Journal of Community Health

, Volume 34, Issue 3, pp 188–194

Trends in Lung Cancer Death Rates in Belgium and The Netherlands: a Systematic Analysis of Temporal Patterns

Authors

    • Division of Public Health PracticeHarvard School of Public Health
  • Zubair Kabir
    • The Digital DepotResearch Institute for a Tobacco Free Society
  • Gregory N. Connolly
    • Division of Public Health PracticeHarvard School of Public Health
Original Paper

DOI: 10.1007/s10900-008-9147-z

Cite this article as:
Van Hemelrijck, M.J.J., Kabir, Z. & Connolly, G.N. J Community Health (2009) 34: 188. doi:10.1007/s10900-008-9147-z

Abstract

Belgium and The Netherlands have fairly similar smoking prevalence patterns, but distinct tobacco control policies. It is our aim to use lung cancer death rates, especially among the youngest age groups (30–39 years), as indicators of past smoking behavioral patterns to evaluate recent tobacco control efforts in both countries. Lung cancer mortality rates from 1954 to 1997 and from 1950 to 2000 were investigated in Belgium and The Netherlands, respectively, using the joinpoint regression modeling technique (log-linear Poisson models) to calculate annual percent change in death rate. In the most recent period (1984–2000) overall male lung cancer death rates have been declining at a faster rate in The Netherlands than in Belgium. In contrast, overall female lung cancer death rates (between 1950 and 2000) have been increasing at a faster rate in The Netherlands than in Belgium. Since 1988, however, APCs in death rates among Dutch females have begun to level off. Interestingly, during this same period, a significant annual decline of 7.7% among the youngest Dutch women (30–39 years) has been observed. Tobacco use prevention and interventions seem to have an impact on smoking prevalence, especially among younger age groups. In The Netherlands, where aggressive anti-tobacco campaigns were introduced a few years earlier than in Belgium, male lung cancer mortality rates have been declining more rapidly, and female lung cancer mortality rates have begun to level off.

Keywords

Lung cancerTemporal patternBelgiumThe Netherlands

Introduction

Tobacco kills over 650,000 people annually in the European Union (2004). Ninety-one percent and 65% of these deaths are due to lung cancer, for males and females, respectively. Lung cancer death rates are thus strongly associated with smoking and can be used as indicators of smoking trends since patterns of smoking prevalence tend to precede patterns of lung cancer mortality by two to three decades [1].

With regard to smoking prevalence and health, Belgium and The Netherlands are fairly comparable countries. However, The Netherlands, ranked the third highest country in the EU for cigarette consumption, is one of the few EU countries with annual per capita cigarette consumption over 2,900. Belgium, on the other hand, has an annual per adult cigarette consumption of 1,837 [2, 3]. In The Netherlands, 25,725 smoking-attributable deaths were estimated in 2000, accounting for 18.3% of the total number of deaths [1, 4, 5]. There were an estimated 18,646 deaths attributable to smoking in Belgium in 2000, accounting for approximately 17.8% of the total number of deaths. Nevertheless, with regard to activities in tobacco control (i.e., taxing cigarettes, combating smuggling of cigarettes, protection from environmental tobacco smoke, display of health warnings…) The Netherlands perform much better than Belgium. In 2006, a measuring activities in tobacco control (MAToc) of 6.3 was reported for The Netherlands and a MAToc of 9.4 for Belgium—the lower the score the better the tobacco control activities in a country [6].

Since Belgium and The Netherlands have fairly similar smoking prevalence patterns but distinct tobacco control efforts, it is our aim to identify and describe temporal patterns for lung cancer mortality in both countries by calculating annual-percent-changes (APCs) in disease rates [7]. This analytical method was proposed by the US national cancer institute and has already been used in several European countries to identify temporal patterns in lung cancer death rates [3, 810]. Changes in lung cancer mortality rates in specific age-groups, particularly the youngest age group (30–39 years), can then be used to indicate temporal changes in smoking patterns, as patterns for lung cancer mortality also tend to follow those for smoking prevalence two to three decades earlier. These results may help to inform the efficacy of specific tobacco prevention and protection interventions previously administered by the Belgian and Dutch governments [1, 9, 11].

