The bubble of 14-day cumulative Covid-19 cases: how to inflate or deflate it?

Since last year life in Latvia has been dependent on a number reported daily by epidemiologists - the 14-day cumulative morbidity rate per 100,000 inhabitants. In the summer, travel opportunities depended on this number. Now - the possibility to go to the store, to school, to work. The traffic light principle developed by the government is also based on this number, i.e., until this number is reduced to 200, don't expect any easing of restrictions.

Last summer, when the 14-day cumulative rate of Covid-19 cases was used to determine which countries are epidemiologically safe to travel to, this rate in Latvia was one of the lowest in the European Union and even the lowest for one week in August at 3.6 per 100 000 inhabitants. Right now it sounds almost unbelievable because yesterday the figure exceeded 400.

Government's traffic light principle

At the beginning of February, the Cabinet of Ministers supported the plan developed by the Ministry of Health to reduce the security measures of Covid-19 according to the so-called traffic light principle. It envisages that the first opportunity to decide on a gradual reduction of restrictions will be only when a total of no more than 200 new cases per 100,000 population are registered over 14 days.

The Traffic Light Principle developed by the government consists of four levels of risk:

* dark red light - more than 200 new cases of Covid-19 per 100,000 population registered in 14 days;

* red light - no more than 200 new cases of Covid-19 per 100,000 population registered in 14 days;

* yellow light - no more than 100 new cases of Covid-19 per 100,000 population registered in 14 days;

* green light - no more than 20 new cases of Covid-19 per 100,000 population registered in 14 days.

We are currently sitting in the dark red traffic light, which means that no measures will be taken to reduce restrictions.

Cumulative is not average

The incidence rate can be calculated in different ways, such as an average per day, per week, per month or any other time scale. In this case, the daily figures are added up and divided by the number of days. If calculated in this way, the 14-day average in Latvia was 573 yesterday. If we calculate this per 100,000 population, the number is 30. Instead of calculating the average, in Latvia, as in other member states of the World Health Organization (WHO), the incidence of Covid-19 is expressed as a 14-day cumulative indicator, as it “shows the trend”. The cumulative indicator is calculated by adding together the number of cases confirmed in the last 14 days. For example, yesterday's 14-day cumulative morbidity rate in Latvia was 8,016. Recalculated per 100,000 inhabitants - 420, which is twice as high as the red zone, which would allow reducing restrictions. In order to reach the government-defined threshold of 200 cases per 100,000 inhabitants, there must be no more than 3815 cases in Latvia in 14 days. There were 8016 cases in the past days. Far too many to ease restrictions, in the government's opinion.

However, even if not a single new positive case of Covid-19 was detected in Latvia in the next three days, the cumulative indicator would still be in the bright red traffic light zone developed by the government.

Only if no new cases of Covid-19 had been recorded for a whole week would the cumulative 14-day Covid-19 rate per 100,000 population finally leave the bright red zone. As it is impossible during a pandemic for all tests to be negative, it looks like Latvia could stay in a bright red zone for quite some time.

It is for this reason that epidemiologists predict that even in May, Latvia will still be in the bright red zone if even stricter conditions are not adopted. And only if even more stringent measures, such as a lockdown, were introduced could the two-week cumulative incidence of Covid-19 be reduced to 60 cases per 100,000 population by 1 May.

The more tests, the higher the number of infected

In absolute numbers (rather than as a percentage), the number of people infected in a pandemic depends on the number of tests performed - the more tests, the more cases. This relationship does not work only in the situation where there is no one infected at all. Then you can test no matter how many people - one, hundred, thousand, million or even all 1,907,675 Latvians - the result will be the same, i.e., zero. If there is an infection in the country, as is currently the case with Covid-19, then the number of tests matters. For example, if yesterday, instead of 17,483 tests, half as many tests had been performed, then knowing that the proportion of positive tests is 3.6%, it can be assumed that 319 cases would have been detected in Latvia instead of 638 cases. If the number of tests had been halved for the entire 14 days, then, mathematically, the number of registered infections and, accordingly, the cumulative 14-day rate would have been halved to 210 cases. Of course, these are just mathematical calculations, because in life the situation may be different, namely, if tests were done only for the highest risk groups, the number of cases would be higher and vice versa, testing only the lowest risk groups, where Covid-19 has little chance of spreading, would result in a lower number of Covid-19 infected.

In the middle of summer, when Latvia was proud of its low morbidity rate - over two weeks, from 3 August to 16 August, 79 cases of Covid-19 were registered in Latvia with 24,326 tests, so there have been positive tests in an average of 0.3% of cases. If six times as many tests had been performed in these two weeks of August, that is, approximately as many as at present, 454 cases or 23 cases per 100,000 population in 14 days would have been recorded if we calculate this mathematically. According to the government traffic light principle, Latvia would have been in the yellow zone. Mathematically calculating how many new cases of Covid-19 there would be in Latvia in 14 days, if as few tests were performed as in summer (24,326 tests in 2 weeks), assuming that 5% of tests are positive (two-week average), they would be 1216 cases or 64 cases per 100 000 population. And we are in the same yellow traffic light zone!

Undoubtedly, the number of tests is important for the reduction of infection, because the ability to identify foci of infection and act faster depends on them. However, is the 14-day cumulative number of cases, without taking into account other indicators, really the only indication of an increase or decrease in infection? Last autumn, the Ministry of Health explained that the stabilization of the situation would be indicated if the number of hospitalized patients would be less than the number of patients discharged from the hospital, the number of cases of unknown origin do not exceed 30%, the proportion of positive tests do not exceed 4%, the number of persons tested per day are at least 4000. The proportion of positive cases against those tested was 3.6% yesterday. In the European Union, the proportion of positive tests is considered significant because, above the 4% threshold, the spread of the disease is considered to be rapid and uncontrolled. In Latvia yesterday, the proportion of positive cases was 3.6 percent, 17,483 tests were performed. 706 patients have been hospitalized over two weeks (629 with moderate disease, 77 with severe disease). Yesterday, 61 patients were admitted to the hospital, 65 were discharged, according to SPKC data.

*****

Be the first to read interesting news from Latvia and the world by joining our Telegram and Signal channels.