In addition to its immediate impact, the Covid-19 pandemic will in all likelihood have indirect, medium-to-long-term effects on mortality and morbidity. Some of the individuals who develop severe forms of the disease are suffering from after-effects that could reduce their life expectancy. But the effects of the pandemic extend well beyond those who have been infected. Healthcare systems have become saturated, generating delays in diagnosis and late detection of serious illnesses. The restrictive measures instituted to manage the crisis have also affected some people psychologically. These indirect effects are expected to be at least partially offset, however. The victims of the epidemic have been first and foremost vulnerable individuals whose life expectancy was already low. At the same time, the epidemic has accelerated far-reaching technological advances. Insurers and reinsurers must take these various factors into account when assessing the risk of death.

Since March 2020, all serious thinking about Covid-19 has been prefaced by the proviso “as of this writing”. This qualification shows how quickly the epidemiological situation is changing, but also indicates that our knowledge is increasing with each passing day.

As of end-October 2021, more than 240 million people worldwide had been infected by Covid-19, and nearly 5 million people had died from it, including 120,000 in France. These figures seem grossly under-estimated when compared with the excess mortality observed since the start of the pandemic, and with average mortality observed in previous years. According to The Economist magazine, the real number of Covid-19 deaths is now probably between 10.2 and 19.3 million.

For insurance companies, the pandemic has impacted more than just mortality risk. It has also led to work stoppages, business losses, disability, long-term care needs and dread diseases.

Nevertheless, we will restrict our analysis here to mortality risk alone. Our article is based on studies of entire populations. In the United States, for example, insured individuals are on average wealthier and have better access to healthcare than the rest of the population, and they have been much less impacted by the pandemic. 

In addition to the immediate effects of Covid-19 on mortality, the epidemic is expected to affect longer-term mortality, directly via potential after-effects or indirectly as a result of lockdowns and other measures. By indirect effects, we mean the effects on persons who have not been infected by the disease.

We intend to paint a broad picture of the medium- and long-term effects of the pandemic on mortality. We have identified five such effects:

• After-effects of Covid-19 – many people who have suffered from severe forms of Covid-19 risk experiencing a diminished life expectancy compared to what they would have enjoyed had they not been infected.
• Delayed diagnosis and treatment – lockdowns have been imposed and healthcare systems reorganized to free up resources to treat Covid-19 patients. As a result, diagnoses have been delayed or treatment has been interrupted,  which could have a significant impact on mortality.
• Psychological effects – the various lockdowns and uncertainty related to the epidemiological and economic situation have caused an increase in psychological disorders, including some serious ones.
• Individual behavior – the various lockdown and social distancing measures have contributed, for some people, to reduced physical activity and less-healthy eating habits.
• Economic impact – we have learned from the numerous past economic crises that poor economic conditions can have a significant impact on access to healthcare and, by extension, on mortality.

We strongly believe that only by understanding these various factors, and through the risk modeling that goes along with them, can we continue, as a reinsurer, to contribute to the resilience of our societies against Covid-19. We will continue to share our knowledge by publishing similar articles throughout the pandemic (1).

What determines the long-term impact of Covid-19?

After-effects

Clinical data shows [Lopez-Leon et al., 2021] that a significant percentage of individuals infected with SRAS-CoV-2 present symptoms over a period of several months that could have negative effects on long-term mortality trends:

• fatigue;
• pulmonary damage;
• heart damage;
• neurological damage.

The after-effects described above relate first and foremost to individuals who have developed a moderate or severe form of the disease. Most of the studies on the after-effects of Covid-19 have been carried out on cohorts of these types of patients. Very little data is available on the potential after-effects developed by asymptomatic individuals or those with benign symptoms. 

Numerous studies have reported cases of “long Covid” in addition to after-effects, even in young people without underlying chronic diseases. The term “long Covid” includes all individuals who have had Covid-19 and present symptoms that last over time or who have taken longer than average to recover.

Because of the after-effects and long forms of the disease, it seems that Covid-19 changes the long-term health status of certain infected individuals. This deterioration in health is likely to lead to a more-or-less significant decrease in the life expectancy of these individuals. 

That said, there could be a strong bias built into the studies, because the economic difficulties suffered as a result of the epidemic can also contribute to the development of neuro- psychological symptoms. In addition, many studies suggest that fewer and fewer patients risk being stricken with a serious form of the disease and therefore run less risk of developing after-effects. There are at least two reasons for this:

• Treatment has improved considerably since the beginning of the pandemic. New therapies could significantly reduce the risk of serious cases and hospitalization;
• Vaccination has helped considerably to reduce the appearance of serious forms of the disease, as demonstrated by the Covid-19 death rate in the countries with the highest vaccination rates. In France, for example, the death rate fell from 3.7% of all cases in mid-December 2020 to around 0.7% as of end-October 2021. The impact of vaccines on after-effects and long forms of Covid-19 remains uncertain, however.

