Deadliest outside the big cities Though the majority of Russia's coronavirus cases have been recorded in Moscow, demographic and epidemiological data suggest more remote areas are most vulnerable. We mapped it.

Based on data available on March 28, most of Russia’s confirmed COVID-19 cases were in Moscow, the Moscow region, and St. Petersburg. This is unsurprising: national capitals everywhere have taken the first blow in this pandemic because they’re the most connected to international travel. The media’s focus on megapolises can be misleading, however, insofar as it creates the impression that coronavirus has somehow spared more remote areas. That is not the case. Given the age demographics of many regions in Russia, it is outside Moscow and St. Petersburg where we can expect to see the most difficult situations, both in terms of the number of patients needing assistance and the number of deaths. At Meduza’s request, demographer Ilya Kashnitsky and epidemiologist Anton Barchuk compared Russia’s regions by the number of people who could need hospitalization and the amount of locally available medical resources. The two experts created an interactive map that helps doctors in Russia assess the seriousness of the COVID-19 crisis in different parts of the country and makes it clear how important it is to delay the epidemic’s peak by as long as possible (to “flatten the curve”).

Attention! For the best reading experience, this article should be viewed from a full Internet browser. The maps embedded below will not display correctly on mobile devices. Also, please note that all Meduza content about the coronavirus pandemic is free to distribute according to Creative Commons CC BY 4.0. You can republish this stuff! (But that doesn't apply to any photographs featured in our stories.)

With COVID-19, the disease caused by the new coronavirus, the risks of complications, hospitalization, need for intensive care, and death are distributed extremely unevenly by age. Children and younger people are nearly out of danger, but the risks are significant for older generations. Additionally, at any age, the disease is more dangerous for men than women. The figure below demonstrates COVID-19’s lethality in Italy as of late March.

Populations’ age structures vary widely between different countries and between different regions within the same country. Given COVID-19’s uneven lethality indexes for age and sex, the logical conclusion, other things being equal, is that severe cases will be greater in relatively more elderly populations. In part, we see the effect of a population’s age structure in the pandemic’s first comparative data by country: the number and proportion of severe cases of the illness are much higher in Italy than in places like South Korea, where the population is relatively younger. We can observe even more significant differences in populations’ age structure between central and peripheral regions within specific European countries. The epidemic naturally begins in the most densely populated major cities. Once the disease is widespread, however, peripheral regions (where populations are generally older) are the most vulnerable. 

On the interactive map below, you can see the total number of people in different regions across Russia who could need intensive care at some point during the coronavirus epidemic. The darker the shading, the greater the proportion of severe COVID-19 cases in a region’s population. Below the map, you can find the same information laid out in a table.

In the event of an uncontrolled or poorly controlled outbreak, the number of seriously ill patients very quickly exceeds the number of available hospital beds. This is especially true for space in intensive care units equipped with ventilators. In this scenario, more remote regions are again the most vulnerable. 

The map below demonstrates a situation where 40 percent of Russia’s population contracts COVID-19 and the epidemic’s peak arrives after five, 10, or 20 weeks. In reality, this timeframe will depend on the effectiveness of Russia’s containment measures. The map displays the predicted number of people in intensive care in each region of the country per available ventilator.

How we counted (methodological details for specialists)

In our calculations, we used available data on Russia’s current sex and age demographics. The proportion of people expected to need intensive care corresponds to projections adopted in a report published on March 19, 2020, by the Imperial College COVID-19 Response Team, titled “Impact of Non-Pharmaceutical Interventions (NPIs) to Reduce COVID-19 Mortality and Healthcare Demand.” These data, in turn, were obtained by modeling and analyzing statistics from China. The proportions were adjusted for the differences in COVID-19’s fatality risks for men and women observed in Italy.

It is very difficult now to estimate the proportion of people who may be infected during the current epidemic, so we modeled multiple scenarios, assuming that the disease will spread to between 40 and 80 percent of Russia’s population. The actual value here will depend on the Rₒ coefficient, which reflects the number of people in the vulnerable population who are infected on average by one carrier of the disease. Our calculations did not take into account differences in the proportion of infected people between different age groups. In reality, the proportion of infected people in Russia could be lower in older age groups because older people generally have fewer social contacts.

To estimate the number of available ventilators in Russia, we used data obtained from public procurement contracts issued between 2017 and 2020 and public contracts for the comprehensive maintenance of medical equipment in 2019 and 2020. To assess the presence of this machinery at facilities where we could not find published contract records, we looked at the distribution of medical apparatuses by device-age/depreciation, the size of surgical departments and intensive care units, and a comparative analysis of the availability of other types of intensive-care equipment. All data about ventilators were provided by the Moscow-based “Headway Company,” and we also used some open-source information.

To calculate the epidemic’s peak surge, we relied on a spreadsheet-based model provided by the Centers for Disease Control and Prevention in the United States, which estimates the surge in demand for hospital-based services during the next influenza pandemic. We modeled three scenarios for the spread of COVID-19, where the peak occurs at the fifth, 10th, and 20th week after the outbreak’s start, and assumed that the average ICU stay is five days. 

In the event of a rapid and effective implementation of measures to contain the epidemic, the peak surge in demand for ICU space will be significantly lower than in the absence of such measures. According to our calculations, should the epidemic peak as soon as its fifth week, the number of patients requiring intensive care will be more than the number of ventilators in most regions across Russia. If Russia is able to “flatten” its epidemiological curve, delaying the COVID-19 peak to the epidemic’s 20th week, there will be sufficient medical resources to treat patients in almost all regions nationwide.

The peak surge per individual ventilator is largely a provisional value — it should be understood foremost as a way to compare regions, taking into account their medical resources and expected numbers of severe COVID-19 cases. The actual peak surge in different regions will depend on many parameters that are very hard to assess now, such as the number of infections, the epidemic’s length, and the timing of the peak surge — all of which are determined by the extent and effectiveness of containment measures.

In real-world conditions, the peak surge will also depend greatly on the use of available medical equipment and resources. Not all ventilators are located at facilities that are likely to get many severe cases of COVID-19. On the other hand, some ventilators can work on multiple patients simultaneously. 

The number of COVID-19 patients we expect to need intensive care in Russia coincides largely with a forecast by Sofia Garushyants and Georgy Bazykin, whose methodology is similar to ours. The differences between our forecasts are due primarily to the fact that our study takes into account the proportion of men and women in Russia’s population. Because the number of more elderly men in most regions is far lower than older women, the number of COVID-19 patients requiring intensive care in these areas will also be lower.

All the source data and code used to make the calculations found above are available here on GitHub.