Lung cancer mortality in European men: Trends and predictions

Lung Cancer 80 (2013) 138–145

Contents lists available at SciVerse ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

Lung cancer mortality in European men: Trends and predictions Matteo Malvezzi a,b , Cristina Bosetti a,∗ , Tiziana Rosso a , Paola Bertuccio a,b , Liliane Chatenoud a , Fabio Levi c , Canzio Romano d , Eva Negri a , Carlo La Vecchia a,b a

Dipartimento di Epidemiologia, Istituto di Ricerche Farmacologiche “Mario Negri”, Milan, Italy Dipartimento di Scienze Cliniche e di Comunità, Università degli Studi di Milano, Milan, Italy Institute of Social and Preventive Medicine (IUMSP), Lausanne University Hospital, Lausanne, Switzerland d Dipartimento di Traumatologia, Ortopedia e Medicina del Lavoro, Università degli Studi di Torino, Turin, Italy b c

a r t i c l e

i n f o

Article history: Received 12 October 2012 Received in revised form 8 January 2013 Accepted 27 January 2013 Keywords: Cancer Europe Male Lung Mortality Projections Trends

a b s t r a c t Lung cancer mortality in men from the European Union (EU) peaked in the late 1980s at an agestandardised (world standard population) rate over 53/100,000 and declined subsequently to reach 44/100,000 in the early 2000s. To provide a comprehensive picture of recent trends in male lung cancer mortality in Europe, we analyzed available data from the World Health Organization up to 2009 and predicted future rates to 2015. Lung cancer mortality rates in EU men continued to fall over recent years, to reach a value of 41.1/100,000 in 2005–2009. The fall was similar at all-ages and in middle-aged men (less than 2% per year over most recent years), but was appreciably larger in young men (aged 20–44 years, over 5% per year). A favourable trend is thus likely to be maintained in the foreseeable future, although the predicted overall EU rate in 2015 is still over 35/100,000, i.e., higher than the US rate in 2007 (33.7/100,000). Over most recent calendar years, overall male lung cancer rates were around 35–40/100,000 in western Europe, as compared to over 50/100,000 in central and eastern Europe. Within western Europe, lung cancer rates were lower in northern countries such as Sweden, but also Finland and the UK (below 30/100,000), where the tobacco-related epidemic started earlier and rates have long been declining, whereas mortality was high in Belgium (51.6), France (42.3), the Netherlands and Spain (around 43.0), where the epidemic started later but is persisting. Widespread measures for smoking control and cessation in middle-aged European men, i.e., in the generations where smoking prevalence used to be high, would lead to appreciable reductions in male lung cancer mortality in the near future. This is particularly urgent in central and eastern European countries. © 2013 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

were still large differences in lung cancer mortality across Europe, with rates between 35 and 45/100,000 in the largest western European countries (France, Germany, Italy, and the UK), and between 55 and 80/100,000 in Hungary, Poland, Russia and a few other central and eastern Europe countries. In order to provide an up-to-date comprehensive picture of trends and predict short-term burden of lung cancer mortality in European men, we analyzed available data to 2009 using joinpoint analysis and age-period-cohort (APC) analysis [4].

Lung cancer mortality in men from the European Union (EU) peaked in the late 1980s at an age-standardised (world standard population) rate over 53/100,000 and declined subsequently to reach 44/100,000 in the early 2000s, with a larger fall in middleaged men, from 80 to 61/100,000 [1]. After the mid/late 1990s, downward trends in male lung cancer mortality were observed in most European countries, including those from central and eastern Europe with the highest mortality and the most unfavourable trends until the late 1990s [1–3]. In the early 2000s, however, there

2. Materials and methods

Abbreviations: APC, age-period-cohort; CIs, confidence intervals; EAPC, estimated annual percent change; EU, European Union; ICD, International Classification of Diseases; OR, odds ratio; PI, prediction intervals; WHO, World Health Organization. ∗ Corresponding author at: Department of Epidemiology, Istituto di Ricerche Farmacologiche “Mario Negri”, Via Giuseppe La Masa 19, 20156 Milan, Italy. Tel.: +39 0239014526; fax: +39 0233200231. E-mail address: [email protected] (C. Bosetti).

