Mortality in Ukraine Over the Last 25 years: How Many Deaths Could Have Been Avoided

Ukraine occupies 2^nd place in the world in death rate

Prior to war, Ukraine was losing 80 people per hour, leading us to a shameful second place  in the world in terms of the number of deaths per thousand of population (15.72 in 2014). Between South Africa (17.49) and Lesotho (14.91), with the EU average being around 10. Figure 1 below shows the dynamics of the crude death rate for Ukraine, Estonia, Russia and the EU average for the sake of comparison. As can be seen, in early 1990s Ukraine was on par with Estonia and somewhat better than Russia. All three countries have experienced a rapid increase in mortality by year 1994. After that only Estonia had seen a steady decline in mortality, with Russia and Ukraine returning to an upward trend till year 2005. From that time onwards both Russia and Ukraine had observed a decline in the overall mortality with the pace of the decline being slower in the latter.

Can Health Care System be Blamed?

Yes and no. Mortality, as the very final measure of population health, depends on a number of factors, among which people’s living conditions, environment, access to state social protection, and individual health related behaviors such as smoking, alcohol consumption, physical activity. Yet, one can see a number of accusations in the media regarding the role of the Health Care system in the increasing number of deaths. What is the share of deaths for which one can held accountable the health care system? As turns out, there is a limited number of causes of death which can be avoidable either by means of medical treatment or via state policies. How has Ukraine done on this front? Is the picture similar to the overall mortality rate dynamics?

The whole concept of ‘avoidability’ of deaths originated in early 1920s in the UK with the thorough investigations of maternal deaths, which should not have occurred following normal pregnancies. Later on, the concept has been picked up by David Rutstein who brought together experts from medicine, epidemiology, public health and related fields to identify a number of conditions for which death or disability should not occur if timely and effective medical care is administered.  The group suggested that any ‘unnecessary disease or disability and unnecessary untimely death’ that occurs in these conditions should serve as a warning signal, a “sentinel health event”, of the poor quality of healthcare received.

Rutstein’s original list of untimely diseases, disability and death has been revised several times to take into account advances in medical care as well as to incorporate changes made to the coding of causes of deaths (ICD codes).  Charlton and colleagues contributed immensely to the development of the concept through several studies of the regional variation in avoidable mortality in England and Wales.  The concept was later broadened to include causes preventable not only via medical intervention but also via public policies.

The concept of avoidable mortality refers, therefore, to all those deaths that, given current medical knowledge and technology, could be avoided by the healthcare system through either prevention and/or treatment. The remainder of deaths include those which cannot be avoided, either by the terminal nature of the disease for which there is no known cure, or due to very old age, or due to an accident for which no policy could be devised.

A potential caveat to the applicability of the concept, is that the selection of conditions/diseases that are amenable to both healthcare treatment and prevention depends on the particular circumstances of any given country at any given point in time.  Thus, some of the deaths which are considered avoidable today in a developed country, half a century ago might not have been.  Likewise, given the state of healthcare systems in a developing nation, many of the diseases for which the cure exists somewhere may be yet unavailable in this particular country.  However, in the era of globalization and openness most technologies, especially health-related, are de facto tradable.  Therefore, what constrains less developed countries from using superior technologies is lack of resources or government priorities not the general availability of the technologies themselves. Therefore, the dynamics of avoidable mortality in a country represents a much accurate outcome measure for the functioning of the country’s existing institutions (both medical care and public health policies) affecting population health than that of the overall mortality.

Avoidable mortality indicator (AM) is constructed from the WHO Mortality database. List of causes amenable to health care is presented in Table 1 (Nolte and McKee 2004). As is evident from the table the deaths from these conditions are only considered avoidable if they occur within the specified age range.  Causes which are preventable via health policies include cancers of trachea, bronchus and lung (which are primarily related to smoking) and road traffic injuries. Low or zero values of the avoidable mortality indicator in the country serves as an evidence of good performance of the health care system (MCI) or public health system (HPI) in the respective area. Data series for Ukraine starts in 1981. However, prior to year 2005 the coding used to classify death did not allow separate identification of all avoidable conditions from those unavoidable. Therefore, one cannot compare earlier period to other countries.

