The Danger of Underestimating Delayed Onset Diseases

This chart illustrates the impact of delaying preventive measures on the incidence of a hypothetical syndrome (such as cancer or heart disease) triggered by a recurring infectious agent (IA). By adjusting the variables, you can observe how waiting for visible signs of delayed harm from infection significantly increases the future burden of disease. The red shaded area represents the cases that are detected and measured before interventions are taken. The blue shaded area shows the additional burden of disease that becomes inevitable due to the delayed response.

As demonstrated, delaying action until enough people are visibly affected guarantees a significantly larger disease burden. This highlights the urgency of taking preventive measures now to mitigate future harm.

Understanding Latent Health Risks of Infectious Diseases

Viruses, parasites, and bacteria are all capable of causing acute illnesses. This is nothing new from a scientific or personal standpoint. We are all familiar with colds and food poisoning that feel awful for a few days then get better. However, how we feel during the acute phase of a disease doesn’t necessarily reflect its potential long-term impact. Many diseases feel mild or are asymptomatic at first, but can take decades to progress into something much worse.

Rabies can take longer than a year to show symptoms. HIV is initially contracted as an asymptomatic or mild flu-like illness. An infected individual may continue to look and feel fine for up to 10 years until AIDS fully develops. Chagas disease, acquired from the parasite in the kissing bug, may present chronic phase symptoms such as an irregular heartbeat, stomach pain, and heart failure 10 to 20 years after initial infection. Similarly, syphilis can remain dormant for years before manifesting neurological symptoms that can result in death.

If someone said they had a mild case of HIV, rabies, or Chagas disease, most people would be alarmed. Hindsight shows us the danger of these diseases, but what about before we understood their long-term outcomes? The beginning of the HIV epidemic was characterized by doubt and stigma.

A Hypothetical Disease with Guaranteed Delayed Onset

Now, imagine a new disease causing asymptomatic to mild initial infections but guaranteeing a heart attack in exactly 10 years. Initially, it would seem like just a common cold. We would probably not act to mitigate this disease until the 10 year mark comes around. However, it will be too late for those who were infected before mitigations were put in place. A hope is that we could detect subtle signs of internal damage or pathogen persistence through various studies, such as in vitro, pathology, or even in silico studies. Usually we would prefer data with more clinical significance, but waiting for that data to come out is too dangerous, as we would be the data.

The Long-Term Threat of COVID-19

SARS-CoV-2 began spreading across the globe at the beginning of 2020. 4 years feels like a long time, but is well before the time it takes for many possible latent health issues to arise after an infection. We are already seeing that Covid-19 can cause significant long-term damage to nearly every organ system (symptomatically and asymptomatically), and may even be capable of viral persistence.

Heart disease, especially for younger populations, has been on the rise since the beginning of the pandemic. The mechanisms behind this are well established and are known to occur after mild, and even asymptomatic, cases.

The brain is another organ showing numerous mechanisms of adverse effects from mild and asymptomatic SARS-CoV-2 infection. One such example is the development of Lewy bodies in 8 out of 8 infected rhesus macaques. Lewy Body Dementia is a well established condition that takes years to develop. It can be difficult to find neurological sequelae because it is easy for patients and physicians to misinterpret symptoms as mental health issues. Consider the case study of two sisters who both reported severe neurocognitive problems after infection with SARS-CoV-2. Their symptoms were initially classified as “psychologic pandemic distress”, but they were later found to have profound brain hypometabolism.

A third long-term concern is cancer. SARS-CoV-2 can affect many mechanisms crucial to cancer onset, including cell cycle regulation and inflammation/proliferation signaling pathways. These mechanistic studies do not claim that cancer will happen, but they highlight the potential risk. How do we intend to find out?

The Urgency of Early Intervention

In conclusion, waiting for observational data to confirm the long-term impacts of COVID-19 is too dangerous. Once we have visible data proving that COVID-19 leads to delayed-onset diseases, it may be too late for many. We must act quickly to prevent transmission and mitigate the potential fallout.

The chart below demonstrates various strategies to prevent the transmission of SARS-CoV-2 and the potential consequences of delayed action.

Methodology

This model forecasts how the incidence and prevalence of a hypothetical syndrome could increase over time due to yearly exposure to a hypothetical infectious agent (IA), which imposes varying levels of risk of developing the aforementioned syndrome, normally distributed around a configurable delayed onset time.

For the purpose of illustration, a default of 18M was chosen for the baseline incidence of the syndrome based on data from the World Cancer Research Fund International. All individuals within the population - everyone on Earth in this case - are assumed to be infected with the IA exactly once per year, with all infections binned to the last day of the year. The extra cases of the syndrome due to the infections that occured within a given year are calculated based on the formula baseline_incidence X (1 - hazard_ratio). This total sum is normally distributed around a date in the future determined by the configurable year delay input. Each of these latency distributions is summed with each other to produce the aggregate curve of increased incidence of the syndrome.

The model has two different methods for halting the creation of latency distributions that simulate different reasons of enacting complete eradication of the IA. The first method is based on a configurable panic threshold in which the population stops transmission of the IA based on a response to the heightened level of the syndrome. The second method is simpler and allows for a date to be selected for when instantaneous eradication of the IA occurs.