Methods

Annual age-standardized lung cancer death rates (per 100,000) were obtained from the world health organization’s (WHO) mortality database [12]. Countrywide world age-adjusted rates by gender were available for Belgium from 1954 to 1997 and for The Netherlands from 1950 to 2004. The age-standardized rates were adjusted to the world population (1960) and lung cancer diagnosis was based on the international classification of disease, 10th revision [12]. Data on adult (age 15+) smoking prevalence and cigarette consumption were acquired from the international smoking statistics which had data available until 1995 [13]. More recent data were obtained from the WHO regional office for Europe, tobacco control database [14]. When several values were reported for 1 year, the average value was used for the annual estimate. Estimates for missing values were calculated by using model-based imputation, more specifically via simple linear regression.

Temporal patterns in lung cancer death rates across different time-periods for each specific age group by gender were analyzed by calculating APCs with log-linear Poisson regression models, using “joinpoints”. Because of fewer lung cancer death rates among the younger age groups, the first four 5-year age categories (30–34, 35–39, 40–44, 44–49) were collapsed into two 10-year age categories (30–39 and 40–49, respectively). On a log scale to lung cancer death rates, the joinpoint analysis fits a series of joined straight lines. Line segments are joined at points called joinpoints. Each joinpoint denotes a statistically significant change in trend. The best-fitting points are the years wherein the lung cancer death rates change significantly (increasing or decreasing trends). The analysis tests how many joinpoints are statistically significant, with a maximum of three joinpoints to be added to the final model.

All analysis was performed using Microsoft Excel® (Microsoft, Redmond, WA, USA), SAS release 9.1.3 (SAS Institute, Cary, NC, USA), and Joinpoint 3.0 (National Cancer Institute, Bethesda, MD, USA) software packages.

Results

Belgium

The absolute number of total annual lung cancer deaths and their corresponding annual age-standardized death rates between 1954 and 1997 are shown in Fig. 1 (top). Male lung cancer death rates increased until 1983, but afterwards a slight decline was noticed. Female lung cancer death rates on the other hand showed an increase for the entire period.
https://static-content.springer.com/image/art%3A10.1007%2Fs10900-008-9147-z/MediaObjects/10900_2008_9147_Fig1_HTML.gif
Fig. 1

Male (left) and female (right) total lung cancer deaths and corresponding age-standardized lung cancer death rates by year in Belgium: 1954–1997 (top). Male (left) and female (right) smoking prevalence and per capita cigarettes/day by year in Belgium: 1954–1997 (bottom)

Table 1 shows APCs for lung cancer death rates in Belgium. Both male and female death rates showed a significant overall annual rise of 1.8 and 2.6%, respectively. For the age-specific death rates, female overall APCs (1954–1997) were approximately 2.5% for all age categories, whereas, males tended to have larger overall APCs among the older age groups (>65 year old), with the largest annual increase among the 85+ age groups (5.5%). Over the last 15 years male lung cancer death rates were significantly declining for all age-groups, with exception of the 50–54 age group.
Table 1

Annual percent changes (APC) with 95% confidence intervals (CI) of lung cancer death rates and joinpoint analysis: age-specific and age-standardized rates in Belgium: 1954–1997

Age-specific

APC (95% CI) 1954–1997

Trend 1

years

(First slope) APC (95% CI)

Trend 2

years

(Second slope) APC (95% CI)

Trend 3

years

(Third slope) APC (95% CI)

Males

    30–39

0.2 (−0.4;0.9)

1954–1960

6.8 (−3.2;17.8)

1960–1979

1.8 (0.2;3.3)

1979–1997

−2.8 (−4.4;−1.2)

    40–49

0.7 (0.4;1.0)

1954–1974

2.1 (1.5;2.7)

1974–1987

0.5 (−0.6;1.7)

1987–1997

−2.9 (−4.4;−1.2)

    50–54

0.4 (0.2;0.7)

1954–1982

1.4 (1.1;1.7)

1982–1989

−3.4 (−6.3;−0.4)

1989–1997

0.8 (−1.2;2.9)

    55–59

0.6 (0.3;1.0)

1954–1964

4.8 (3.4;6.3)

1964–1984

0.8 (0.4;1.3)

1984–1997

−2.2 (−2.9;−1.4)

    60–64

1.0 (0.6;1.5)

1954–1963

7.8 (6.1;9.4)

1963–1981

1.7 (1.2;2.2)

1981–1997

−1.8 (−2.3;−1.3)

    65–69

1.6 (1.1;2.2)

1954–1964

8.9 (7.6;10.3)

1964–1977

3.6 (2.9;4.3)

