© 1993 Oxford University Press
research-article |
Effect of Age, Birth Cohort and Period of Death on Italian Liver Cirrhosis Mortality, 1972–1986
* Centro Interdipartimentale di Epidemiologia, Università degli Studi di L'Aquila, Via Verdi 28, 67100 L'Aquila, Italy
** Institute of Medical Statistics, University of Milan Italy
Division of Gastroenterology, Ospedale Mauriziano Umberto 1 Turin, Italy
Liver cirrhosis is one of the main causes of death in Mediterranean countries. A trend towards a global reduction in the mortality rate has been recently reported. In order to clarify better this trend and in an attempt to hypothesize the future pattern of mortality, we analysed data from 254 834 Italian subjects aged 30–79 who died from liver cirrhosis during the period 1972–1986. We used a log-linear Poisson model to examine the effects of age, calendar period of death and birth cohort. Our data confirm that both in the population as a whole and after stratification for three geographical areas (Northern, Central and Southern Italy) the mortality rate is decreasing. The age-effect analysis showed an exponentially rising effect in the Southern population, in accordance with the viral aetiology of cirrhosis, whereas an increased effect followed by a decreased effect was observed in the Northern and Central population, suggesting the alcoholic aetiology for the disease. The results from the birth-cohort effect suggested that in the Northern and Central populations mortality should continue to decrease over the next decade, possibly due to the implementation of better prevention programmes for cirrhotics and to decreased alcohol consumption in Italy. In the Southern population, however, mortality is still rising and this will probably continue for the next decade, as the generations born between 1940 and 1950 who are at high risk of carrying chronic hepatitis B virus infection, reach the age of higher risk of death from liver cirrhosis. We conclude that such an epidemiological approach may help in understanding the variations of the trends of mortality from liver cirrhosis and make it possible to hypothesize the future trend pattern, thus helping to target more appropriately health care and prevention programmes.
Received 1 November 1992