overall (hazard ratio [HR]: 1.50) and hospitalized (HR: 3.06) COVID-19 cohorts, but not non-hospitalized group. COVID-19 infection increased post-acute mortality from circulatory, digestive, genitourinary, neurological, respiratory, and external causes, as well as neoplasms (though elevated cancer mortality without prior diagnoses may reflect detection bias). Hospitalized cases showed elevated risks across 11/11 organ systems and 27/37 diseases; non-hospitalized cases showed increased risks for external-cause and neurological outcomes. Advanced age, atrial fibrillation, chronic kidney disease, and hypertension exacerbated post-acute all-cause mortality; atrial fibrillation also amplified respiratory and neurological risks. Conclusion: Clinically apparent COVID-19 was associated with elevated post-acute and overall mortality across multiple systems, with hospitalized cases exhibiting the broadest risk spectrum. As controls may include unrecorded infections, these estimates are likely conservative. Sustained monitoring is warranted, particularly for older survivors and those with high-risk comorbidities.

The coronavirus disease 2019 (COVID-19) pandemic has led to substantial health and financial burdens worldwide, and vaccines provide hope for reducing the burden of this pandemic. However, vaccinated people remain at risk for SARS-CoV-2 infection. Genome-wide association studies (GWASs) may identify potential genetic factors involved in the development of COVID-19 breakthrough infections (BIs); however, very few or no GWASs have been conducted for COVID-19 BI thus far. We conducted a GWAS and detailed bioinformatics analysis on COVID-19 BIs in a European population via the UK Biobank (UKBB). We conducted a series of analyses at different levels, including SNP-based, gene-based, pathway, and transcriptome-wide association analyses, to investigate genetic factors associated with COVID-19 BIs and hospitalized infections. The polygenic risk score (PRS) and Hoeffding’s test were performed to reveal the genetic relationships between BIs and other medical conditions. Two independent loci (LD-clumped at r² = 0.01) reached genome-wide significance (p < 5 × 10⁻⁸), including rs36170929, which mapped to LOC102725191/VWDE, and rs28645263, which mapped to RETREG1. A pathway enrichment analysis highlighted pathways such as viral myocarditis, Rho-selective guanine exchange factor AKAP13 signaling, and lipid metabolism. The PRS analyses revealed significant genetic overlap between COVID-19 BIs and heart failure and between HbA1c and type 1 diabetes. Genetic dependence was also observed between COVID-19 BIs and asthma, lung abnormalities, schizophrenia, and type 1 diabetes on the basis of Hoeffding’s test. This GWAS revealed two significant loci that may be associated with COVID-19 BIs and a number of genes and pathways that may be involved in BIs. Genetic overlap with other diseases was identified. Further studies are warranted to replicate these findings and elucidate the mechanisms involved.