1977–1997

−1.1 (−1.4;−0.8)

    70–74

2.6 (1.8;3.3)

1954–1972

8.4 (7.8;9.0)

1972–1980

3.1 (1.6;4.7)

1980–1997

−1.3 (−1.6;−0.9)

    75–79

3.8 (3.0;4.6)

1954–1968

9.7 (7.2;12.1)

1968–1981

6.6 (4.9;8.4)

1981–1997

−0.8 (−1.7;0.1)

    80–84

4.8 (4.1;5.6)

1954–1978

9.1 (8.5;9.7)

1978–1984

4.7 (1.2;8.4)

1984–1997

−0.1 (−0.8;0.7)

    85+

5.5 (4.8;6.3)

1954–1983

8.5 (7.9;9.0)

1983–1993

1.9 (0.5;3.5)

1993–1997

−4.0 (−9.1;1.2)

Age standardized

    Overall (0–85+)

1.8 (1.4;2.3)

1954–1965

6.1 (5.4;6.8)

1965–1980

3.3 (2.9;3.6)

1980–1997

−1.1 (−1.4;−0.9)

    Truncated (30–84)

1.7 (1.3;2.2)

1954–1965

6.2 (5.5;6.9)

1965–1980

3.1 (2.8;3.5)

1980–1997

−1.3 (−1.5;−1.0)

Females

    30–39

1.4 (0.6;2.2)

1954–1984

0.3 (−1.3;1.9)

1984–1995

5.2 (−1.8;12.8)

1995–1997

−12.7 (−62.5;103.1)

    40–49

2.6 (2.1;3.0)

1954–1986

1.7 (1.0–2.3)

1986–1987

9.7 (−10.4;34.3)

1987–1997

1.9 (−2.8;6.9)

    50–54

2.8 (2.4;3.2)

1954–1962

7.5 (1.4;14.0)

1962–1978

−5.8 (−41.4;51.4)

1978–1997

3.1 (2.5;3.6)

    55–59

2.7 (2.5;3.1)

1954–1956

−11.28 (−38.6;28.1)

1956–1979

1.8 (−2.4;6.1)

1979–1997

3.0 (2.6;3.5)

    60–64

2.9 (2.6;3.2)

1954–1971

1.1 (−0.4;2.7)

1971–1991

4.0 (3.0;5.1)

1991–1997

0.6 (−3.9;5.4)

    65–69

2.8 (2.4;3.2)

1954–1991

2.2 (1.9;2.6)

1991–1994

13.1 (−11.0;43.5)

1994–1997

−0.4 (−11.0;11.3)

    70–74

2.5 (2.2;2.7)

1954–1989

2.3 (1.9;2.6)

1989–1992

5.2 (−17.4;34.0)

1992–1997

2.3 (−2.6;7.3)

    75–79

2.0 (1.7;2.4)

1954–1957

−5.1 (−24.7;19.5)

1957–1995

2.0 (1.6;2.4)

1995–1997

12.4 (−15.6;49.7)

    80–84

2.2 (1.7;2.6)

1954–1963

9.9 (4.8;15.2)

1963–1966

−4.9 (−36.6;42.7)

1966–1997

2.0 (1.5;2.6)

    85+

2.5 (1.9;3.1)

1954–1956

−21.1 (−65.4;79.6)

1956–1965

12.1 (2.9;22.0)

1965–1997

1.6 (0.9;2.3)

Age standardized

    Overall (0–85+)

2.6 (2.5;2.7)

1954–1956

−0.9 (−16.4;17.4)

1956–1984

2.3 (2.1;2.6)

1984–1997

3.3 (2.7;3.9)

    Truncated (30–84)

2.6 (2.5;2.8)

1954–1969

1.7 (1.0;2.5)

1969–1984

2.6 (1.9;3.3)

1984–1997

3.4 (2.8;4.1)

Annual adult smoking prevalence and per capita cigarette consumption (age 15+) patterns are shown in Fig. 1 (bottom). Both prevalence and consumption declined for Belgian males. In contrast, female cigarette consumption increased over time while prevalence remained rather stable (about 20%), with the exception of a temporary decrease in both consumption and prevalence among females during the early 1990s.

