Competing causes of death

European Society of Cardiology CongressParis, August 28, 2011

Ulrich Keil and Angela Spelsberg

Institute of Epidemiology and Social Medicine, University of Münster

Tumorzentrum Aachen e.V., Aachen

No conflict of interest

Competing causes of death

Mortality race: cardiovascular versus cancer?

Record female life expectancy from 1840 to the present

Source: Oeppen J, Vaupel JW. Broken limits to life expectancy. Science 2002; 296: 1029–31

Sans S et al. Eur Heart J 1997; 18: 1241

The burden of cardiovascular disease mortality in Europe, 1970–92Men, age 45–74 years

Time trends in CVD mortality Annual % change in CVD mortality rates

Evolution of mean age-adjusted mortality rates fromtotal cardiovascular diseases in different world regions, 1970–1996 Men, age 75–84 years

Source: Kesteloot H et al. Evolution of all-causes and cardiovascular mortality in the age-group 75–84 years in Europe during the period 1970–1996.Eur Heart J 2002; 23: 392

EAE = Eastern Europe

WE = Western Europe

OCE = Oceania

NA = North America

JAP = Japan

Time trends in TCV mortality in 1970–2000 (A) men aged 45–74 years ; (B) women aged 45–74 years.

Kesteloot H et al. Eur Heart J 2006;27:107-113© The European Society of Cardiology 2005.

All cancers Lung Intestines Breast Prostate

Males Females Males Females Males Females Females Males

Year

1982a 187.0 108.3 53.2 8.1 19.7 14.3 20.7 13.8

1987a 191.1 107.9 55.0 9.1 20.2 14.0 21.5 14.9

1992a 187.8 105.5 53.4 9.9 20.5 13.5 21.2 15.3

1997a 177.8 100.5 49.2 10.6 20.1 12.6 19.8 15.1

2002 166.5 95.2 44.4 11.4 18.8 11.1 17.9 13.9

Change in rate, 1987–1997

Absolute     −13.3 −7.4 −5.8 +1.5 −0.1 −1.4 −0.7 +0.2

Percent     −7.0 −6.9 −10.5 +16.5 −0.5 −10.0 −7.9 +1.3

Annual %b     −0.78 −0.74 −1.16 +1.36 −0.15 −1.08 −0.80 +0.14

Change in rate, 1997–2002

Absolute     −11.3 −5.3 −4.8 +0.8 −1.3 −1.5 −1.9 −1.2

Percent     −6.4 −5.3 −9.8 +7.5 −6.5 −11.9 −9.6 −7.9

Annual %b     −1.22 −1.04 −1.91 +1.65 −1.60 −2.52 −1.72 −1.41

Age-standardised mortality rates per 100 000 (world standard population) from all cancers, lung, intestinal, breast and prostate cancers by gender in the European Union, 1982–2002

a Average annual rate in quinquennial periods 1980–1984, 1985–1989, 1990–1994 and 1995–1999. Levi F. et al., Ann Oncol 2007; 18:593-5 b From a log-linear model based on single calendar year rates. Copyright ©  2011 European Society for Medical Oncology

Trends in age-standardised (per 100 000, world population) mortality rates for cancers by gender in the 25 countries of the European Union, 1980–2002.

Levi F et al. Ann Oncol 2006;18:593-595

© 2006 European Society for Medical Oncology

Average life expectancy at birth

The level of mortality within a population can be expressed

as average life expectancy.

The average life expectancy of a specific calendar year is

calculated using mortality rates of each age group within

the observed year.

To analyse changes of life expectancy between two points

in time, Pollard’s method calculates the contribution of each

age group and within this group of each cause of death to

the observed changes of life expectancy.

Pollard JH: The expectation of life and its relationship to mortality. J Inst Actuaries 1982; 109: 225–40

Contribution of different disease groups on the increase in average life expectancy for men between 1980 and 2002 in Germany

Weiland SK, Rapp K, Klenk J, Keil U. Zunahme der Lebenserwartung. Dtsch Ärztebl 2006; 103(16):A 1072–7.

external causes

other diseases

perinatal diseases

diseases of the respiratory system

diseases of the cardiovascular system

malignant neoplasms

infections

Contribution of different disease groups on the increase in average life expectancy for women between 1980 and 2002 in Germany

Weiland SK, Rapp K, Klenk J, Keil U. Zunahme der Lebenserwartung. Dtsch Ärztebl 2006; 103(16):A 1072–7.

external causes

other diseases

perinatal diseases

diseases of the respiratory system

diseases of the cardiovascular system

malignant neoplasms

infections

Annual and cumulative contribution of different disease groups to the changes in life expectancy between 1962 and 2008 in Germany, men

Adapted from: Klenk J, Rapp K, Büchele G, Keil U, Weiland SK. Increasing life expectancy in Germany: quantitative contributions from changes in age- and disease-specific mortality. Eur J Public Health 2007; 17:587–592.

