Causal Modeling to Mitigate Selection Bias and Unmeasured Confounding in Internet-Based Epidemiology of COVID-19: Model Development and Validation.

BACKGROUND: Selection bias and unmeasured confounding are fundamental problems in epidemiology that threaten study internal and external validity. These phenomena are particularly dangerous in internet-based public health surveillance, where traditional mitigation and adjustment methods are inapplicable, unavailable, or out of date. Recent theoretical advances in causal modeling can mitigate these threats, but these innovations have not been widely deployed in the epidemiological community. OBJECTIVE: The purpose of our paper is to demonstrate the practical utility of causal modeling to both detect unmeasured confounding and selection bias and guide model selection to minimize bias. We implemented this approach in an applied epidemiological study of the COVID-19 cumulative infection rate in the New York City (NYC) spring 2020 epidemic. METHODS: We collected primary data from Qualtrics surveys of Amazon Mechanical Turk (MTurk) crowd workers residing in New Jersey and New York State across 2 sampling periods: April 11-14 and May 8-11, 2020. The surveys queried the subjects on household health status and demographic characteristics. We constructed a set of possible causal models of household infection and survey selection mechanisms and ranked them by compatibility with the collected survey data. The most compatible causal model was then used to estimate the cumulative infection rate in each survey period. RESULTS: There were 527 and 513 responses collected for the 2 periods, respectively. Response demographics were highly skewed toward a younger age in both survey periods. Despite the extremely strong relationship between age and COVID-19 symptoms, we recovered minimally biased estimates of the cumulative infection rate using only primary data and the most compatible causal model, with a relative bias of +3.8% and -1.9% from the reported cumulative infection rate for the first and second survey periods, respectively. CONCLUSIONS: We successfully recovered accurate estimates of the cumulative infection rate from an internet-based crowdsourced sample despite considerable selection bias and unmeasured confounding in the primary data. This implementation demonstrates how simple applications of structural causal modeling can be effectively used to determine falsifiable model conditions, detect selection bias and confounding factors, and minimize estimate bias through model selection in a novel epidemiological context. As the disease and social dynamics of COVID-19 continue to evolve, public health surveillance protocols must continue to adapt; the emergence of Omicron variants and shift to at-home testing as recent challenges. Rigorous and transparent methods to develop, deploy, and diagnosis adapted surveillance protocols will be critical to their success.

Methodological evaluation of bias in observational coronavirus disease 2019 studies on drug effectiveness.

BACKGROUND AND OBJECTIVE: Observational studies may provide valuable evidence on real-world causal effects of drug effectiveness in patients with coronavirus disease 2019 (COVID-19). As patients are usually observed from hospital admission to discharge and drug initiation starts during hospitalization, advanced statistical methods are needed to account for time-dependent drug exposure, confounding and competing events. Our objective is to evaluate the observational studies on the three common methodological pitfalls in time-to-event analyses: immortal time bias, confounding bias and competing risk bias. METHODS: We performed a systematic literature search on 23 October 2020, in the PubMed database to identify observational cohort studies that evaluated drug effectiveness in hospitalized patients with COVID-19. We included articles published in four journals: British Medical Journal, New England Journal of Medicine, Journal of the American Medical Association and The Lancet as well as their sub-journals. RESULTS: Overall, out of 255 articles screened, 11 observational cohort studies on treatment effectiveness with drug exposure-outcome associations were evaluated. All studies were susceptible to one or more types of bias in the primary study analysis. Eight studies had a time-dependent treatment. However, the hazard ratios were not adjusted for immortal time in the primary analysis. Even though confounders presented at baseline have been addressed in nine studies, time-varying confounding caused by time-varying treatment exposure and clinical variables was less recognized. Only one out of 11 studies addressed competing event bias by extending follow-up beyond patient discharge. CONCLUSIONS: In the observational cohort studies on drug effectiveness for treatment of COVID-19 published in four high-impact journals, the methodological biases were concerningly common. Appropriate statistical tools are essential to avoid misleading conclusions and to obtain a better understanding of potential treatment effects.

COVID-19-The impact of variable and "low normal" pulse oximetry scores on Oximetry@Home services and clinical pathways: Confounding variables?

COVID-19 Oximetry@Home services have been commissioned nationally. This allows higher-risk patients with mild COVID-19 symptoms to remain at home, being supplied with a Pulse Oximeter to measure their oxygen saturation (SpO2 ) two to three times daily for two weeks. Patients record their readings manually or electronically which are monitored by a clinical team. Clinical decisions, using an algorithm, are based on SpO2 readings in a narrow range with 1-2 point changes potentially affecting care. In this article, we discussed the problem that multiple factors affect SpO2 readings, and that some "normal" individuals will have "low-normal" scores at the threshold of clinical management, without any known respiratory problem. We discuss the potential magnitude of this problem based on the associated literature and consider how this will have an impact on the use of the Oximetry@home services, potentially partially confounding their purpose; to reduce face-to-face medical care.