The Netherlands

The absolute number of total annual lung cancer deaths and their corresponding annual age-standardized death rates between 1950 and 2004 are shown in Fig. 2 (top). Male lung cancer death rates increased until 1982. Female lung cancer death rates, on the other hand, were constant up until the late 1960s, but increased steeply thereafter.
https://static-content.springer.com/image/art%3A10.1007%2Fs10900-008-9147-z/MediaObjects/10900_2008_9147_Fig2_HTML.gif
Fig. 2

Male (left) and female (right) total lung cancer deaths and corresponding age-standardized lung cancer death rates by year in The Netherlands: 1950–2004 (top). Male (left) and female (right) smoking prevalence and per capita cigarettes/day by year in The Netherlands: 1950–2000 (bottom)

Table 2 show APCs for lung cancer death rates as well as the joinpoint analyses for The Netherlands. Both male and female death rates showed a significant annual rise overall, 0.9 and 4.2%, respectively. For the age-specific death rates, we found that during the last 15 years approximately male rates were significantly declining for all age-groups. In contrast, all female age groups showed a significant annual increase in lung cancer mortality in the most recent periods, except for the youngest age group who showed an annual decrease of 7.7% since 1996.
Table 2

Annual percent changes (APC) with 95% confidence intervals (CI) of lung cancer death rates and joinpoint analysis: age-specific and age-standardized rates in The Netherlands: 1950–2000

Age-specific

APC (95% CI) 1950-2000

Trend 1

years

(First slope) APC (95% CI)

Trend 2

years

(Second slope) APC (95% CI)

Trend 3

years

(Third slope) APC (95% CI)

Males

    30–39

−1.7 (−2.0;−1.3)

1950–1954

13.4 (−3.3;33.1)

1954–1957

−7.3 (−39.5;42.0)

1957–2000

−1.8 (−2.2;−1.3)

    40–49

−0.9 (−1.2;−0.6)

1950–1961

2.1 (0.6;3.6)

1961–1979

−0.1 (−0.8;0.6)

1979–2000

−2.7 (−3.2;−2.2)

    50–54

−0.7 (−1.0;−0.4)

1950–1961

4.5 (3.1;5.9)

1961–1984

−0.4 (−0.8;0.0)

1984–2000

−3.3 (−3.8;−2.7)

    55–59

−0.4 (−0.8;0.0)

1950–1966

4.3 (3.7;5.0)

1966–1988

−0.7 (−1.1;−0.4)

1988–2000

−4.1 (−4.7;−3.5)

    60–64

0.1 (−0.4;0.7)

1950–1969

5.7 (5.1;6.3)

1969–1993

−1.1 (−1.4;−0.8)

1993–2000

−5.2 (−6.5;−3.9)

    65–69

0.8 (0.2;1.4)

1950–1967

8.2 (7.5;9.0)

1967–1981

1.6 (0.9;2.2)

1981–2000

−2.6 (−2.9;−2.3)

    70–74

1.6 (0.9;2.2)

1950–1969

7.8 (7.2;8.4)

1969–1981

3.6 (2.8;4.4)

1981–2000

−2.3 (−2.6;−2.0)

    75–79

2.4 (1.6;3.1)

1950–1977

7.3 (7.1;7.6)

1977–1986

2.0 (1.0;3.0)

1986–2000

−5.6 (−2.9;−2.3)

    80–84

3.1 (2.3;3.9)

1950–1978

8.0 (7.0;9.0)

1978–1988

3.0 (0.3;5.8)

1988–2000

−2.5 (−3.7;−1.4)

    85+

3.5 (2.8;4.1)

1950–1982

7.2 (6.7;7.7)

1982–1993

1.3 (−0.1;2.7)

1993–2000

−2.5 (−3.7;−1.3)

Age standardized

    Overall (0–85+)

0.9 (0.4;1.4)

1950–1968

5.6 (5.3;5.9)

1968–1984

1.5 (1.2;1.7)

1984–2000

−2.7 (−2.9;−2.5)

    Truncated (30–84)

0.9 (0.4;1.4)

1950–1968

5.6 (5.3;5.9)

1968–1984

1.5 (1.2;1.7)

1984–2000

−2.7 (−2.9;−2.5)

Females

    30–39

2.6 (2.0;3.1)

1950–1981

1.9 (0.6;3.2)

1981–1996

5.5 (2.4;8.8)

1996–2000

−7.7 (−13.8;−1.3)

    40–49

4.7 (4.4;5.0)

1950–1952

−25.0 (−56.4;29.1)

1952–1969

3.0 (0.6;5.5)

1969–2000

5.3 (4.9;5.8)

    50–54

4.8 (4.5;5.1)

