This Metadvice blog series explores the mysteries surrounding the complexities of SARS-CoV-2 infection on people and healthcare systems. We also explore how advanced analytics and technologies like artificial intelligence (AI), developed by companies like Metadvice can help navigate these complexities.

People infected with COVID-19 show a variety of symptoms which range in severity from nil, through mild, to life-threatening. The majority of symptomatic carriers regardless of symptom severity have a fever and sore throat which is in line with the disease being a respiratory disease and affecting the body through inflammation. However, many symptoms beyond the respiratory tract have been added to the list according to the World Health Organization (WHO). These include neurological (loss of smell and taste) and digestive tract symptoms along with cutaneous lesions. It is now clear that COVID-19 is a multisystem disease. The tropic properties of SARS-CoV2 are far broader than initially suspected.

Data also show that sex, age, weight, comorbidities, ethnicity, social deprivation, occupational exposure and housing occupancy affect susceptibility to and symptom severity of COVID-19. Blood type has also emerged as a factor determining relative risk of COVID-19 infection. The role of comorbidities will be considered in detail in a separate publication in this series, so we look here at ethnic groups, particularly ethnic minorities such as Black, Asian, and minority ethnic (BAME) who are at higher risk of displaying more severe symptoms and dying from COVID-19 according to data from both the UK and the US. So far, there is no clear physiological or genetic evidence that explains the predisposition of different ethnic groups to COVID-19 infection. Let’s consider some plausible hypotheses that are currently being tested.

1. First of all, systemic issues seem to largely explain the difference in infection and death rates experienced by ethnic minorities. These include higher rates of comorbidities such as diabetes, hypertension and cardiovascular disease, social deprivation, a lower likelihood to have access to health services, and a higher proportion of essential workers belonging to ethnic minorities. People from Asian and black groups are at markedly increased risk of in-hospital death from COVID-19. This excess risk, however, appears to be only partially attributable to pre-existing clinical conditions or deprivation so the question remains as to why certain individuals show milder or more severe symptoms, even within the same ethnic group. Therefore, further research into the drivers of this association is required. Genetic and lifestyle factors are currently being explored in well characterised datasets such as the UK Biobank.
2. What about children? Here, the picture has evolved considerably in the last few months. A study from China reported that children are as likely as adults to be infected but the majority will show mild or no symptoms. European data have confirmed this view, while noting that a small percentage of children at the severe end of the clinical spectrum may progress to the point of requiring mechanical ventilation. It is presently unclear whether children have a lower risk than adults of becoming infected after exposure or are less likely to transmit the virus to others. These considerations are presently being weighed with respect to reopening schools in some countries. In general, inflammatory markers are also less common in children except in the case of paediatric multi-system inflammatory syndrome which, although uncommon, can have severe clinical consequences. Of note, children under the age of five are usually more prone to catch viral infections as they lack the antibodies required to fight common infections such as the flu and the common cold. This is not the case with COVID-19. One possible explanation for this with the relationship of the virus with ACE2 (angiotensin converting enzyme-2) receptor. ACE2 has been identified as the functional receptor of SARS-CoV2. While the exact mechanisms remain unclear, ACE2 is produced in higher concentrations in children who also have constitutionally elevated lymphocyte counts.

COVID-19 is a novel coronavirus characterised by considerable heterogeneity and many unanticipated challenges. Multiple factors have been identified that may modulate individual susceptibility and which may influence the clinical course of the illness. Over the past decade, many developments have been made in applying AI to healthcare . These advances have the potential to usefully complement standard clinical research methods. Since human beings have a limited capacity to process information - especially in the context of challenging clinical scenarios - incorporating AI into healthcare by processing a multitude of information, from the quantity and severity of symptoms to age, gender, ethnicity, social background or comorbidities, offers the prospect of aiding physicians, healthcare personnel and pharmaceutical companies to make better and more informed decisions.

Metadvice aims to help highlight the challenges posed by COVID-19 by bringing together our AI expertise with clinical insights to provide actionable insights. Metadvice is focused on complex multi-morbid conditions, several of which were identified at an early stage of the pandemic as risk modulators in the context of COVID-19. We believe that Metadvice risk and therapy knowledge bases could be of value during the exit strategy from the current wave of infection. By determining risk associated with specific sub-groups of individuals not yet infected we can help map risk to reduce infection rates. Metadvice could also provide guidance for stratification of vaccination and anti-viral pharmacotherapeutics priorities as these become available.