Delayed diagnosis and treatment

During the various waves of the epidemic, healthcare systems have focused largely on the treatment of Covid-19 patients, forcing healthcare personnel to postpone certain operations or medical consultations. Patients have been asked to postpone non-critical appointments to the extent possible. This was particularly true during the first wave. So it is likely that certain patients with serious diseases requiring specific treatment and regular follow-up have not gone to the hospital or to their doctor’s office since the onset of the pandemic, out of fear of being infected. Similarly, new cases of serious illness might not have been identified.

These factors might have a significant impact on long-term mortality in general and cancer mortality in particular. In the United States, there was a severe shortage of cancer diagnostics and treatment services during the first six months of the pandemic. 21st Century Oncology, a company specialized in the treatment of cancer, with 300 sites around the country, has reported that around 18% of its patients newly diagnosed with breast cancer between January and August 2020 were in an advanced stage of the disease, vs. 12% in 2019. Many cancers advance rapidly, so a number of months without detection could translate into fewer treatment options and a greater risk of death. A UK study [McPhail et al. 2015] has demonstrated that a delayed start to treatment has an effect on the increase in ten-year mortality, in particular for cancers diagnosed at an early stage. In France, the Institut Gustave Roussy [Bardet et al., 2020] has estimated that its oncology services will experience excess mortality of 2-5% between 2021 and 2025 as a result of delayed diagnoses between March and June 2020. The results observed in other pathologies are similar to those in oncology. In cardiology, the number of catheterizations has declined significantly since the start of the pandemic. As a result, the profile of patients who come for treatment has, on average, deteriorated. As these patients have postponed their medical visits, more of them now have multivessel coronary disease and need either an aorto-coronary bypass or a stent. Patients with certain other pathologies, however, have not experienced significant deterioration on average. Diabetes, for example, seems on average to have been controlled to a similar level before and during the pandemic [Patel et al., 2021].

Lastly, medical visits are still lagging behind their pre-pandemic levels and did so even during periods when the rate of Covid-19 incidence was very low. The main reason seems to be that some people are afraid to go to a location where they might come in contact with individuals infected by Covid-19.

Psychological effects

Anxiety and depression are two recurrent after-effects of Covid-19. Through social distancing and lockdown, the epidemic has given rise to feelings of uncertainty and has weighed negatively on mental health. Although mental health problems do not necessarily lead to excess mortality, they can increase risky behavior through less-healthy lifestyles and can also increase the number of suicides.

Studies carried out in the wake of the 2003 SRAS epidemic and the 2014 Ebola outbreak [Hawryluck et al., 2004; Preti et al., 2020] evaluated the psychological impact of quarantine and pandemic-type situations. The authors recorded anger, anxiety, boredom, confusion, fear, depression, emotional exhaustion, frustration, irritability and stress in some of the individuals they interviewed. Their work highlighted certain behaviors individuals developed, such as avoiding public places or crowds, social detachment, as well as the impact of the pandemic on alcohol consumption, death by overdose and suicide.

The SRAS studies showed that moderate-to-severe symptoms of anxiety, depression and post-traumatic stress disorders lasted two-and-a-half to three years in 20-30% of individuals who developed the disorders. The risk of persistent symptoms and psychiatric disorders was particularly high among healthcare workers.

Individual behavior

The pandemic also generated various types of eating disorders and changes in physical activity that could have a negative long-term impact on health and mortality. Zachary et al. [2020] studied the impact of lockdowns on body weight. Twenty-two percent of individuals reported an increase of 2-5 kg (4.4-11 lbs.) in body weight since the start of the pandemic (data as of May 2020). During lockdown periods, not only have people eaten more, but they have also had a tendency to eat less-healthy foods, especially those who were already overweight. Poelman et al. [2021] highlighted that obese or overweight individuals reported eating fewer healthy foods during lockdown than did individuals with a normal body mass index. Excess weight, obesity and stress being significant risk factors for hypertension, cardiovascular diseases, diabetes, metabolic syndrome and cancer, these pandemic-linked behavioral factors could have a lasting impact on mortality.

The economic impact

The Covid-19 pandemic and the lockdown measures implemented as a result of it initially caused a sharp recession. Numerous examples from past crises have shown that a deterioration in economic conditions can have a deleterious effect on access to healthcare, which can in turn potentially increase mortality over the long term.

