**2. Development of the concept of avoidable mortality**

The concept of avoidable mortality was developed by Rutstein et al. [3]. They suggested that several diseases at certain ages should not occur in the presence of timely and effective health care. Additionally, they distinguished the diseases that should be amenable by the quality of health care (e.g. diabetes mellitus treated with insulin) and those that were influenced by public health policy interventions (e.g. lung cancer prevented by smoking elimination). Their list of causes of death included more than 90 diseases considered as unnecessary, untimely causes of death and disabilities. Many research studies have tried to renew the list over time adjusting to the new medical and technological advances. In 1983, Charlton et al. [4] modified the number of conditions on 14 disease groups and excluded deaths that were not directly associated with health care, for example, deaths avoided by public health prevention programs comprising alcohol or tobacco consumption. At the end of the 1980s, the concept was proceeded by several researchers [5, 6], but the highest progress was achieved by Holland [7] who created a European Community atlas of avoidable mortality modifying the previous authors. As for a main benefit of the atlas, strict distinguishing between types of health care services on primary care, hospital care and collective health services was interpreted. In 1993 and 1997, second and third editions of atlas adjusting the number of diseases were created by Holland [8, 9] again, and further developed by other authors [10–13]. In spite of changing list of causes of death, age limit was mostly set at 65 years, what was about the average life expectancy in developed countries in those years. According to experts, above this age, the treatment of selected diseases is less obvious and appearance of co‐morbidities becomes problematic.

system performance. Evaluating the effectiveness of health care systems requires pre‐defined objectives or the expected health outcomes that are usually measured by mortality and morbidity. More specifically, length of life and quality of life are examined. These aggregate indicators are not able to capture a clear impact of health care activities, especially quality of health care, on the health status of population. Therefore, more specific health outcome indicators were developed, e.g. avoidable mortality by selected causes of death, infant mortality,

prevalence or incidence of chronic diseases, avoidable hospitalisations, and others [1].

population characteristics [2].

72 Advances in Health Management

and international levels?

The question of how much health care contributes to the health of populations has been discussed for several decades. Although there is no indicator that would comprehensively reflect the performance of health care system, nevertheless, the suitable measurement seems to be a concept of avoidable mortality. The concept of avoidable mortality, as an indicator for the quality of health care services, defines premature deaths from selected disease groups that are considered either treatable through the medical services or preventable by influencing the

From the beginning 1970s, many researchers have tried to renew the list of causes of death considered as amenable by health care or preventable by health interventions (see Chapter 2). Unfortunately, many studies did not demonstrate the selection process of avoidable causes of death. This is a bias that raises a question if this concept is not influenced by subjective approach of the given researchers. Has this concept a potential to be applied both at national

The main objective of this study is to analyse the impact of the four lists of causes of death on amenable mortality by country, sex and cause of death. By application of several methods of avoidable mortality, we have an ambition to point out possible fluctuations in their results and limitations of international comparison. These raise an appeal for confrontation of the scientific teams at national and international levels and for the development of comparative international baseline. The most innovative part of the study is disputation whether the concept

This chapter consists of six sections. Theoretical background of avoidable, amenable and preventable mortality, together with a literature review follows Section 1. Section 3 presents the description of the two recent modifications of the concept including cause of death structure. Section 4 deals with the empirical analyses of amenable mortality differentials across the European Union countries and describes the data and methods used. Section 5 provides a discussion about the potentials or limitations of the concept applied. The most meaningful

The concept of avoidable mortality was developed by Rutstein et al. [3]. They suggested that several diseases at certain ages should not occur in the presence of timely and effective health care. Additionally, they distinguished the diseases that should be amenable by the quality

of avoidable mortality is reliable for international health systems comparison or not.

conclusions are summarised at the end of the chapter.

**2. Development of the concept of avoidable mortality**

In 2001, Tobias and Jackson [14] derived the weights for primary, secondary and tertiary health interventions on the basis of a medical expert consensus. For example, avoidability of deaths from HIV/AIDS was distributed according to the primary level with weights 0.9, the secondary level with weights 0.05 and the tertiary level weighted 0.05. To compare, deaths from hypertensive disease were avoidable first by secondary interventions with weights 0.65, second by tertiary interventions with weights 0.3 and finally by primary interventions weighted 0.05. Unfortunately, all above‐mentioned proposals of the concept of avoidable mortality did not consider the availability of health care resources such as current technology, medical skills, human resources or health expenditures in a certain country.