We obtained official male death certification data for lung cancer from the World Health Organization (WHO) database for 33 European countries from 1970 to 2009, when available [5]. Mortality data for the EU as a whole from 1970 to 2007 were generated by aggregating data from its 27 member states as defined in January 2007, with the exclusion of Cyprus for which data were available for a limited number of most recent years only. No interpolation was made for missing data in the computation of single country

0169-5002/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.lungcan.2013.01.020

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

rates. For the calculation of the EU rates, when data were not available for a country, the nearest available data (i.e., generally that of the previous or subsequent year) were replicated. In Belgium, 2004 data were used for 2000–2003, and 2005 data were used for 2005–2007; in Denmark, 2006 was used for 2007; in Germany, 1973 was used for the whole 1970–1974 quinquennium, and 1978

139

was used for 1979; in Italy, 2003 was used for 2004, and 2006 for 2005; in Latvia, the 1980–1984 quinquennium was replicated over 1970–1974 and 1975–1979; in Luxembourg, 1971 was used for 1970; in Poland, 1996 was used for 1997, and 1999 for 1998; in Portugal, 1971 was used for 1970 and 2003 was replicated over the whole 2003–2007; in Romania, 1978 was used for 1979; in Slovenia

Fig. 1. Trends in age-standardized (world standard population) death rates for lung cancer per 100,000 men (all ages and 35–64 years) in selected European countries and truncated 35–64 years . the EU as a whole from 1970 to 2009. All ages

140

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

Table 1 World standardized death rates from lung cancer per 100,000 men (at all ages and truncated at 35–64 and 20–44 years) in selected European countries and in the European Union in the periods 2000–2004 and 2005–2009, and corresponding percent changes. Country

Austria Belarus Belgium Bulgaria Croatia Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Norway Poland Portugal Romania Russian Federation Slovakia Slovenia Spain Sweden Switzerland Ukraine United Kingdom United Kingdom, England and Wales United Kingdom, Northern Ireland United Kingdom, Scotland European Union

All ages

35–64 years

2000–2004

2005–2009

35.81 55.12 49.99 39.41 62.11 57.41 41.60 59.55 31.57 44.07 39.17 47.73 79.46 26.14 36.24 45.44 56.36 54.23 44.32 37.49 48.76 30.59 67.64 28.72 46.76 57.11 53.63 50.61 45.71 20.51 32.06 47.16 37.05 35.79 38.07 49.88 44.66

32.35 47.79 51.60 44.34 57.06 48.10 41.03 54.21 28.24 42.31 34.99 46.45 72.57 28.17 32.95 38.89 53.44 51.51 37.26 32.54 43.03 28.66 61.95 28.55 47.93 51.21 46.69 46.93 43.19 19.27 29.20 40.93 33.12 31.95 36.36 44.29 41.05

% change −10 −13 3 13 −8 −16 −1 −9 −11 −4 −11 −3 −9 8 −9 −14 −5 −5 −16 −13 −12 −6 −8 −1 2 −10 −13 −7 −6 −6 −9 −13 −11 −11 −4 −11 −8

the quinquennium 1985–1989 was replicated over the 1970–1974, 1975–1979, and 1980–1984 quinquennia; in Spain, 1971 was used for 1970; in Estonia and Lithuania, the average of 1981 and 1982 data was used for the 1980–1984 quinquennium, and then replicated over the 1970–1974 and 1975–1979; data for Czechoslovakia were used for the 1970–1991 period while, from 1992 onwards, data for the Czech Republic and Slovakia were used separately. During the calendar period considered, three different revisions of the International Classification of Diseases (ICD) were used. Since differences in lung cancer mortality classification between various revisions were minor, they were re-coded for all countries according to the Tenth Revision of the ICD (code C34) [6]. Estimates of the resident population, based on official censuses, were obtained from the same WHO database [5]. Population projection estimates for the year 2015 were obtained from the EUROSTAT database [7]. From the matrices of certified deaths and resident populations, we computed age-specific rates and numbers of death for each age group (0, 1–4, 5–9, . . ., 80–84, 85+ years) and calendar year. Agestandardized rates, per 100,000 men, at all ages, truncated at age 35–64 and 20–44 years were computed using the world standard population with the direct method [8]. A joinpoint regression model was used to analyze mortality rate trends over the studied period [9,10]. According to this model, the standardized rates, measured on a logarithmic scale, change linearly with calendar years over specific time intervals where the slope of the trend remains constant. Joinpoint regression can be used to identify the points, called the “joinpoints”, where the linear

20–44 years

2000–2004

2005–2009

51.45 92.86 63.50 77.46 91.57 84.68 45.67 83.32 32.27 72.71 52.29 69.56 140.15 28.74 39.91 53.82 86.41 82.31 55.86 42.05 49.65 35.94 101.15 45.35 91.00 94.58 78.87 69.91 67.96 23.44 40.57 83.87 39.56 38.06 40.62 54.45 62.66