Background: COVID-19 is a major public health concern. Given the extent of the pandemic, it is urgent to identify risk factors associated with disease severity. More accurate prediction of those at risk of developing severe infections is of high clinical importance. Objective: Based on the UK Biobank (UKBB), we aimed to build machine learning models to predict the risk of developing severe or fatal infections, and uncover major risk factors involved. Methods: We first restricted the analysis to infected individuals (n=7846), then performed analysis at a population level, considering those with no known infection as controls (ncontrols=465,728). Hospitalization was used as a proxy for severity. A total of 97 clinical variables (collected prior to the COVID-19 outbreak) covering demographic variables, comorbidities, blood measurements (eg, hematological/liver/renal function/metabolic parameters), anthropometric measures, and other risk factors (eg, smoking/drinking) were included as predictors. We also constructed a simplified (lite) prediction model using 27 covariates that can be more easily obtained (demographic and comorbidity data). XGboost (gradient-boosted trees) was used for prediction and predictive performance was assessed by cross-validation. Variable importance was quantified by Shapley values (ShapVal), permutation importance (PermImp), and accuracy gain. Shapley dependency and interaction plots were used to evaluate the pattern of relationships between risk factors and outcomes. Results: A total of 2386 severe and 477 fatal cases were identified. For analyses within infected individuals (n=7846), our prediction model achieved area under the receiving-operating characteristic curve (AUC–ROC) of 0.723 (95% CI 0.711-0.736) and 0.814 (95% CI 0.791-0.838) for severe and fatal infections, respectively. The top 5 contributing factors (sorted by ShapVal) for severity were age, number of drugs taken (cnt_tx), cystatin C (reflecting renal function), waist-to-hip ratio (WHR), and Townsend deprivation index (TDI). For mortality, the top features were age, testosterone, cnt_tx, waist circumference (WC), and red cell distribution width. For analyses involving the whole UKBB population, AUCs for severity and fatality were 0.696 (95% CI 0.684-0.708) and 0.825 (95% CI 0.802-0.848), respectively. The same top 5 risk factors were identified for both outcomes, namely, age, cnt_tx, WC, WHR, and TDI. Apart from the above, age, cystatin C, TDI, and cnt_tx were among the top 10 across all 4 analyses. Other diseases top ranked by ShapVal or PermImp were type 2 diabetes mellitus (T2DM), coronary artery disease, atrial fibrillation, and dementia, among others. For the “lite” models, predictive performances were broadly similar, with estimated AUCs of 0.716, 0.818, 0.696, and 0.830, respectively. The top ranked variables were similar to above, including age, cnt_tx, WC, sex (male), and T2DM. Conclusions: We identified numerous baseline clinical risk factors for severe/fatal infection by XGboost. For example, age, central obesity, impaired renal function, multiple comorbidities, and cardiometabolic abnormalities may predispose to poorer outcomes. The prediction models may be useful at a population level to identify those susceptible to developing severe/fatal infections, facilitating targeted prevention strategies. A risk-prediction tool is also available online. Further replications in independent cohorts are required to verify our findings.

Effective therapies for COVID-19 are still lacking, and drug repositioning is a promising approach to address this problem. Here, we adopted a medical informatics approach to repositioning. We leveraged a large prospective cohort, the UK-Biobank (UKBB, N ~ 397,000), and studied associations of prior use of all level-4 ATC drug categories (N = 819, including vaccines) with COVID-19 diagnosis and severity. Effects of drugs on the risk of infection, disease severity, and mortality were investigated separately. Logistic regression was conducted, controlling for main confounders. We observed strong and highly consistent protective associations with statins. Many top-listed protective drugs were also cardiovascular medications, such as angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARB), calcium channel blocker (CCB), and beta-blockers. Some other drugs showing protective associations included biguanides (metformin), estrogens, thyroid hormones, proton pump inhibitors, and testosterone-5-alpha reductase inhibitors, among others. We also observed protective associations by influenza, pneumococcal, and several other vaccines. Subgroup and interaction analyses were also conducted, which revealed differences in protective effects in various subgroups. For example, protective effects of flu/pneumococcal vaccines were weaker in obese individuals, while protection by statins was stronger in cardiovascular patients. To conclude, our analysis revealed many drug repositioning candidates, for example several cardiovascular medications. Further studies are required for validation.