1950–1974

2.8 (1.5;4.2)

1974–1981

10.7 (2.4;19.6)

1981–2000

4.1 (3.5;4.8)

    55–59

4.8 (4.4;5.1)

1950–1971

1.1 (−0.4;2.6)

1971–1987

7.9 (6.1;9.8)

1987–2000

3.7 (2.8;4.6)

    60–64

4.6 (4.2;5.0)

1950–1975

0.9 (0.1;1.7)

1975–1992

8.8 (7.7;10.0)

1992–2000

1.7 (0.7;2.8)

    65–69

4.5 (4.1;5.0)

1950–1973

−0.3 (−1.1;0.6)

1973–1996

7.4 (6.7;8.1)

1996–2000

2.9 (1.0;4.8)

    70–74

3.9 (3.4;4.3)

1950–1971

0.4 (−0.6;1.3)

1971–1984

3.8 (1.9;5.8)

1984–2000

6.3 (5.7;6.9)

    75–79

3.2 (2.8;3.5)

1950–1968

−0.2 (−1.6;1.3)

1968–1993

3.3 (2.5;4.1)

1993–2000

6.8 (5.0;8.6)

    80–84

2.8 (2.5;3.0)

1950–1961

6.5 (2.3;10.8)

1961–1966

−4.8 (−20.2;13.6)

1966–2000

3.2 (2.8;3.6)

    85+

1.7 (1.3;2.0)

1950–1978

1.5 (0.4;2.6)

1978–1981

8.8 (−38.8;93.6)

1981–2000

0.6 (−0.5;1.6)

Age standardized

    Overall (0–85+)

4.2 (4.0;4.5)

1950–1973

1.2 (0.8;1.6)

1973–1988

6.4 (5.8;7.1)

1988–2000

4.4 (4.1;4.8)

    Truncated (30–84)

4.3 (4.0;4.5)

1950–1971

0.9 (0.5;1.4)

1971–1993

5.8 (5.5;6.1)

1993–2000

4.1 (3.5;4.6)

Annual adult smoking prevalence and per capita cigarette consumption (age 15+) patterns are shown in Fig. 2 (bottom). Both prevalence and consumption declined for Dutch males with the exception of an increase in consumption during the 1970s. Female cigarette consumption increased over time, but prevalence was consistent at 40% until 1985, and then decreased rapidly before leveling off again at 30%.

Discussion

Even though Belgium and The Netherlands are currently comparable with respect to smoking prevalence and tobacco control interventions, we observed significant temporal differences in lung cancer mortality patterns over the study period between the two countries.

Firstly, overall and age-truncated (30–84) lung cancer death rates among males in The Netherlands declined at a faster rate in the most recent period (1984–2000) compared to Belgium. This decrease began about two decades earlier in the younger age groups in The Netherlands, which resulted in a cohort effect that might explain the faster rate of decrease in the most recent period. The earlier smoking interventions in The Netherlands, e.g., the “Gentlemen’s agreement” about tobacco publicity and health between tobacco companies in 1965, probably explain the more steeply decreasing rates witnessed recently in The Netherlands (Legacy tobacco document, 2501265717/5718 [15]). This favorable trend in lung cancer death rates is comparable to that achieved among males in other European countries who have pursued early interventions for tobacco prevention; for instance Ireland has recently experienced a sustained period of annual decline in rate of death due to lung cancer among males [9]. Furthermore, both countries show a similar lag time between decrease in smoking prevalence and APCs for overall lung cancer mortality rates. The APCs started to level off about on to two decades after the increase in smoking, followed by a decrease in APC about two decades later. This confirms the previously suggested lag of two to three decades between patterns for smoking prevalence and lung cancer mortality rates [1, 11].

Secondly, the magnitude of APCs was similar for men aged 40–49 years in both countries, but only from 1980 onwards these changes in APCs happened approximately at the same time for both countries. More youth-intensive tobacco cessation programs in Belgium may explain this trend [16].

Thirdly, as in many other European countries, overall female lung cancer mortality rates increased for both Belgium and The Netherlands [3, 10]. However, overall female lung cancer death rates have increased much more rapidly in The Netherlands than in Belgium. Interestingly though, the only significant annual percent decline in lung cancer mortality among females was observed among the youngest Dutch women (30–39) since 1996. These findings most likely reflect the patterns of smoking prevalence among Dutch women about 15 years earlier [1, 11]. In The Netherlands, overall smoking prevalence was declining, presumably indicative of the smoking patterns among the younger age group.