A new perspective view on the concept was presented by Nolte and McKee [15] in 2004. They conducted a broad review of randomised controlled trials providing the evidence of impact of health services on survival taking into account advances in medical knowledge and technology across the European Union countries during the 1980s and 1990s. The previous lists of causes of death created by Mackenbach et al. [6] or Charlton et al. [4] were changed on 34 groups of diseases comprising amenable, preventable conditions and ischaemic heart disease separately. Ischaemic heart disease was represented as a separate group because the highest number of these deaths could bias the influence of health services on other diseases. Additionally, the concept considers only 50% of deaths from ischaemic heart disease. Another reason was that ischaemic heart disease could be understood partially as amenable but also as preventable cause of death. Some causes of death were added to the list and some were removed. For example, malignant neoplasm of prostate was not included because an available time trends analysis of cancer mortality showed a small decrease of mortality from prostate cancer, together with the uncertain impact of screening. On the other hand, they included colorectal cancer on the basis of randomised controlled trials providing that curative resection had a significant impact on survival. Establishing an upper age limit varied across diseases. The vast majority were set at 75 years, with the exception of diabetes mellitus (lower than 50 years), some infectious and respiratory diseases (lower than 15 years), malignant neoplasm of cervix uteri and body uterus, as well as leukaemia (lower than 45 years). This was the result of studies that reported substantial improvements in mortality from these diseases relating to advances in treatment before mentioned age limits.

The concept was further renewed analysing European and non‐European countries due to the works by Nolte and McKee [16] in 2008 and Tobias and Yeh [17] in 2009. Nolte and McKee closely followed up their last list of causes of death from 2004, while Tobias and Yeh discussed some new inclusion and exclusion criteria. Infectious diseases varied significantly. While Nolte and McKee concentrated on infectious disease of children before the age of 15, Tobias and Yeh focused on selective invasive bacterial infections such as scarlet fever, meningococcal infection, etc. They argued that early detection and treatment by antibiotic therapy decrease mortality substantially. Moreover, only half of the mortality from cerebrovascular diseases, ischaemic heart disease and diabetes mellitus are considered as amenable by appropriate health care according to Tobias and Yeh, because the second half can be preventable by health behaviours (e.g. healthy lifestyle, obesity prevention). The authors of mentioned lists of diseases have different opinions on setting age limit for some causes of death; however, there are more similarities than discrepancies between these two lists of diseases (**Table 1**).


Assessment of Avoidable Mortality Concepts in the European Union Countries, Their Benefits and Limitations http://dx.doi.org/10.5772/67818 75

age limit varied across diseases. The vast majority were set at 75 years, with the exception of diabetes mellitus (lower than 50 years), some infectious and respiratory diseases (lower than 15 years), malignant neoplasm of cervix uteri and body uterus, as well as leukaemia (lower than 45 years). This was the result of studies that reported substantial improvements in mortality from these diseases relating to advances in treatment before mentioned age

The concept was further renewed analysing European and non‐European countries due to the works by Nolte and McKee [16] in 2008 and Tobias and Yeh [17] in 2009. Nolte and McKee closely followed up their last list of causes of death from 2004, while Tobias and Yeh discussed some new inclusion and exclusion criteria. Infectious diseases varied significantly. While Nolte and McKee concentrated on infectious disease of children before the age of 15, Tobias and Yeh focused on selective invasive bacterial infections such as scarlet fever, meningococcal infection, etc. They argued that early detection and treatment by antibiotic therapy decrease mortality substantially. Moreover, only half of the mortality from cerebrovascular diseases, ischaemic heart disease and diabetes mellitus are considered as amenable by appropriate health care according to Tobias and Yeh, because the second half can be preventable by health behaviours (e.g. healthy lifestyle, obesity prevention). The authors of mentioned lists of diseases have different opinions on setting age limit for some causes of death; however, there are more similarities than discrepancies between these two lists of

**Cause of death Nolte and McKee [16] ICD‐10 Tobias and Yeh [17] ICD‐10**

Tuberculosis A15–A19, B90 A15–A19, B90

Intestinal infectious diseases A00–A09 (age 0–14) Non‐classified Whooping cough A37 (age 0–14) Non‐classified Measles B05 (age 1–14) Non‐classified Tetanus and Diphtheria A35–A36 Non‐classified Sepsis A40–A41 A40–A41 Scarlet fever Non‐classified A38 Meningococcal infection Non‐classified A39

Acute poliomyelitis A80 Non‐classified Influenza J10–J11 Non‐classified Pneumonia J12–J18 J13–J15, J18

Erysipelas Non‐classified A46 Legionnaires disease Non‐classified A481 Malaria Non‐classified B50–B54

limits.

74 Advances in Health Management

diseases (**Table 1**).

**Infectious disease**

Selected invasive infections:



**Table 1.** Causes of death selected in the amenable mortality list of Nolte and McKee [16] and Tobias and Yeh [17].

The second latest study 'Amenable mortality in the European Union: toward better indicators for the effectiveness of health systems' (AMIEHS) [18] in 2011 introduced an empirical evidence of selecting diseases into the lists of causes of death. Finally, a recent project has referred to the avoidable mortality indicators defined according to the Eurostat 'Satellite List' Task Force [19] in 2013. A common objective of these studies is to reach a consensus by countries of the European Community about the definition and selection of causes of avoidable deaths. Both studies are further described in Section 3 more specifically for the purposes of our analysis.
