Since April, at least 74.9 thousand people died in coronavirus hospitals. How the regions hide these data: the investigation "Mediazons"

As Mediazona found out, from April to November in Russia, at least twice as many people died in Koronaviral Patients, which is taken into account in the official statistics of the Stopkoronavirus.rf site for the entire time of the pandemic. Regional and federal authorities have operational data on daily mortality in coronavirus hospitals and use it for internal calculations, but half of the regions publish numbers, underestimating them many times or even dozens of times.

The material was prepared with the participation of the Medusa editorial office .

Mediazona gained access to the data of the information center for monitoring the situation with the coronavirus (ICC) of the Ministry of Health.

The user of ITSK, who wished to remain anonymous, got in touch with the Mediazona. He presented the editorial office with evidence that he really has access to the system, and sent an archive with files downloaded from ICC. In the source archive there were several hundred files in PDF and XLSX formats with template names “Operational monitoring of the readiness of regional healthcare systems for hospitalization of patients with pneumonia” and dates with. In addition, the archive had files with the names “Consolidated information on the readiness of the bed fund”, but only for November 2020.

According to these data, from the beginning of the ICC system on March 31 to November 22, inclusive in Russia, 74,866 patients died for coronavirus patients. The Ministry of Health never published these figures, but the authorities used them to assess the situation in each region, and President Vladimir Putin instructed his envoy to follow this base and report to him weekly. Rospotrebnadzor leads publicly affordable statistics from the victims of coronavirus-and, according to his data, during the same period of victims is half as much: 36,516 people died from Covid-19 .

The data of the Ministry of Health, on the one hand, do not take into account those who died from the Covid-19 in non-specialized hospitals or were not hospitalized, on the other hand, take into account people who died in coronavirus bunks not from Covid-19.

Rosstat, who at the time of the release of the text published data on the victims of Covid-19 from April to September inclusive, estimates the number of victims of 55.6 thousand people; For the same period-April-September-ITSK contains data on the deaths of 41.3 thousand patients of coronavirus hospitals.

What is ICC and how did we check the data?

The Ministry of Health is responsible for the information center for monitoring the situation with coronavirus (ICC). The register was created so that different departments could collect information about the course of the epidemic in Russia in one place. On March 31, the head of government, Mikhail Mishustin, signed a decree on the creation of an information system, the functionality of which coincides with the functionality of the ICC, but there is no name of the system itself in the document. For the same date-March 31-the first reports on the situation with the Covid-19 in the country were loaded into the system.

For the first time, Mishustin mentioned the abbreviation of ICS at a meeting of Putin with the heads of regions about the situation with coronavirus on April 9. According to the results of this meeting, Putin instructed to use the ICC register to interact all departments that are involved in the fight against the epidemic. Then the president ordered his plenipotentiary representatives in the federal districts to monitor information from this registry and report to it weekly about the results. The ITSK data was used by Rospotrebnadzor for the phased quarantine removal in the regions at the end of the first wave of Covid-19 in Russia.

At the same time, the federal official who wished to remain anonymous in an interview with Medusa claimed that the ICC system operators are primarily interested in data on the availability of beds in hospitals, and the death of patients in this base is secondary. “[The database is not suitable for analysis of mortality data - it is the most operational, but also the most inaccurate,” he says.

In addition to data on the deaths of people with Covid-19, in the archive we received there really were data on the employment of beds and the number of people for Ivl in each of the regions. Data on patients for Ivl coincided with the data of the opera Staff of the Altai Territory, the Jewish Autonomous Region, Buryatia, Ingushetia, Ossetia, Tyvs; Data on employed beds-with publications of operastans of the Kabardino-Balkarian Republic, Buryatia, Ingushetia, the Republic of Crimea, the Republic of North Ossetia. In Moscow, the data on the Ivl coincided completely, the rest of the regions do not publish these indicators.

The all-Russian graphics shows that the data of Rospotrebnadzor about the dead from the Covid-19 never coincided with the number of dead in coronavirus bunks and have always been less than twice.

In November, when federal reports reported 400-500 deaths per day, 800-900 people actually died in coronavirus hospitals. The peak of mortality in the data available to the Mediazone came on November 17 - then 932 people died in hospitals: Rospotrebnadzor reported only about 456 dead.

The background on the mortality schedule we have imposed data on busy coronavirus bunks - these values ​​are inextricably linked. On all schedules, the values ​​on the right vertical scale are mortality, and on the left - to the number of beds.

The total number of those who died on the graph is less than in the heading: in the archive we received, data on the total number of victims were made in a separate line, but for infographics we calculated this indicator as the sum of those who died for each day separately.

Thus, data on deaths until March 31, as well as on September 14, did not get on the schedule - there was no file for this day in the archive. The figure for this day is averaged at neighboring values ​​and is not taken into account in the sums.

All mortality data in the regions according to the operatives are given on the site Stopkoronavirus.rf

After analyzing, as reported by each region reports separately, we divided them into four large groups, depending on how the data from the ICC that publish operational headquarters are connected or not related.

To exclude the influence of weekends and holidays, we used a centered sliding. This means that to build graphs, we do not use the number itself, which is reported by the operaste or ICC every day, but the average between this number and numbers for each of the three days before and three days after.

Employment of beds

The main object of analysis in the obtained "media" data set is the readiness of the medical system for taking new patients. Each region reports daily how many coronavirus beds were occupied by hospitalized. Unlike other indicators - infection and mortality - data on beds on a Russian scale have not yet been available.