^{Source: Inserm-Centre d'épidémiologie sur les causes médicales de décès.}

During economic crises, certain fragile healthcare systems can suffer a reduction in public or private financial support. A study on the treatment of diabetes in Greece showed the nefarious effects of the 2010 financial crisis in terms of poor nutrition, chronic stress and reduced adherence to medication. 

Recourse to laboratory testing and medical imaging declined. The monitoring of vascular complications also worsened, in particular for poorly controlled cases of diabetes [Frangos et al., 2012; Aloumanis and Papanas, 2014]. Other studies have confirmed the negative impact of a weakened economy on cancer mortality [Maruthappu et al., 2016]. Specifically, data from the World Health Organization and the World Bank on more than two billion people between 1990 and 2010 showed that an increase in unemployment was significantly tied to an increase in cancer mortality. An analysis of the data by type of cancer showed that the rise in unemployment increased the specific mortality of each cancer, with the exception of lung cancer among women. The study highlighted an excess of around 260,000 cancer deaths in OECD countries due to the economic crisis between 2008 and 2010. Using the same data, the authors demonstrated that universal healthcare and public spending on healthcare offers protection against an excess of cancer deaths. The same principle applies to patients with health insurance, who would therefore be less impacted by an economic crisis linked  to Covid-19. In the United States, for example, where insurance is generally provided by the employer, jobless people can be less well protected than their employed counterparts. Data gathered in the U.S. between 1997 and 2010 shows that an increase in the unemployment rate of one percentage point led to an increase in the suicide rate of 1-1.3% [Luo et al., 2011].

The negative effects of the pandemic on long-term mortality have been partially offset

While the Covid-19 pandemic is expected to have an overall negative long-term impact, certain factors should contribute to reducing mortality deriving from other causes over the coming years.

Firstly, mortality related to influenza and pneumonia has declined since the beginning of the pandemic. In France, infectious and parasitic diseases (a category that includes influenza and pneumonia) were responsible for around 10,000 deaths in 2016, the most recent year for which data is available from the center for epidemiology and medical causes of death at the French national institute for health and medical research (Centre d'épidémiologie sur les causes médicales de décès de l’Inserm). Lockdowns and safety protocols have been the principal reasons for this decline. These behaviors and the positive effects they have had on mortality might last over time (or at least, as long as the pandemic continues).

In addition, a significant percentage of the individuals who have died from Covid-19 were elderly and had comorbidities. A significant number of these people would have died anyway over the next few years, even had they not developed a severe form of Covid-19. This so-called “harvesting effect”, less grimly known as a “mortality shift”, must be taken into account when estimating the medium-to-long-term impact of the pandemic on mortality. Peaks in mortality among the elderly during heatwaves, for example, have often been documented. Elderly people have a diminished thermoregulatory capacity, and there is a greater likelihood that they live alone, are physically inactive, have chronic illnesses and take medication. Heatwaves provoke an increase in sudden deaths among the elderly in the first few days, but mortality is then lower than expected in the following months [Guo et al., 2011;  Klenk et al., 2010; Izraelewicz, 2012; Toulemon and Barbieri, 2008; Basu and Malig, 2011].

Catastrophes such as wars and epidemics often lead to an acceleration in scientific progress. The Covid-19 epidemic has been no exception. Technological progress has stepped up the development of monoclonal antibodies and messenger RNA vaccines, for example, which are very promising for the prevention of diseases such as cancer [Liu et al., 2018; Verbeke et al., 2019] or autoimmune diseases such as multiple sclerosis [Krienke et al., 2021].

Finally, with regard to insurance companies, insured individuals have been less affected by the pandemic than the population at large, because they have been more able to protect themselves from it and have experienced less loss of access to healthcare and fewer delayed diagnoses.

Conclusion

Looking beyond the horrific daily Covid-19 death count, there will be many deaths either directly or indirectly linked to the pandemic over the next few years regardless of how the pandemic evolves.
 
Five principal factors have been identified: Covid-19 after-effects, delayed diagnosis and treatment, psychological effects, changes in individual behavior and economic recession. These excess deaths will be only very partially offset by the harvesting effect, by pandemic-induced medical progress or by a reduction in deaths owing to influenza or pneumonia. Many unknowns remain as to the development of the pandemic, in particular with the potential advent of new variants that could bypass immunity already acquired through infection or vaccination. This makes it difficult to quantify the long-term impact of Covid-19 on mortality. But identifying the principal factors contributing to changes in mortality is a first step that will enable insurers to make the assumptions necessary to assess the risk.
 

Acknowledgements: The authors would like to thank the medical community of SCOR Global Life for their advice and recommendations in the preparation of this article, particularly Gabriela Buffet.