47.02 75.95 63.97 83.80 87.71 67.01 45.01 70.20 29.70 69.63 47.02 66.64 129.92 29.53 35.96 43.96 76.57 74.74 43.50 33.44 44.07 31.40 87.31 45.92 89.58 80.95 65.26 67.27 63.85 20.90 38.33 71.28 34.69 33.35 38.84 47.07 57.18

% change −9 −18 1 8 −4 −21 −1 −16 −8 −4 −10 −4 −7 3 −10 −18 −11 −9 −22 −20 −11 −13 −14 1 −2 −14 −17 −4 −6 −11 −6 −15 −12 −12 −4 −14 −9

2000–2004

2005–2009

% change

2.10 3.84 2.34 5.30 4.58 2.37 2.05 3.48 0.77 4.61 2.32 3.47 8.19 2.29 1.36 2.14 3.26 2.97 3.46 1.53 2.18 1.26 3.80 3.49 6.30 4.36 2.89 3.47 4.23 0.79 1.76 4.67 1.32 1.30 1.28 1.54 3.05

1.53 2.95 2.32 4.90 3.13 1.67 1.99 1.30 0.97 3.26 1.78 2.63 4.85 0.31 1.22 1.61 2.73 1.83 2.56 1.35 1.73 0.73 2.57 2.88 4.66 3.53 2.14 2.25 2.99 0.70 1.48 3.75 1.28 1.25 0.95 1.73 2.29

−27 −23 −1 −7 −32 −29 −3 −63 25 −29 −23 −24 −41 −87 −10 −25 −16 −38 −26 −11 −21 −42 −32 −18 −26 −19 −26 −35 −29 −11 −16 −20 −3 −4 −26 12 −25

slope of the trend changes significantly. Models with a maximum of 4 joinpoints (corresponding to up to 5 different trends) were considered. For a trend described by the relationship y = a + bx, where y is ln(rate) and x is calendar year, the estimated annual percent change (EAPC) is calculated by 100 × (eb − 1). Furthermore, for seven selected European countries (i.e., France, Germany, Hungary, Italy, Poland, Spain, and the UK) and the EU as a whole, we provided estimates of numbers of lung cancer deaths and rates for the year 2015. These were derived by fitting a joinpoint model to each 5-year age specific (age groups 0–4, . . ., 75–79, 80+) number of certified deaths, assuming a Poisson distribution, in order to identify the most recent trend segment; a linear regression was then performed on mortality data from each age group over the time period identified by the joinpoint model in order to compute the predicted age-specific number of deaths, the corresponding 95% confidence intervals (CIs) and prediction intervals (PIs) (i.e., the confidence intervals for the prediction of a single future value, calculated using a standard error that accounted for the variability of the possible new observation) [4]. Standardized death rates (and 95% CIs and PIs) for 2015 were computed using the predicted population data from EUROSTAT [7]. The age-specific mortality rates per 100,000 population for 5year age groups (from 30–34 to 75–79 years) and, for 5-year periods (from 1970–74 to 2005–09) were used for the APC analysis [11]. Cohorts were defined according to their central year of birth. Thus the earliest possible cohort, centred in 1895, relates to individuals aged 75–79, who died in the quinquennium 1970–1974, and could

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

141

Fig. 2. Trends in age-standardized (world standard population) death rates from lung cancer per 100,000 men at 20–44 years in selected European countries and the EU as a whole from 1970 to 2009.

have been born between 1890 and 1899. The effects of age, cohort of birth and period of death were evaluated with a log-linear Poisson model using a likelihood penalizing function method to solve the identifiability problem [12,13]. The age effects are interpretable in terms of mean age-specific death rates throughout the periods

considered; the cohort and period of death effects are expressed in relative terms against their weighted average set to unity. Since this modelling technique does not allow for the calculation of CIs in a conventional analytic manner, a parametric bootstrap simulation technique was used [14].