BACKGROUND COVID-19 is a major public health concern. Given the extent of the pandemic, it is urgent to identify risk factors associated with disease severity. More accurate prediction of those at risk of developing severe infections is of high clinical importance. OBJECTIVE Based on the UK Biobank (UKBB), we aimed to build machine learning models to predict the risk of developing severe or fatal infections, and uncover major risk factors involved. METHODS We first restricted the analysis to infected individuals (n=7846), then performed analysis at a population level, considering those with no known infection as controls (ncontrols=465,728). Hospitalization was used as a proxy for severity. A total of 97 clinical variables (collected prior to the COVID-19 outbreak) covering demographic variables, comorbidities, blood measurements (eg, hematological/liver/renal function/metabolic parameters), anthropometric measures, and other risk factors (eg, smoking/drinking) were included as predictors. We also constructed a simplified (lite) prediction model using 27 covariates that can be more easily obtained (demographic and comorbidity data). XGboost (gradient-boosted trees) was used for prediction and predictive performance was assessed by cross-validation. Variable importance was quantified by Shapley values (ShapVal), permutation importance (PermImp), and accuracy gain. Shapley dependency and interaction plots were used to evaluate the pattern of relationships between risk factors and outcomes. RESULTS A total of 2386 severe and 477 fatal cases were identified. For analyses within infected individuals (n=7846), our prediction model achieved area under the receiving-operating characteristic curve (AUC–ROC) of 0.723 (95% CI 0.711-0.736) and 0.814 (95% CI 0.791-0.838) for severe and fatal infections, respectively. The top 5 contributing factors (sorted by ShapVal) for severity were age, number of drugs taken (cnt_tx), cystatin C (reflecting renal function), waist-to-hip ratio (WHR), and Townsend deprivation index (TDI). For mortality, the top features were age, testosterone, cnt_tx, waist circumference (WC), and red cell distribution width. For analyses involving the whole UKBB population, AUCs for severity and fatality were 0.696 (95% CI 0.684-0.708) and 0.825 (95% CI 0.802-0.848), respectively. The same top 5 risk factors were identified for both outcomes, namely, age, cnt_tx, WC, WHR, and TDI. Apart from the above, age, cystatin C, TDI, and cnt_tx were among the top 10 across all 4 analyses. Other diseases top ranked by ShapVal or PermImp were type 2 diabetes mellitus (T2DM), coronary artery disease, atrial fibrillation, and dementia, among others. For the “lite” models, predictive performances were broadly similar, with estimated AUCs of 0.716, 0.818, 0.696, and 0.830, respectively. The top ranked variables were similar to above, including age, cnt_tx, WC, sex (male), and T2DM. CONCLUSIONS We identified numerous baseline clinical risk factors for severe/fatal infection by XGboost. For example, age, central obesity, impaired renal function, multiple comorbidities, and cardiometabolic abnormalities may predispose to poorer outcomes. The prediction models may be useful at a population level to identify those susceptible to developing severe/fatal infections, facilitating targeted prevention strategies. A risk-prediction tool is also available online. Further replications in independent cohorts are required to verify our findings.

Background: COVID-19 is a major public health concern, yet its risk factors are not well-understood and effective therapies are lacking. It remains unclear how different drugs may increase or decrease the risks of infection and severity of disease. Methods: We studied associations of prior use of all level-4 ATC drug categories (including vaccines) with COVID-19 diagnosis and outcome, based on a prospective cohort of UK Biobank(UKBB). Drug history was based on general practitioner(GP) records. Effects of prescribed medications/vaccinations on the risk of infection, severity of disease and mortality were investigated separately. Hospitalized and fatal cases were categorized as ‘severe’ infection. We also considered different study designs and conducted analyses within infected patients, tested subjects and the whole population respectively, and for 5 different time-windows of prescriptions. Missing data were accounted for by multiple imputation and inverse probability weighting was employed to reduce testing bias. Multivariable logistic regression was conducted which controls for main confounders. Results: We placed a greater focus on protective associations here, as (residual) confounding by indication and comorbidities tends to bias towards harmful effects. Across all categories, statins showed the strongest and most consistent protective associations. Significant protective effects against severe infection were seen among infected subjects (OR for prescriptions within a 12-month window, same below: 0.50, 95% CI:0.42-0.60), tested subjects (OR=0.63, 0.54-0.73) or in the general population (OR=0.49, 0.42-0.57). A number of top-listed drugs with protective effects were also cardiovascular medications, such as angiotensin converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blocker and beta-blockers. Some other drugs showing protective associations included biguanides (metformin), estrogens, thyroid hormones and proton pump inhibitors, among others. Interestingly, we also observed protective associations by numerous vaccines. The most consistent association was observed for influenza vaccines, which showed reduced odds of infection (OR= 0.73 for vaccination in past year, CI 0.65-0.83) when compared cases to general population controls or test-negative controls (OR=0.60, 0.53-0.68). Protective associations were also observed when severe or fatal infection was considered as the outcome. Pneumococcal, tetanus, typhoid and combined bacterial and viral vaccines (ATC code J07CA) were also associated with lower odds of infection/severity. Further subgroup and interaction analyses revealed difference in protective effects in different clinical subgroups. For example, protective effects of flu and pneumococcal vaccines were weaker in obese individuals, while we observed stronger protective effects of statins in those with cardiometabolic disorders, such as diabetes, coronary artery disease, hypertension and obesity. Conclusions: A number of drugs, including many for cardiometabolic disorders, may be associated with lower odds of infection/severity of infection. Several existing vaccines, especially flu vaccines, may be beneficial against COVID-19 as well. However, causal relationship cannot be established due to risk of confounding. While further studies are required to validate the findings, this work provides a useful reference for future meta-analyses, clinical trials or experimental studies.