Anticipating the Total Mortality Impact of Coronavirus in Israel
Author: Alex Weinreb, Dov Chernichovsky
March 29, 2020
This study present rough estimates of the number of coronavirus fatalities under a number of scenarios, drawing on estimated case-fatality rates (CFRs) associated with the coronavirus in China and Italy. The authors highlight the risk of a significant increase in Israel’s overall mortality due to the increased diversion of medical resources – which are limited to begin with – to deal with the virus.
There is a great deal of uncertainty surrounding both the levels of coronavirus infection and its case fatality rate (that is, the number of people infected with coronavirus who die). In addition to the direct mortality from the virus, it will almost certainly also have indirect “collateral” mortality effects, due to the reallocation of medical resources, which are limited to begin with, to address the outbreak. Despite their inherent uncertainty, mortality estimates are an important tool for policymaking.
A new Taub Center study, conducted by demographer Prof. Alex Weinreb and health economist Prof. Dov Chernichovsky, lays out the direct and indirect effects of mortality from the coronavirus pandemic in Israel. The assessment includes a number of possible scenarios based on the virus mortality rates in China and Italy and mortality rates for from other causes in Israel and the EU.
Israel’s relatively young age structure means that the mortality rate from the virus is expected to be significantly lower than the rate in Italy or the Hubei Province in China
Mortality from the coronavirus is low for young people, and increases with age. Because some of those diagnosed with the virus do not show any symptoms, estimates based on the number of deaths among confirmed cases overstate mortality. On the other hand, the fact that there is still not long-term tracking of infected people might result in an underestimation of mortality rates. In Italy, where a large share of the population is elderly, the virus mortality rate up to age 70 is lower than the rate in China but, above age 70, is 15% higher.
Therefore, the crude mortality rate in Italy is more than twice the rate in China’s Hubei Province – 5.8% compared to 2.4%. In general, Israel’s young age structure (meaning a large percentage of the population is under 65, and a much larger share than in China or Italy is under 30) indicates that the mortality rate from the disease will be lower in Israel than in China and Italy.
A series of projections apply the age-specific mortality data from China to the Israeli population in order to forecast the number of coronavirus deaths in Israel under a number of different scenarios. The range of scenarios cover:
- Infection rates in the population as a whole of 0.1%, 0.5%, 1%, 2%, 5%, 10%, 20%, and 30%.
- Coronavirus mortality rates that range from the same as the adjusted rates in Hubei Province to 50% and 75% reductions in those rates. These reductions reflect the anticipated improvements in dealing with the virus as time passes and as the basic ability of the Israeli medical system to cope with emergencies increases.
The scenarios posit infection rates from coronavirus ranging from a minimum of 9,300 infected people (0.1%) to 2.79 million (30%). The Taub Center study presents three mortality scenarios according to the different infection rates: at an infection rate of 0.1% with the same mortality rate as in Hubei, Israel could expect to have zero deaths among those ages 20 and younger and up to about 100 deaths among those ages 70 and older, resulting in approximately 150 virus-related deaths total across all age-groups.
The higher the infection rate, the higher the expected number of deaths: at the infection rate of 10% at the Hubei mortality rate, the number of deaths is expected to reach up to 15,400 Israelis, and at an infection rate of 30%, up to 46,000.
In each scenario, significant medical successes reducing the mortality rate to a quarter of the rate observed in Hubei Province would cut the number of deaths accordingly. But it would still result in 39 deaths at a 1% infection rate and 11,000 deaths at a 30% infection rate.
By comparison, in 2016 there were 43,964 deaths from other causes in the country. About 11,000 of these deaths were from cancer and 6,800 from heart disease – the two most common causes of death in Israel. In the worst-case scenario (infection rates about 8% with Hubei mortality rates), coronavirus could become the leading cause of death in Israel.
However, Taub Center researchers Weinreb and Chernichovsky warn of the indirect effects that may result from the fight against the virus: “The coronavirus could potentially affect other types of mortality. We expect mortality rates from other causes to rise as hospitalization of coronavirus patients increases. The Israeli hospital system is deficient to begin with, as shown by another Taub Center study published a few months ago.”
Worst case scenario – coronavirus will itself become the leading cause of death in Israel, but it will also increase deaths from other causes: this will be the indirect mortality effect of the virus
In addition to the direct effect of coronavirus on mortality rates, the indirect effects on mortality rates from other causes must also be taken into account. This is due to the poor state of Israel’s hospital system. In Israel, there are 2.2 beds for general care per 1,000 residents (compared to 3.6 in the OECD and 4.1 in countries with a similar health system to Israel). The hospital bed occupancy rate is 94% (compared to 75% in the OECD), and the average length of stay in hospitals is relatively short – 5.2 days per patient (compared with 6.7 in the OECD).
The number of visits to hospital emergency rooms in Israel is twice that of countries with similar healthcare systems. In addition, even before the outbreak of the pandemic, Israel’s general hospitals were operating at full capacity, with almost no emergency reserves, and no alternatives to hospitalization available in the community or nursing homes.
Another reason for the expansion of indirect mortality is damage the virus has inflicted on Israel’s medical personnel, causing a reduction in available medical staff (about 3,000 had been sent into quarantine by March 22, 2020), which was limited to begin with due to a lack of positions. The current situation inevitably leads to medical resources being diverted to cope with the coronavirus and is expected to come at the expense of other life-saving treatments such as heart catheterization and cancer detection.
In Italy, the burden on the hospital system has resulted in the indirect mortality of “normal” patients whose treatments have been disregarded due to the virus, and in England millions of operations have been postponed due to increasing pressure on the health system.
The authors estimate that in Israel, a moderate 2% increase in mortality rates from other causes would increase the overall number of deaths over a six-month period by about 460. At that time, a 20% increase in mortality from causes other than the coronavirus would result in an increase of about 4,600 deaths.
Both these estimates assume that the pandemic will end within six months. If it were to continue beyond that, the number of deaths from other causes would be higher yet.
The sharp increase expected in the number of deaths alongside the possible increase in infection rates bolster the Israeli government’s social distancing and economic shutdown policies – policies that should slow the spread of the virus and “flatten the curve.”
Slowing down the rate of transmission will make things easier for the hospital system’s weak infrastructure and provide the system with time to add hospital beds and other essential equipment, and for its staff to hone their clinical skillset in relation to this virus.
However, similar efforts should also be invested in protecting against mortality from other causes. Minimizing mortality from coronavirus at the cost of higher levels of indirect mortality should not be used to evade responsibility for long-term neglect of parts of the Israeli medical system.