142

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

Table 2 Joinpoint analysis for lung cancer mortality in men from major European countries and in the European Union (at all ages, and truncated at 35–64 and 20–44 years), 1970–2009. Country

Trend 1

Trend 2

Trend 3

Trend 4

Trend 5

Years

EAPC

Years

EAPC

Years

EAPC

Years

EAPC

All ages France Germany Hungary Italy Poland Spain United Kingdom European Union

1970–1978 1973–1978 1970–1986 1970–1982 1970–1984 1971–1975 1970–1978 1970–1979

3.7a 1.3a 3.7a 3.3a 3.9a 5.9a −0.2 1.6a

1978–1989 1978–1993 1986–1996 1982–1988 1984–1991 1975–1989 1978–1985 1979–1987

1.3a −0.3a 1.8a 0.8a 1.8a 3.3a −1.7a 0.6a

1989–1998 1993–2010 1996–2009 1988–1994 1991–2002 1989–1995 1985–1991 1987–1993

−0.2 −2.2a −1.4a −1.2a −0.5a 1.2a −2.5a −0.5a

1998–2008

−1.1a

1994–2008 2002–2009 1995–2008 1991–2000 1993–2007

−2.6a −1.9a −1.0a −4.0a −1.8a

35–64 years France Germany Hungary Italy Poland Spain United Kingdom European Union

1970–1978 1973–1990 1970–1986 1970–1980 1970–1984 1971–1990 1970–1976 1970–1979

3.6a 0.5a 5.6a 3.4a 4.6a 3.4a −1.5a 1.5a

1978–1991 1990–2010 1986–1994 1980–1987 1984–1991 1990–1999 1976–1985 1979–1987

1.4a −2.4a 2.7a 0.4 1.6a 0.1 −2.5a 0.8a

1991–2008

−0.7a

1994–2009 1987–2008 1991–2009 1999–2008 1985–1999 1987–1993

−1.3a −3.7a −2.4a −1.4a −4.3a −1.0a

1999–2009 1993–2000

−2.7a −2.5a

20–44 years France Germany Hungary Italy Poland Spain United Kingdom European Union

1970–1997 1973–1986 1970–1978 1970–1984 1970–1980 1971–1993 1970–1996 1970–1979

1.8a 0.4 8.6a −0.2 3.7a 4.5a −4.0a 1.3a

1997–2008 1986–1999 1978–1992 1984–2008 1980–1995 1993–2008 1996–2001 1979–1995

−6.2a −2.6a 4.3a −4.8a −0.8a −5.3a −7.3a −0.1

1999–2010 1992–1998

−5.1a −1.2

1998–2009

−9.1a

1995–2009

−6.4a

2001–2009 1995–2007

−0.5 −5.6a

Years

EAPC

2000–2009

−2.3a

2000–2007

−1.7a

EAPC, estimated annual percent change. a Significantly different from 0 (p < 0.05).

3. Results Table 1 gives the age-standardised mortality rates from male lung cancer (at all ages and truncated at 35–64 and 20–44 years) in 33 European countries (plus England and Wales, Northern Ireland, and Scotland separately) and the EU as a whole, in the 2000–2004 and 2005–2009 quinquennia with the corresponding percentage changes. Between 2000–2004 and 2005–2009, in the EU, male lung cancer mortality fell by 8% overall, from 44.7 to 41.1/100,000, by 9% in the 35–64 age group, from 62.7 to 57.2/100,000, and by 25% in the 20–44 age group, from 3.1 to 2.3/100,000. Downwards trends were observed in most European countries, the only exceptions being Belgium, Bulgaria, Iceland, and Romania. In 2005–2009, the highest mortality rates were in Hungary (72.6/100,000), Poland (62.0), Croatia (57.1), and Baltic countries (around 52–54/100,000); Sweden had the lowest rates (19.3/100,000), followed by other Nordic countries and Portugal (around 28/100,000). Similar geographical patterns and time trends were observed in middle-aged and young men, although trends were generally more favourable in men 20–44 years old. Only in Finland and Scotland, increased rates for this age group were registered. Fig. 1 shows the trends in age-standardised mortality rates from lung cancer in men at all ages and at 35–64 years in selected European countries and the EU as a whole. Fig. 2 shows the corresponding trends in men aged 20–44 years. At all ages and in the middle-aged, mortality rates have been declining since the 1970s in Austria, Finland, and the UK (Fig. 1). In most other European countries considered, mortality rates had a peak in the 1980s or the 1990s, to then start levelling off or declining. In particular, in countries of central and eastern Europe (as Bulgaria, Hungary, Poland, Romania, but also Russia) mortality has been increasing up to the 1990s to decline thereafter. In several countries the difference between all ages and middle-aged rates has been closing. Only in Bulgaria, Iceland, Romania and Portugal mortality has been still