History of heart failure in patients with coronavirus disease 2019: Insights from a French registry.

BACKGROUND: Although cardiovascular comorbidities seem to be strongly associated with worse outcomes in patients with coronavirus disease 2019 (COVID-19), data regarding patients with preexisting heart failure are limited. AIMS: To investigate the incidence, characteristics and clinical outcomes of patients with COVID-19 with a history of heart failure with preserved or reduced ejection fraction. METHODS: We performed an observational multicentre study including all patients hospitalized for COVID-19 across 24 centres in France from 26 February to 20 April 2020. The primary endpoint was a composite of in-hospital death or need for orotracheal intubation. RESULTS: Overall, 2809 patients (mean age 66.4±16.9years) were included. Three hundred and seventeen patients (11.2%) had a history of heart failure; among them, 49.2% had heart failure with reduced ejection fraction and 50.8% had heart failure with preserved ejection fraction. COVID-19 severity at admission, defined by a quick sequential organ failure assessment score>1, was similar in patients with versus without a history of heart failure. Before and after adjustment for age, male sex, cardiovascular comorbidities and quick sequential organ failure assessment score, history of heart failure was associated with the primary endpoint (hazard ratio [HR]: 1.41, 95% confidence interval [CI]: 1.06-1.90; P=0.02). This result seemed to be mainly driven by a history of heart failure with preserved ejection fraction (HR: 1.61, 95% CI: 1.13-2.27; P=0.01) rather than heart failure with reduced ejection fraction (HR: 1.19, 95% CI: 0.79-1.81; P=0.41). CONCLUSIONS: History of heart failure in patients with COVID-19 was associated with a higher risk of in-hospital death or orotracheal intubation. These findings suggest that patients with a history of heart failure, particularly heart failure with preserved ejection fraction, should be considered at high risk of clinical deterioration.

Syndemics at play: chronic kidney disease, diabetes and COVID-19 in Pakistan.

Although coronavirus disease 2019 (COVID-19) is a pandemic, it has several specificities influencing its outcomes due to the entwinement of several factors, which anthropologists have called "syndemics". Drawing upon Singer and Clair's syndemics model, I focus on synergistic interaction among chronic kidney disease (CKD), diabetes, and COVID-19 in Pakistan. I argue that over 36 million people in Pakistan are standing at a higher risk of contracting COVID-19, developing severe complications, and losing their lives. These two diseases, but several other socio-cultural, economic, and political factors contributing to structured vulnerabilities, would function as confounders. To deal with the critical effects of these syndemics the government needs appropriate policies and their implementation during the pandemic and post-pandemic. To eliminate or at least minimize various vulnerabilities, Pakistan needs drastic changes, especially to overcome (formal) illiteracy, unemployment, poverty, gender difference, and rural and urban difference.

Risk factors for positive and negative COVID-19 tests: a cautious and in-depth analysis of UK biobank data.

BACKGROUND: The recent COVID-19 outbreak has generated an unprecedented public health crisis, with millions of infections and hundreds of thousands of deaths worldwide. Using hospital-based or mortality data, several COVID-19 risk factors have been identified, but these may be confounded or biased. METHODS: Using SARS-CoV-2 infection test data (n = 4509 tests; 1325 positive) from Public Health England, linked to the UK Biobank study, we explored the contribution of demographic, social, health risk, medical and environmental factors to COVID-19 risk. We used multivariable and penalized logistic regression models for the risk of (i) being tested, (ii) testing positive/negative in the study population and, adopting a test negative design, (iii) the risk of testing positive within the tested population. RESULTS: In the fully adjusted model, variables independently associated with the risk of being tested for COVID-19 with odds ratio >1.05 were: male sex; Black ethnicity; social disadvantage (as measured by education, housing and income); occupation (healthcare worker, retired, unemployed); ever smoker; severely obese; comorbidities; and greater exposure to particulate matter (PM) 2.5 absorbance. Of these, only male sex, non-White ethnicity and lower educational attainment, and none of the comorbidities or health risk factors, were associated with testing positive among tested individuals. CONCLUSIONS: We adopted a careful and exhaustive approach within a large population-based cohort, which enabled us to triangulate evidence linking male sex, lower educational attainment and non-White ethnicity with the risk of COVID-19. The elucidation of the joint and independent effects of these factors is a high-priority area for further research to inform on the natural history of COVID-19.

Statistical power in COVID-19 case-control host genomic study design.

