top of page

COVID-19 Regional Safety Assessment 

250 Countries, Regions & Territories

Correlation between ratio of fatal cases and other indicators

The correlation coefficient provides both the strength and direction of the relationship between the independent and dependent variables. Its values range between - 1.0 and + 1.0. When it is positive, the relationship between x and y is positive, and when it is negative, the relationship is also negative. A correlation coefficient close to zero indicates that there is no correlation between x and y.


Correlation of Fatal cases ratio and Total cases per 1 million with the most significant indicators

Of the greatest importance is the effect of the elderly demographic (in % of total population) on Ratio of fatal cases (0.35), though it does not lead to higher Total cases per 1M (slightly above zero - 0,06). For the total number per 1 million population (i.e. spread of virus) the only (and a fairly high) correlation is revealed for the population densities (coefficient value is 0.21) and also obesity (0.25).

Correlation of Fatal cases ratio and Total cases per 1 million with the incidence and mortality from some other diseases


Correlation coefficients suggest that there is at least no effect on the number of COVID-19 cases and its mortality from the spread of Tuberculosis, and Endocrine Disorders Death Rate. However, there is a positive correlation with the Total cases per 1M of population for Obesity reflected in % of total population (Correlation coefficient value is 0.25) and Diabetes prevalence (0.13). 

Correlation of Fatal cases ratio and Total cases per 1 million with the healthcare-related indicators.

First of all, when analyzing the graphs of the healthcare system related indicators vs. fatal cases ratio (in percentage points) it is important to avoid the kind of heuristic oversimplifications that often occur wherever rapid decisions are made. These sometimes result in basic miscalculations when forming strategic agendas. Many economic indicators related to GDP have been used as performance criteria, often as a basis for setting normative values. However, this approach is often favoured out of a desire to cut corners, to find simple and quick solutions where none are possible. By the time this becomes apparent, it is usually too late and errors have taken root. Thus, in many cases a high level of health expenditure to GDP (in %) is an unreliable way of attaining a lower rate of fatal cases of COVID-19 (calculated by comparing fatal cases to the total number of cases). Surprisingly, a high level of health expenditure to GDP (in %) in a number of developed countries coincides with a relatively high level of fatal cases of COVID-19. Then again, a high level of health expenditures in developing countries may be due to poor reasons and motives, such as:

- Negligible budget combined with state monopoly on healthcare (a large portion of a negligible national budget arbitrarily assigned to healthcare).

- High morbidity rate due to humanitarian crises and lack of facilities, imposing a heavy burden on health systems and creating a cycle of underfunding.

- Financial aid, including to a large extent donations specifically for items that relate to healthcare (when not given in-kind).

- Corruption and inefficient spending policies. 


At the very least, it seems less likely that health expenditures, represented in US$ thousand per capita, or the number of doctors per thousand people, might include the kind of statistical distortions that are normally encountered when calculating indicators related to GDP. And it is not only health expenditures (both as % of GDP and in US$ per capita) that have more or less significant positive correlations, but almost all indicators associated with a surplus in the scale and assumed efficiency of the healthcare systems coincides with the higher mortality rate and number of total cases per 1 million. The correlation coefficients for healthcare expenditures, number of doctors and nurses, Global Health Safety Index, Healthcare Access and Quality Index, Quantity of ICU-CCB beds per 100 000 inhabitants -  all have values from 0.1 to 0.3 when compared with the ratio of fatal cases. And in general for this range, the same values coincide with higher mortality. There is an insignificant correlation between the mortality from COVID-19 and the number of hospital beds, though it is also positive.

The relationship between the ratio of fatal cases or total cases per 1M of population vs. healthcare expenditures (both in relative terms as a ratio to GDP, and as an amount in USD per capita) either indicates that high spending does not guarantee the required results, or even an inverse relationship. At the very least, the calculated values of the ratio of fatal cases are often similar for underfunded healthcare systems as for those countries and regions which have negligible health spending. 


The irregular distribution of fatal cases in relation to healthcare expenditure, raises the possibility of a non-linear approach to the analysis of healthcare economic efficiency. It is irregular because the economic system and the rate of virus spread are linked by diverse complex factors. Given that healthcare systems worldwide deal with a multitude of diseases in a multitude of circumstances using a vast range of different tools, resources and practices, the relationships between healthcare system funding and COVID deaths defies analysis. 


The diversity of significant indicators attributed to healthcare scale and quality, such as number of doctors per 1000 citizens, number of nurses per 1000 citizens, healthcare safety index, health care access and quality index and even those associate with availability of equipment, does not enable us to make conclusions on the expectable interrelationship between a lower number of total cases per 1M of population and the ratio of fatal cases from COVID-19. As already mentioned, one can observe that those countries and regions which have less extensive healthcare systems often are those where mortality from COVID-19 and the spread of the virus is lowest. And, at the same time, countries and regions with higher levels of fatal cases often have not only large-scale, but also high quality healthcare systems. Calculations based on different approaches to correlation indexes and use of various other indicators is required for more evidence. Nevertheless, it is obvious that by attempting to tackle COVID-19 only by expanding healthcare systems, countries and regions may miss the other underlying reasons which are not fully explored. Either way, case-by-case analysis of the identified patterns associated with the regions, types of healthcare economic policies and applicable non-monetary measures, likely will bring desirable results. Especially if we do not allow ourselves to get complacent while these issues grow more urgent. More explicit patterns that can be attributed to the relationship between expenditures on healthcare systems (e.g. if calculating it per types and items expenses) and the efficiency of COVID-19 treatment in terms of avoiding fatal cases are yet to be identified. In addition, an institutional analysis of the health economy, which covers more than just economic indicators, could be of some value. 

bottom of page