5.12.3. Data description and analysis

Graphic representations of cirrhosis mortality for men and women since the 1970’s have been reported in Figures 5.12.1 (A and B) and 5.12.2 (A and B).

Figure 5.12.1.a. Joinpoint analysis of cirrhosis mortality for men - 35-64 years of age in selected European countries, 1970-2002

Figure 5.12.1.b. Joinpoint analysis of cirrhosis mortality for men all ages in selected European countries, 1970-2002

Figure 5.12.2.a. Joinpoint analysis of cirrhosis mortality for women - 35-64 years of age in selected European countries, 1970-2002.

Figure 5.12.2.b. Joinpoint analysis of cirrhosis mortality for women all ages in selected European countries, 1970-2002.

Table 5.12.1 gives the age-standardized death certification rates from cirrhosis in men at all ages and at 35 to 64 years of age for various European countries in 1980-82, 1990-92, and 2000-02, plus the corresponding changes in percentage.

Table 5.12.1. Age-adjusted (world population) mortality rates from cirrhosis per 100,000 men at all ages and at 35-64 years of age from various European countries in 1980-82, 1990-92 and 2000-02, and corresponding change in rates.

For European men, the highest rates in the early 1980s were in Southern and Central Europe, i.e. Portugal (38.7/100,000), Italy (35.0), France (30.9), as well as in Austria (34.5) and Hungary (33.9), while the lowest ones were in the England and Wales (3.6/100,000), Ireland (3.4), Norway and the Netherlands (around 5.5). In the subsequent two decades, substantial declines in male cirrhosis mortality were observed in Southern Europe, and in the early 2000s France, Italy and Spain had rates between 12 and 15/100,000, and Portugal of around 18/100,000. Similar declines were observed in Greece, Malta and Sweden, where rates were around 10-13/100,000 men in the early 1980s and around 5/100,000 in 2000-02. Consequently, these countries had the lowest male cirrhosis mortality in 2000-02, together with the Netherlands and Norway, whose rates remained around 5/100,000 men throughout all the calendar period considered. The falls were smaller in Austria, Germany and the Czech Republic, whose rates declined from 27-34 in 1980-82 to 18-20/100,000 in 2000-02, and even smaller in Slovakia and Slovenia, whose rates in 2000-02 were 31-35/100,000 men. Eight European countries diverged from the favourable trends of cirrhosis mortality in European men: Denmark whose rates increased from 11.3 in 1980-82 to 15/100,000 in 2000-2002, Finland from 7.4 to 13.6, Ireland from 3.4 to 4.8, England and Wales from 3.6 to 8.9, Scotland from 7.8 to 20.3, Bulgaria from 12.5 to 19.4, Poland from 13.0 to 15.9, and mainly Hungary, where mortality from cirrhosis rose from 33.9 in 1980-82 to 67.3 in 1990-92, and stabilized around 68 in the following decade. Consequently, in 2000-02 there was still an over ten-fold difference in mortality from cirrhosis in European men, although the pattern had totally changed as compared to only two decades earlier. The downward trends in several European countries in truncated rates (35-64 years of age) from cirrhosis in men were similar to those observed in overall rates. However, the rises were larger – at least in absolute terms – in some European countries, with truncated rates reaching 172.2/100,000 men in Hungary and 106.1 in Romania. Slovakia and Slovenia showed also extremely high mortality rates from cirrhosis in middle-aged men (around 75/100,000).

Mortality rates from cirrhosis were lower in women from all countries (Table 5.12.2), but trends were similar to those of men, with substantial declines throughout Southern Europe. As in men, rises were observed in women from a few Nordic countries (Denmark and Finland), the UK, central and Eastern European countries (Bulgaria, Romania and particularly Hungary). In European women, the highest rates in 2000-02 were 22.6/100,000 in Romania, followed by Hungary with 20.9, and the lowest ones, below 2/100,000, were in Greece and Malta. Taking into consideration truncated rates (35-64 years of age), in 2000-02 the highest values for women were 53.9/100,000 in Hungary and 44.3 in Romania. Slovakia and Slovenia had also extremely high truncated mortality rates from cirrhosis (around 24-30/100,000).

Table 5.12.2. Age-adjusted (world population) mortality rates from cirrhosis per 100,000 women at all ages and at 35-64 years of age from various European countries in 1980-82, 1990-92 and 2000-02, and corresponding change in rates.

The main findings from joinpoint regression analysis of cirrhosis mortality rates over the 1970-2002 period in selected European countries are given in Table 5.12.3 for men and in Table 5.12.4 for women. Given the substantial variation in absolute values across countries, a limited number of different scales has been chosen. Thus, it is important to check the scale for each country before making an inference on trends. Joinpoint analysis indicates that the rates have been steadily declining in several Western and Southern European countries since the mid or late 1970s (APC between -5 and -1.5% in both sex in the last decade only). In contrast, rates have been rising in Eastern European countries (such as Bulgaria, Hungary, Poland and Romania) up to the mid 1990s. Subsequent declines have, however, been observed also in these countries in more recent years particularly in men, with an APC between -4 and -3%. In the UK and other Nordic countries, mortality from cirrhosis was still steadily rising, with an APC around +7% in men and +3% in women from England and Wales, around +9% in men and +7% in women from Scotland and around +9% in men and +4% in women from Ireland.

Table 5.12.3. Joinpoint regression analysis for age-standardized cirrhosis mortality rates in men from selected EUGLOREH countries, 1970-2002

Table 5.12.4. Joinpoint regression analysis for age-standardized cirrhosis mortality rates in women from selected EUGLOREH countries, 1970-2002.

This updated analysis of mortality from cirrhosis shows a general favourable trend in several countries of southern Europe, with a history of high rates. Conversely, persisting upward trends up to the more recent calendar periods were observed in Ireland, the UK (mainly Scotland), and in a few countries from central and Eastern Europe (e.g., Hungary, Romania, Croazia, Slovakia), which in the early 2000s had extremely high cirrhosis mortality.

Changes in trends over time cannot be attributed to changes in the ICD, since no meaningful changes were observed around the years when subsequent Revisions of the ICD were introduced. Some of the variation may, however, be due to changed validity of cirrhosis diagnosis and certification, which has been stigmatized and may consequently not be systematically reported in death certificates in several countries. Improvements in the management of cirrhosis – including endoscopic banding, drug treatment and transjugular intraheatic portosystemic stent shunt to prevent variceal bleeding, better control of liver functions and liver transplantation – may also have had some favourable impact on mortality from the disease in selected countries (de la Pena et al, 2005; Ioannou et al, 2003; Poynard et al, 1999; Sauer et al, 2002). However, the impact of these factors on national mortality rates is difficult to quantify and unlikely to be substantial.

At least part of these changes may be due to the classification as deaths from cirrhosis of deaths from acute liver intoxication or failure (as well as from cardiovascular conditions) in subjects with cirrhotic liver; further attention to diagnosis and death certification validity from liver disease is therefore required.