As can be seen from Figure 2, deaths which can be avoided via medical and policy intervention represent only about a third of overall mortality (for example in 2010 the total mortality was 15 per 1,000 or 1500 per 100,000 while avoidable mortality in this period was around 600 for males and 400 for females), yet the pattern of avoidable mortality in Ukraine is still alarming. Deaths which could have been avoided with existing state of medical technologies have been on the rise in Ukraine since early 1990s. Another disturbing feature is the startling difference in the avoidable mortality rates between males and females. Although this is the feature commonly observed in other countries, but the magnitude of this difference in Ukraine is huge.

It is important to stress that the years prior to 2005 represent a significant underestimation of the avoidable mortality rate, because in early years Ukraine submitted statistics on mortality classified according to the Soviet system and some causes of deaths were grouped into broad categories. This made it impossible to separate many causes of deaths which are avoidable from those unavoidable. Yet, after 2005 there is some positive dynamics observed. Sadly, the series stops with the data from 2012 since Ukraine has not reported mortality statistics to the WHO afterwards.

How does Ukraine compare to other countries in terms of avoidable mortality?

To avoid differences in the classification of causes of deaths across countries, the time period past 2005 has been chosen, when all the countries switched to the same ICD-10 classification. As Figures 3-4 show, past 2005 Ukraine is slightly worse off than Russia in terms of avoidable mortality among males, and it is considerably worse off in terms of avoidable mortality among females. If compared to Estonia, one of the three former Soviet Republics which are the new EU member states, the situation looks extremely dramatic – avoidable mortality in Estonia is almost twice smaller throughout the period, and the difference is even greater for females than it is for males.

As expected, avoidable mortality in the United Kingdom is considerably smaller compared to the three other countries, showing that there is a tremendous potential for health care systems and policies to improve population health.

Is there a room for improvement outside Health Care systems?

Figure 5 shows the dynamics in mortality avoidable via policy intervention, which basically includes smoking related cancers and road traffic injuries and is reported consistently over time. There are several interesting observations from this Figure. First of all, it reflects gender dimension of the automobile usage and smoking in Ukraine. In spite of the alarming messages regarding the increasing smoking prevalence among women, men in Ukraine are much more likely to smoke. And they are more likely to be driving vehicles as well. Thus, higher rate of mortality in this category among females in the UK than in Ukraine. Second, it shows that among males policy-related avoidable mortality in Ukraine is still much higher than that in the United Kingdom and, although it is decreasing, there is still a lot of potential for improvement. Third, there is virtually no improvement in the policy-related avoidable mortality among females in Ukraine over the whole period considered, which should be an alarming sign for policy makers.

To sum up

Since its independence Ukraine has been experiencing a significant increase in mortality rates among both males and females. Although some positive dynamics has been observed after year 2005, the situation is likely to worsen given the war and resulting disturbances. But at the moment this is impossible to analyse due to the lack of data. Contrary to the common rhetoric, health care system can only be held accountable for part of these deaths – avoidable mortality. Ukraine compares very poorly to other countries on both the levels and the dynamics of avoidable mortality. Estonia serving a particularly great example as it started with the same levels of avoidable mortality as Ukraine in 1991 and now is in a much better position approaching the levels of developed countries. Analysis of the avoidable mortality which can be decreased via policy intervention shows great potential for the country both in terms of smoking policies and road traffic interventions.

References:

Castelli, Adriana and Olena Nizalova. 2011. “Avoidable Mortality: What It Means And How It Is Measured”, University of York Centre for Health Economics Research Paper 63.
Rutstein, David D., William Berenberg, Thomas C. Chalmers, Charles G. Child, Alfred P. Fishman, and Edward B. Perrin. 1976. Measuring the Quality of Medical Care. New England Journal of Medicine 294 (11):7.
Charlton, J. R. H, R. M. Hartley, R. Silver, and W. W. Holland. 1983. Geographical variation in mortality from conditions amenable to medical intervention in England and Wales. Lancet 1 (8326):691-696.
Nolte, Ellen, and C. Martin McKee. 2004. Does Health Care Save Lives? Avoidable mortality revisited. London: The Nuffield Trust.