increasing over more recent years. The trends in young men were favourable in all countries considered, at least since the 1990s, apart from Finland and Scotland where they were levelling off over recent calendar years (Fig. 2). Table 2 gives the results from the joinpoint analysis of male lung cancer mortality rates (at all ages and truncated at age 35–64 and 20–44) for seven selected European countries and the EU as a whole. The EU showed significant rising trends in mortality up to 1987 that levelled off subsequently with a significant decline by 1.8% per year between 1993 and 2007 overall, and by 1.7% between 2000 and 2007 in middle-aged men; in young men mortality in the EU levelled off about 10 years earlier, and declined by 5.6% per year in 1995–2007. Downward, but widely variable, trends were observed in various European countries: the EAPCs over the most recent calendar years were between −1.0% (Spain) and −2.6% (Italy) at all ages, between −0.7% (France) and −3.7% (Italy) in middle-aged men, and between −4.8% (Italy) and −9.1% (Hungary) in young men. Only in the UK, declines in overall mortality rates were recorded over the whole studied period, although EAPCs showed smaller falls in the last calendar periods as compared to previous ones. Fig. 3 displays the results from the APC analysis performed on male lung cancer mortality data from seven selected European countries and the EU as a whole. The first graph shows lung cancer mortality trends stratified by both quinquennium of age (horizontal black lines) and cohort of birth (diagonal grey lines) plotted (on a log scale) against the period of death; the second graph illustrates the age curve (rates/100,000 against age at death) with 95% CIs; while the third graph shows the effects of cohort of birth and period of death (with corresponding 95% CIs) against central year of birth and death respectively. The age specific trends were favourable, particularly over the more recent periods and younger age groups, in all the countries considered. Italy, Poland and Spain shared a similar cohort effect structure, with rising effects in the older cohorts that reach a peak in the 1930s in Italy, the 1940s in Poland, and the

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

143

Fig. 3. Trends in lung cancer mortality rates in men from selected European countries and the European Union as a whole, stratified by both quinquennium of age (horizontal black lines) and cohort (diagonal grey lines) plotted (on a log scale) against period of death (first panel); effect estimates of age (30–34 to 75–79 years, second panel), cohort (central years of birth 1895–1985) and period (1970–1974 to 2005–2009, third panel) from age-period-cohort analysis. Age effects are expressed as mortality rates per 100,000 men; cohort and period effects are expressed as multiplicative effects relative to the age effects. Effect estimates are in black, approximate 95% confidence interval limits in grey.

1950s in Spain, to subsequently show a steady decrease up to the most recent cohorts of birth. Germany and the UK showed an overall declining cohort effect pattern, with a rising effect only seen in the earliest cohorts of birth. The cohort effect plot for the EU had three distinct peaks – that correspond to the peaks in cohort effects in the main country groups – while in the cohorts born after the 1960s showed strong declines. Persisting favourable trends in overall male lung cancer mortality rates are predicted up to 2015 Fig. 4 and (Supplemental Table 3): in the EU as a whole the predicted number of lung cancer deaths in men in 2015 is about 187,000, corresponding to a standardized mortality rate of 35.5/100,000. Corresponding predicted values are 24,200 deaths (rate 38.7/100,000) in France, 29,000 (29.9) in Germany, 5600 (68.7) in Hungary, 24,500 (30.3) in Italy, 17,000 (52.3) in Poland, 18,400 (38.5) in Spain, and 20,000 (28.5) in the UK (Supplemental Table 3).

4. Discussion This up-to-date analysis shows that the declines in lung cancer mortality rates in EU men since the late 1980s [1] have continued over recent years, and are projected to persist for the near future. The decline over more recent calendar years is, however, less than

2% per year, and the predicted overall EU rate in 2015 is still over 35/100,000, i.e., higher than the US rate in 2007 (33.7/100,000) [15]. The shape of the decline was similar in all-ages and in middle-aged men, but was appreciably larger in young men (aged 20–44 years), where the decline was over 5% per year over most recent years. Since lung cancer trends in the young are reliable indicators of future lung cancer rates [16–18], this suggests that lung cancer trends are likely to be more favourable in the future. Indeed, about 25% of young (age 20–25 years) European men now smoke, as compared to over 60% in the 1980s [19], indicating that lung cancer rates are due to decline substantially when younger generations of European men reach middle-age. In Central and Eastern European countries (as Bulgaria, Hungary, Poland, Romania, but also the Czech Republic) – after increasing trends observed up to the late 1990s – mortality from lung cancer showed appreciable declines over the most recent decades, with particularly favourable patterns in young men. National mortality patterns and time trends from lung cancer essentially reflect the tobacco spread and the subsequent decline of smoking prevalence – following the knowledge of the health consequences of smoking and related campaigns and policies to control and limit tobacco use [20–22] – in subsequent cohorts of men in various countries [23–25]. Occupational exposures, including mainly asbestos, were estimated to account for 5–10% of male