Annual and cumulative contribution of different disease groups to the changes in life expectancy between 1962 and 2008 in Germany, women

Adapted from: Klenk J, Rapp K, Büchele G, Keil U, Weiland SK. Increasing life expectancy in Germany: quantitative contributions from changes in age- and disease-specific mortality. Eur J Public Health 2007; 17:587–592.

Annual and cumulative contribution of specific types of cancer to the changes in life expectancy between 1962 and 2005 in Germany, men

Analog zu: Klenk J, Rapp K, Büchele G, Keil U, Weiland SK. Increasing life expectancy in Germany: quantitative contributions from changes in age- and disease-specific mortality. Eur J Public Health 2007; 17:587–592.

Annual and cumulative contribution of specific types of cancer to the changes in life expectancy between 1962 and 2005 in Germany, women

Analog zu: Klenk J, Rapp K, Büchele G, Keil U, Weiland SK. Increasing life expectancy in Germany: quantitative contributions from changes in age- and disease-specific mortality. Eur J Public Health 2007; 17:587–592.

Annual and cumulative contribution of different disease groups to the changes in life expectancy by sex between 1987 and 1997 in East Germany

Klenk J, Rapp K, Büchele G, Keil U, Weiland SK. Increasing life expectancy in Germany: quantitative contributions from changes in age- and disease-specific mortality. Eur J Public Health 2007; 17:587–592.

Frequency of causes of death from CVD and cancer in different age groups in men and women, Germany 2002

Source: Weiland SK, Klenk J, Keil U et al. 2005

0

5.000

10.000

15.000

20.000

25.000

30.000

35–39 40–44 45–49 50–54 55–59 60–64 65–69 70–74 75–79 80–84

Age group

Fre

qu

en

cyCardiovascular diseases (CVD) – MenCardiovascular diseases (CVD) – WomenMalignant neoplasms – MenMalignant neoplasms – Women

Women

Men

Summary (1)

• CVD mortality rates have been decreasing in western countries by about 2 % annually (since 1970); this is particularly true for higher age groups such as 75-84 years, and 85 years and older

• Mortality rates from all cancers have been decreasing in the European Union from 1987-1997 by 0.8 % and 0.7 % annually for males and females, respectively; and from 1997-2002 by 1.2 % and 1.0 % annually for males and females, respectively.

• From 1980-2002 life expectancy in Germany increased by 5.8 years in men and 4.6 years in women: nearly 50% of this increase in both men and women was due to declining CVD mortality rates; declining cancer mortality rates contributed by about 10 % !

• From 1990–1997 life expectancy in the eastern part of Germany increased by 3.2 years (!) for men and women. Declining CVD mortality rates contributed even more to this rapid increase than in the western part of the country. Contribution from declining cancer mortality rates was minimal.

• Age groups 65 years and older and CVD mortality declines contribute most to the increase in life expectancy between 1962-2008

(men 10.3 years; women 9.5 years).

• Contributions to life expectancy from declining CVD mortality began rather late in Germany, around 1980.

Summary (2)

• In spite of the impressive increases in life expectancy in Germany in recent decades, every 5th man (18.4%) and every 10th woman (9.6%) are dying before reaching their 65th birthday (present pension age).

• Cardiovascular diseases are the most frequent causes of death in Germany and other western countries; however up to age 65–69 years, deaths from cancer are predominating in women and men.

Summary (3)

Summary (4)

„ Given the inevitibility of death, a decline in the proportionate mortality from one cause must be compensated by a rise in the proportion of deaths ascribed to others. It is, therefore, probable that persons who previously would have succumbed to one of the CVDs are now dying from cancer.

However, what is important is not whether these „saved“ persons are dying from cancer (or any other cause) but the age at which they are dying from competing causes.“

Source: Lopez AD: Competing causes of death. Annals NY Academy of Sciences 1990; 609:58-74

Conclusions

Obviously, what is good for the prevention of cardiovascular

diseases is also effective against cancer. This is why cancer mortality rates are also declining, though on a slower pace.

To use the mortality race picture: In recent decades cardiovascular mortality (and incidence) has been shifted to higher age groups much faster than cancer mortality, but according to recent predictions, cancer may catch up.