Reporting the number of hospitalized, the leaders of hospitals and local Ministry of Health are unlikely to be trying to underestimate the numbers, since each new patient entails expenses that are covered from the federal budget. By this, the new indicator is very different from both infectionality data and mortality.

In many countries, the number of hospitalizations, the employment of beds and the availability of Ivl are public information. The authorities of Great Britain , the USA , Germany , the Czech Republic and other countries report this.

Regions in which the data is completely coincided

Astrakhan region, Volgograd region, Trans -Baikal Territory, Kamchatka Territory, Krasnoyarsk Territory, Kursk region, Magadan region, Moscow, Moscow region, Nizhny Novgorod region, Omsk region, Oryol region, Penza region, Republic of Adygea, Republic of Buryatia, Republic of Ingushetia, Republic of Kalmykia, Republic of Mari El, Rostov Region, Ryazan region, Sakhalinsky Region, Sakhalinsky Region, Sakhalinsky Region, Sakhalinsky Oblast, Sakhalinsk region, Sevastopol, Stavropol Territory, Tambov region, Tver region, Tula region, Tyumen region, Khabarovsk Territory, Chukotka Autonomous Okrug, Yaroslavl region

The operational headquarters of the 31 regions publish exactly those that are in the ICC system. The total amount of the dead coincides, and the form of the schedule throughout the epidemic.

This approach can be called the most transparent for the message of operational data: according to the methodological recommendations of the Ministry of Health, to clarify the final (pathologist) diagnosis, up to 45 days. Since the reports are made for every day, it is impossible to get final diagnoses for all dead in 24 hours. In such a situation, the message about the total number of dead in coronavirus bunks of day-in-day reflects the real operational picture of what is happening, which can then be clarified as diagnoses when transmitting data to Rosstat.

Separately, it is worth noting the situation in the Omsk region: until November 6, the data entered in the ICC coincided with the fact that the operastble reported. After November 6, a sharp increase in mortality, which was not reported publicly, can be seen in ITS.

Regions that underestimate statistics using the coefficient

Altai Territory, Voronezh region, Ivanovo region, Irkutsk region, Kabardino-Balkarian Republic, Kemerovo region, Lipetsk region, Novosibirsk region, Primorsky Territory, Republic of Dagestan, Komi Republic, the Republic of Crimea, the Republic of Khakassia, St. Petersburg, Smolensk Region, Ulyanovsk Region, Khamao, Chuvashia, Yamalo-Nenets Autonomous District District Autonomous District District

The operational headquarters of 19 regions daily publish data on deaths on coronavirus bunks from ICC, multiplying them by a reducing coefficient - that is, underestimating mortality indicators, for example, by ⅓ or on another meaning. As a result, the graphs are similar, but the total amount of deaths in the daily official reports of the opera station.

The authorities of St. Petersburg come in a similar way. During the first wave, they daily reported the number of the dead, which corresponds to data from the ICC, multiplied by ⅔. Since the beginning of July, they began to act differently and began to report daily the number of deaths, which is exactly 15 less than in the ICC.

Theoretically, the difference between the deaths on coronavirus bunks and data reported by the operastble can be explained by the features of counting coronavirus victims; However, this difference cannot be naturally the same daily for seven months.

What is happening is clearly visible on the graphs.

Fixed on December 1 at 21:34. Initially, this paragraph stated that Petersburg multiplies the number of those killed by a coefficient of 2/3. They specified their methodology.

Regions in which the data of the ICC are in no way related to the data of the operational headquarters

Amur Region, Arkhangelsk Region, Bryansk Region, Vladimir Region, Vologda Oblast, Jewish Autonomous Region, Kaliningrad Region, Kaluga Region, Karachay-Cherkess Republic, Kirov Region, Krasnodar Territory, Leningrad Region, Nenets Autonomous District, Novgorod Region, Orenburg Region, Perm Territory, Pskov Region, Republic of Republic, Republic, Republic, Republic, Republic of Republic. Altai, the Republic of Karelia, the Republic of Mordovia, the Republic of North Ossetia - Alania, the Republic of Tatarstan, Samara Region, Saratov Region, Sverdlovsk Region, Tomsk Region, Udmurtia, Chechen Republic

The operational headquarters of 29 regions publish numbers that are in no way related to data from the ICC: we have not found any pattern that would explain this data.

The authorities of the Chechen Republic underestimated the published data about the dead ten times. Starting in July, they practically stopped entering the ICC on deaths on coronavirus bunks. At the same time, a much larger number of those killed from Covid-19 in Chechnya is also confirmed by excess mortality: 1,691 people died here than usual-the increase in mortality was 50%.

The authorities of the Bryansk region have ceased to submit data about the dead in ICS, starting on May 25. It was in May that in the region a total mortality rate was recorded in the entire epidemic. At the same time, the operative continued to publish data on deaths.

Regions that publish data from ICC with a delay

Belgorod region, Kostroma region, Murmansk region, Yakutia, the Republic of Tuva, Chelyabinsk region

Six regions publish numbers comparable to data from ICC, but with a delay of several days. Despite the discrepancies in the schedules, the total amount of dead in beds for coronavirus patients coincides with the data of the operational headquarters.

Editor: Dmitry Treshchin

Text: Maxim Litavrin

Data analysis: Mikhail Zelensky, Dmitry Treshchin, Yegor Skovoroda

Infographics: David Frenkel