144

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145

the epidemic is still ongoing and may remain at high levels in the future, when the same generations reach middle-age. This calls for effective and urgent measures for tobacco control in those countries. Even more urgent is the implementation of comprehensive tobacco control policies in central and eastern European countries, where – notwithstanding the steady declines in mortality from lung cancer over more recent years – death rates remain 50% to over 100% higher than in most western Europe, corresponding to over 20,000 excess lung cancer deaths per year in that area of the continent as compared to average western European rates. Such a persistent epidemic is attributable to the higher smoking prevalence among men from those countries as compared to western countries, and the delays in the implementation of tobacco preventive measures [2,42–45]. Our short-term lung cancer mortality predictions are comparable to those proposed by others [46]. Assuming that the projected trends in lung cancer mortality will be observed, in 2015 there will be about 30,000 fewer lung cancer deaths in the EU as compared to those observed in 2007. In any case, a favourable trend is likely to be maintained for the foreseeable future, given the decreased tobacco prevalence in young generations of EU men. Still, over 187,000 lung cancer deaths are predicted for 2015 in the EU. Consequently, only widespread measures for smoking control and cessation in middle-aged European men, i.e., in the generations where smoking prevalence used to be extremely high, will lead to appreciable reductions in male lung cancer mortality in the near future [47]. Role of funding sources

Fig. 4. Age-standardized (world standard population) death rates from lung cancer in men from selected European countries and the European Union as a whole from 1970 to 2009, and predicted rates for 2015.

This work was conducted with the contribution of the Italian Association for Cancer Research (AIRC grant no. 10264), the Italian and Swiss Leagues Against Cancer, and the Swiss Foundation for Research Against Cancer (KFS-02437-08-2009). The sponsors had no role in study design, collection, analysis and interpretation of data. Conflict of interest statement

lung cancer deaths in industrialized countries, including eastern Europe [23,26–29]. Indoor air pollution has also been a relevant lung carcinogen in western, but mostly in central and eastern Europe, with relative risks of 1.2–1.3 for solid fuel use [30]. The control of asbestos and other industrial carcinogens has contributed to the fall in male lung cancer mortality over the last few decades, but its contribution is likely to be minor in the future, since asbestos has a late stage effect on lung cancer [31,32] and massive exposure to asbestos has occurred in the 1970s and 1980s in most European countries. The decline in outdoor [33] and indoor [34] air pollution may also have played some role, which, however, is masked within the overwhelming impact of tobacco on lung cancer rates. Male lung cancer mortality trends are unlikely to have been affected by changes in survival, since only modest improvements in the diagnosis and treatment have been introduced in the last decades and the prognosis for this neoplasm remains poor [35–37]. The trends and magnitude of lung cancer mortality rates in European men have been variable according to their different phases in the tobacco epidemic [38,39]. Over most recent calendar years, overall male lung cancer rates were around 35–40/100,000 in western Europe, as compared to over 50/100,000 in central and eastern Europe. Also within western Europe, lung cancer rates were lower in northern countries such as Finland and the UK, where the tobacco-related epidemic broke out earlier and rates have long been declining, whereas mortality has been comparably high in Belgium, France, the Netherlands and Spain, where the epidemic started later but is now persisting longer [40,41]. Those countries also have comparatively high rates in young men, indicating that

The authors have no conflict of interest to declare. Acknowledgement The authors thank Mrs. Ivana Garimoldi for editorial assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.lungcan.2013.01.020. References [1] La Vecchia C, Bosetti C, Lucchini F, Bertuccio P, Negri E, Boyle P, et al. Cancer mortality in Europe, 2000–2004, and an overview of trends since 1975. Ann Oncol 2010;21:1323–60. [2] Tyczynski JE, Bray F, Aareleid T, Dalmas M, Kurtinaitis J, Plesko I, et al. Lung cancer mortality patterns in selected Central, Eastern and Southern European countries. Int J Cancer 2004;109:598–610. [3] Bray FI, Weiderpass E. Lung cancer mortality trends in 36 European countries: secular trends and birth cohort patterns by sex and region 1970–2007. Int J Cancer 2010;126:1454–66. [4] Malvezzi M, Bertuccio P, Levi F, La Vecchia C, Negri E. European cancer mortality predictions for the year 2012. Ann Oncol 2012;23:1044–52. [5] World Health Organization Statistical Information System. WHO mortality database. Available at: http://www3.who.int/whosis/menu.cfm [accessed July 2012]. [6] World Health Organization. International statistical classification of disease and related health problems. 10th revision ed. Geneva; 1992.