Fourthly, female lung cancer death rates in both Belgium and The Netherlands do not follow the general trends for female lung cancer mortality in the European Union, where a 2.6% annual increase was estimated between the early 1970s and the late 1980s, and a 1.4% annual increase thereafter [3]. In Belgium, APCs over the corresponding time periods were 2.3 and 3.3%, respectively, indicating that APCs have yet to level off. In The Netherlands, the corresponding APCs were 6.4 and 4.4%, respectively, suggesting that some leveling off has occurred, but that APCs still remain higher than European average values.

However, anomalous recent increases in APCs were noticed among Belgian women aged 50–59 and among Dutch women aged 40–49. This may predict future mortality trends among middle and elderly age women, and further indicate that smoking cessation interventions need to specifically target middle age women as a unique increase in APCs has been observed in this group. These observations are perhaps less concerning than the observed increasing APCs in lung cancer mortality among the youngest females in Ireland [9]. However, smoking cessation interventions focused at this age group can still have long-term consequences on the burden of tobacco-related diseases in Belgium and The Netherlands.

Finally, regarding smoking trends, an increase in cigarette consumption was observed for women in both the countries, despite declines in smoking prevalence across genders for both Belgium and The Netherlands. During the mid-1990s, however, transient dips in consumption were observed, probably as a result of the smoking advertising bans taking effect around this time. The Belgian government officially banned direct and indirect advertising of tobacco products in 1999, but several other restrictions and bans had already been launched in previous years, such as the previous smoking ban which applied to public transportation (1982). In The Netherlands most interventions took effect a few years earlier: direct advertising was officially banned on TV and radio in 1988, and in cinemas and on billboards in 1996 [7, 14]. In addition, The Netherlands introduced the use of health warnings around the late 1980s, whereas in Belgium this happened around the mid-1990s. As of 1995, both countries implemented the EU directive on labeling [14]. Furthermore, price and price index for a 20 units pack of Marlboro showed a wide variation between Belgium and The Netherlands in April 2000. Marlboro was used as the reference product since this is the dominating cigarette brand in the European Union [17]. In Belgium, the purchasing power parities (PPP) adjusted price was 2.55 Euro with an adjusted price index of 97, whereas, the PPP adjusted price for a 20 units pack of Marlboro was 2.15 Euro and the adjusted price index was 81 [18]. These contrasts in price and price index might indicate why the differences in smoking patterns are not extremely different despite the earlier interventions in The Netherlands. A possible explanation for the overall decrease in smoking prevalence, but simultaneous increase in cigarette consumption, might be derived from relatively high quitting rates among those who smoked fewer numbers of cigarettes. Evidence indicates that quitting rates were lower among the apparently “heavy” smokers compared to lighter smokers [19].

Conclusions

In conclusion, our findings suggest that both recent and past tobacco control efforts seem to have had an impact on smoking prevalence and lung cancer mortality rates, especially among the younger age cohorts in the two European countries studied. In The Netherlands, male lung cancer mortality rates declined faster, and this may be due to anti-tobacco campaigns being introduced a few years earlier in The Netherlands than in Belgium. As a result, the increasing lung cancer mortality rates among females also seem to have leveled off in the last decade in The Netherlands, whereas they have continued to increase in Belgium. Also of interest are anomalous recent annual increases in lung cancer death rates among middle aged women in both countries. In addition, a recent rise in cigarette consumption, despite a simultaneous reduction in smoking prevalence in both sexes was observed. Targeted smoking cessation interventions, particularly in middle-age women, in addition to youth smoking prevention programs, could be an effective strategy to accelerate further declines in lung cancer death rates in Belgium and in The Netherlands.
  • Recent and past tobacco control efforts had impact on smoking prevalence and lung cancer mortality rates, especially among the younger age cohorts in Belgium and The Netherlands.

  • Faster decline in male lung cancer mortality in The Netherlands due to anti-tobacco campaigns being introduced a few years earlier in The Netherlands than in Belgium.

  • Targeted smoking cessation interventions, particularly in middle-age women, in addition to youth smoking prevention programs, could be an effective strategy to accelerate further declines in lung cancer death rates in Belgium and in The Netherlands.

Acknowledgments

The authors want to thank Jacob Sattelmair for his language support.

Copyright information

© Springer Science+Business Media, LLC 2009