The identification of genetic variation that directly impacts infection susceptibility to SARS-CoV-2 and disease severity of COVID-19 is an important step towards risk stratification, personalized treatment plans, therapeutic, and vaccine development and deployment. Given the importance of study design in infectious disease genetic epidemiology, we use simulation and draw on current estimates of exposure, infectivity, and test accuracy of COVID-19 to demonstrate the feasibility of detecting host genetic factors associated with susceptibility and severity in published COVID-19 study designs. We demonstrate that limited phenotypic data and exposure/infection information in the early stages of the pandemic significantly impact the ability to detect most genetic variants with moderate effect sizes, especially when studying susceptibility to SARS-CoV-2 infection. Our insights can aid in the interpretation of genetic findings emerging in the literature and guide the design of future host genetic studies.

On the relationship between BCG coverage and national COVID-19 outcome: could 'heterologous' herd immunity explain why some countries are better off?

The COVID-19 pandemic has affected most parts of the global society since its emergence, and the scientific community has been challenged with questions urgently demanding answers. One of the early hypotheses on COVID-19 outcome was that some protection could be offered by the tuberculosis vaccine (BCG), and several clinical studies were initiated along with the emergence of numerous observational studies on the relationship between BCG and COVID-19 severity. In the present work, I demonstrate a strong correlation between the number of years that countries implemented BCG vaccination plans and age-standardized mortality rates during the first months of the pandemic in Europe. Further analyses of age groups in two European countries with comparably few confounding factors and easily identifiable groups of BCG-vaccinated and non-vaccinated subgroups suggest a population-level effect of BCG on national outcomes of COVID-19. This phenomenon of 'heterologous herd immunity' deserves further investigation, both in epidemiological and experimental studies.

Non-steroidal Anti-inflammatory Drugs and the Risk of Pneumonia Complications: A Systematic Review.

There have been concerns regarding the safety of nonsteroidal antiinflammatory drugs (NSAIDs) in patients with respiratory infections. However, to date, the quality of the evidence has not been systematically assessed. The purpose of this systematic review was to evaluate the role of NSAIDs on pneumonia complications. OVID MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Database of Abstracts of Reviews of Effects, and Google Scholar were searched. Studies that examined pneumonia complications in patients who had taken NSAIDs before onset of symptoms were identified. Quality assessment was conducted using the Risk of Bias in Non-randomized Studies - of Interventions (ROBINS-I) assessment tool, which was adapted to include biases that were pertinent to this question. The search strategy identified 1721 potential studies through the 5 primary databases and searching reference lists. Of these, 10 studies met the inclusion criteria, including 5 nested case-control studies, 2 population-based case-control studies, and 3 cohort studies. In total, 59,724 adults were included from 4 of the studies (range = 57-59,250) and 1217 children from 5 studies (range = 148-540). All studies demonstrated a positive association; in adults (odds ratio/risk ratio range = 1.8-8.1) and children (odds ratio/risk ratio range = 1.9-6.8). Studies were limited by moderate or serious risk of confounding bias, exposure misclassification, and protopathic biases and sparse data bias. The results of this review demonstrate that published studies on the effect of NSAIDs use and risk of pneumonia complications are subject to a number of biases. These results should not be extrapolated as evidence of harm for NSAIDs, including ibuprofen, in respiratory ailments but highlight the need for more methodologically robust studies to evaluate this potential relationship.

Increased Risk of COVID-19 Among Users of Proton Pump Inhibitors.

INTRODUCTION: Proton pump inhibitors (PPIs) increase the risk for enteric infections that is likely related to PPI-induced hypochlorhydria. Although the impact of acid suppression on severe acute respiratory syndrome coronavirus 2 is unknown thus far, previous data revealed that pH ≤3 impairs the infectivity of the similar severe acute respiratory syndrome coronavirus 1. Thus, we aimed to determine whether use of PPIs increases the odds for acquiring coronavirus disease 2019 (COVID-19) among community-dwelling Americans. METHODS: From May 3 to June 24, 2020, we performed an online survey described to participating adults as a "national health survey." A multivariable logistic regression was performed on reporting a positive COVID-19 test to adjust for a wide range of confounding factors and to calculate adjusted odds ratios (aORs) and 95% confidence intervals (CIs). RESULTS: Of 53,130 participants, 3,386 (6.4%) reported a positive COVID-19 test. In regression analysis, individuals using PPIs up to once daily (aOR 2.15; 95% CI, 1.90-2.44) or twice daily (aOR 3.67; 95% CI, 2.93-4.60) had significantly increased odds for reporting a positive COVID-19 test when compared with those not taking PPIs. Individuals taking histamine-2 receptor antagonists were not at elevated risk. DISCUSSION: We found evidence of an independent, dose-response relationship between the use of antisecretory medications and COVID-19 positivity; individuals taking PPIs twice daily have higher odds for reporting a positive test when compared with those using lower-dose PPIs up to once daily, and those taking the less potent histamine-2 receptor antagonists are not at increased risk. These findings emphasize good clinical practice that PPIs should only be used when indicated at the lowest effective dose, such as the approved once-daily label dosage of over-the-counter and prescription PPIs. Further studies examining the association between PPIs and COVID-19 are needed.