M. Malvezzi et al. / Lung Cancer 80 (2013) 138–145 [7] European Commission. Eurostat population database. Available at: http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/ [accessed October 2011]. [8] Doll R, Smith PG, Waterhouse JAH, Muir CS, Shanmugaratnam K, Powell J, et al. Comparison between registries: age-standardized rates. Cancer incidence in five continents, vol. IV. IARC Sci Publ No. 42.; 1982. pp. 671–5. [9] Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med 2000;19:335–51. [10] National Cancer Institute. Joinpoint Regression Program, version 3.5. Available at: http://srab.cancer.gov/joinpoint/2011 [11] Malvezzi M, Bonifazi M, Bertuccio P, Levi F, La Vecchia C, Decarli A, et al. An age-period-cohort analysis of gastric cancer mortality from 1950 to 2007 in Europe. Ann Epidemiol 2010;20:898–905. [12] Osmond C. Using age, period and cohort models to estimate future mortality rates. Int J Epidemiol 1985;14:124–9. [13] Decarli A, La Vecchia C. Age, period and cohort models: review of knowledge and implementation in GLIM. Riv Stat Appl 1987;20:397–410. [14] Efron B, Tibshirani R. An introduction to the bootstrap. Boca Raton: Chapman & Hall/CRC; 1993. [15] Jemal A, Thun MJ, Ries LA, Howe HL, Weir HK, Center MM, et al. Annual report to the nation on the status of cancer, 1975–2005, featuring trends in lung cancer, tobacco use, and tobacco control. J Natl Cancer Inst 2008;100:1672–94. [16] Doll R. Progress against cancer: an epidemiologic assessment. The 1991 John C. Cassel Memorial Lecture. Am J Epidemiol 1991;134:675–88. [17] Muir CS, Fraumeni Jr JF, Doll R. The interpretation of time trends. Cancer Surv 1994;19–20:5–21. [18] Jemal A, Cokkinides VE, Shafey O, Thun MJ. Lung cancer trends in young adults: an early indicator of progress in tobacco control (United States). Cancer Causes Control 2003;14:579–85. [19] World Health Organization Statistical Information System. Global Adult Tobacco Survey; 2011. Available at: http://www.who.int/tobacco/surveillance/ gats/en/index.html [accessed 02.10.11]. [20] Levy DT, Chaloupka F, Gitchell J. The effects of tobacco control policies on smoking rates: a tobacco control scorecard. J Public Health Manag Pract 2004;10:338–53. [21] Joossens L, Raw M. The tobacco control scale 2010 in Europe. Bruxels, Belgium: Association of European Cancer Leagues; 2011. [22] Martinez-Sanchez JM, Fernandez E, Fu M, Gallus S, Martinez C, Sureda X, et al. Smoking behaviour, involuntary smoking, attitudes towards smoke-free legislations, and tobacco control activities in the European Union. PLoS One 2010;5:e13881. [23] Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer in the United States today. J Natl Cancer Inst 1981;66:1191–308. [24] World Health Organization. Tobacco or Health: a global status report. Geneva, Switzerland: WHO; 1997. [25] Forey B, Lee P. Estimation of sex-specific smoking statistics by standardized age groups and time periods. Supplement 1. In: Forey B, Hamling J, Lee P, Wald N, editors. International Smoking Statistics, a collection of historical data from 30 economically developed countries. 2nd ed. Sutton, Surrey: P N Lee Statistics and Computing Ltd.; 2002. [26] Doll R, Peto R. Effects on health of exposure to asbestos. In: Health and safety commission. London: Her Majesty’s Stationery Office; 1985. [27] Carel R, Olsson AC, Zaridze D, Szeszenia-Dabrowska N, Rudnai P, Lissowska J, et al. Occupational exposure to asbestos and man-made vitreous fibres and risk of lung cancer: a multicentre case–control study in Europe. Occup Environ Med 2007;64:502–8. [28] Bardin-Mikolajczak A, Lissowska J, Zaridze D, Szeszenia-Dabrowska N, Rudnai P, Fabianova E, et al. Occupation and risk of lung cancer in Central and Eastern Europe: the IARC multi-center case–control study. Cancer Causes Control 2007;18:645–54.

145

[29] Olsson AC, Gustavsson P, Zaridze D, Mukeriya A, Szeszenia-Dabrowska N, Rudnai P, et al. Lung cancer risk attributable to occupational exposures in a multicenter case–control study in Central and Eastern Europe. J Occup Environ Med 2011;53:1262–7. [30] Lissowska J, Bardin-Mikolajczak A, Fletcher T, Zaridze D, Szeszenia-Dabrowska N, Rudnai P, et al. Lung cancer and indoor pollution from heating and cooking with solid fuels: the IARC international multicentre case–control study in Eastern/Central Europe and the United Kingdom. Am J Epidemiol 2005;162: 326–33. [31] Pira E, Pelucchi C, Piolatto PG, Negri E, Discalzi G, La Vecchia C. First and subsequent asbestos exposures in relation to mesothelioma and lung cancer mortality. Br J Cancer 2007;97:1300–4. [32] La Vecchia C, Boffetta P. Role of stopping exposure and recent exposure to asbestos in the risk of mesothelioma. Eur J Cancer Prev 2012;21: 227–30. [33] Gallus S, Negri E, Boffetta P, McLaughlin JK, Bosetti C, La Vecchia C. European studies on long-term exposure to ambient particulate matter and lung cancer. Eur J Cancer Prev 2008;17:191–4. [34] Straif K, Baan R, Grosse Y, Secretan B, El Ghissassi F, Cogliano V. Carcinogenicity of household solid fuel combustion and of high-temperature frying. Lancet Oncol 2006;7:977–8. [35] Karim-Kos HE, de Vries E, Soerjomataram I, Lemmens V, Siesling S, Coebergh JW. Recent trends of cancer in Europe: a combined approach of incidence, survival and mortality for 17 cancer sites since the 1990s. Eur J Cancer 2008;44:1345–89. [36] Coleman MP, Forman D, Bryant H, Butler J, Rachet B, Maringe C, et al. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995–2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. Lancet 2011;377:127–38. [37] Pastorino U, Rossi M, Rosato V, Marchiano A, Sverzellati N, Morosi C, et al. Annual or biennial CT screening versus observation in heavy smokers: 5-year results of the MILD trial. Eur J Cancer Prev 2012;21:308–15. [38] Youlden DR, Cramb SM, Baade PD. The International Epidemiology of Lung Cancer: geographical distribution and secular trends. J Thorac Oncol 2008;3:819–31. [39] Thun M, Peto R, Boreham J, Lopez AD. Stages of the cigarette epidemic on entering its second century. Tob Control 2012;21:96–101 [discussion 101–102]. [40] Hill C. Trends in tobacco smoking and consequences on health in France. Prev Med 1998;27:514–9. [41] Fernandez E, Schiaffino A, Borras JM, Shafey O, Villalbi JR, La Vecchia C. Prevalence of cigarette smoking by birth cohort among males and females in Spain, 1910–1990. Eur J Cancer Prev 2003;12:57–62. [42] Fagerstrom K, Boyle P, Kunze M, Zatonski W. The anti-smoking climate in EU countries and Poland. Lung Cancer 2001;32:1–5. [43] Brennan P, Crispo A, Zaridze D, Szeszenia-Dabrowska N, Rudnai P, Lissowska J, et al. High cumulative risk of lung cancer death among smokers and nonsmokers in Central and Eastern Europe. Am J Epidemiol 2006;164:1233–41. [44] Jha P, Peto R, Zatonski W, Boreham J, Jarvis MJ, Lopez AD. Social inequalities in male mortality, and in male mortality from smoking: indirect estimation from national death rates in England and Wales, Poland, and North America. Lancet 2006;368:367–70. [45] Hruba F, Fabiaova E, Bencko V, Cassidy A, Lissowska J, Mates D, et al. Socioeconomic indicators and risk of lung cancer in Central and Eastern Europe. Cent Eur J Public Health 2009;17:115–21. [46] Eilstein D, Uhry Z, Lim TA, Bloch J. Lung cancer mortality in France, Trend analysis and projection between 1975 and 2012, using a Bayesian age-period-cohort model. Lung Cancer 2008;59:282–90. [47] Peto R, Darby S, Deo H, Silcocks P, Whitley E, Doll R. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case–control studies. BMJ 2000;321:323–9.