Jump to ContentJump to Main Navigation
International Smoking StatisticsA collection of historical data from 30 economically developed countries$

Barbara Forey, Jan Hamling, Peter Lee, and Nicholas Wald

Print publication date: 2002

Print ISBN-13: 9780198508564

Published to Oxford Scholarship Online: September 2009

DOI: 10.1093/acprof:oso/9780198508564.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 25 February 2017

(p.799) Appendix V

(p.799) Appendix V

Source:
International Smoking Statistics
Publisher:
Oxford University Press

Estimating past smoking habits by an indirect method, with application to Great Britain

Nationally-based surveys of smoking habits are now carried out regularly in many countries. However, such surveys existed in relatively few countries before 1970, as has been demonstrated in the chapters of this book. Several researchers have therefore attempted to estimate the prevalence of smoking for earlier periods indirectly, using data on past smoking habits recalled in a recent survey. Based on current age, smoking status (never/former/current), age of starting to smoke and, for former smokers, age of stopping, each subject’s smoking status can be derived for each year throughout life, thus allowing estimation of the prevalence of smoking by calendar year. A list of references to such attempts, relevant to the countries in this book, is given below.

We have carried out such an analysis for Great Britain and compared direct and indirect estimates back to 1948. The results are described briefly in this Appendix, a fuller description, including a discussion of a number of sources of potential bias in the indirect estimates, being available on request.

Two surveys were used as sources of indirect data.

(1) The 1984-1985 Health and Lifestyle Study (HALS) (Cox et al (1987)) of 9 003 subjects aged 18+, was chosen because it was large, representative of the British population, and enquired in moderate detail about smoking. Questions related to manufactured and hand-rolled cigarettes combined, with less detailed information available for pipe and cigar smoking.

(2) The 1978-1981 Alderson Hospital In-Patient Study (AHIP) (Alderson et al (1985)) was chosen because it used an exceptionally detailed questionnaire on manufactured cigarette (MC) smoking history. The 4 950 subjects selected for study were aged 35-74 and hospitalized in England with non-smoking-associated diseases. Questions related to regular smoking, separately for MC, hand-rolled cigarettes, pipes and cigars. Detailed questions on the amounts of MC smoked at up to 10 time points since age 16 allowed for estimates on the level of smoking to be calculated.

Direct estimates were obtained from surveys conducted by the Tobacco Advisory Council (TAC) and predecessor organizations (see Chapter 27, survey source 1).

Table A5.1 presents estimates of prevalence of smoking from the three sources, for selected years. The age groups presented are those most commonly used by TAC. The average difference over the period 1948-1975 between the direct and indirect estimates is also shown. There is generally a good correspondence between the three sets of estimates, with the same time trends evident at all age groups for both sexes, except that the AHIP estimates fall more steeply in the final 5 years than do the other sources (an effect which may be associated with ill health, even though the subjects’ diseases were not smoking associated). For the central age groups (about age 20-60) the HALS and AHIP estimates for cigarette smoking are generally a little higher than TAC for males, but lower for females. For older male smokers of all products TAC estimates are consistently highest.

Table A5.2 presents the average MC smoked per MC smoker from two sources, for selected years. AHIP estimates are almost always higher than TAC estimates except after 1975. However both sources show the male consumption to be higher than the female consumption by a factor of 1.5 to 2, with consumption rising with time in each sex and age group. Similar age patterns are seen at each year for both sources, with the amount smoked increasing up to age 25 and little variation between ages 25 and 60. The AHIP estimates (p.800) decline less at age 60+ than do the TAC estimates, which may in part be an artefact of the different age distributions.

Tables A5.3-A5.5 show the estimates from the two indirect sources by sex and birth cohort going back to 1920.

In the fuller version of this appendix (available on request) we discuss potential sources of bias in indirect estimates, including choice of subjects, definitions of smoking, quit periods, missing data, and inaccurate reporting. We demonstrate that bias due to differential mortality between smokers and nonsmokers (only survivors being able to contribute to the survey) should be minor if attention is limited to subjects aged under 70 at interview. The high level of agreement between the prevalence estimates suggests that most of the sources of potential bias have not in fact had a large influence. Indeed it is surprising that the differences in the definitions of ‘cigarette’ smoking (with hand-rolled cigarettes included only for HALS) has not had a noticeable impact on the estimated cigarette smoking prevalence for older men. Conversely, none of the factors discussed explains why estimates of the level of smoking are higher based on indirect rather than on direct methods.

For both prevalence and level of smoking, the indirect estimates show patterns similar to the direct estimates in respect of sex, age, cohort and time. Indirect estimates therefore represent a useful approach when direct estimates are not available.

References

Bibliography references:

Alderson M.R., Lee P.N. and Wang R. (1985). Risks of lung cancer, chronic bronchitis, ischaemic heart disease, and stroke in relation to type of cigarette smoked. J Epidemiol Community Health, 39, 286-93.

Cox B.D., Blaxter M., Buckle A.L.J., Fenner N.P., et al (1987). The health and lifestyle survey. Preliminary report of a nationwide survey of the physical and mental health, attitudes and lifestyle of a random sample of 9 003 British adults. Health Promotion Research Trust, London.

Published papers presenting indirect estimates or similar analyses

Bibliography references:

Berrino F., Merletti F., Zubiri A., Del Moral A., et al (1988). A comparative study of smoking, drinking and dietary habits in population samples in France, Italy, Spain and Switzerland. II. Tobacco smoking. Rev Epidemiol Sante Publique, 36, 166-76.

Birkett N.J. (1997). Trends in smoking by birth cohort for births between 1940 and 1975: a reconstructed cohort analysis of the 1990 Ontario health survey. Prev Med, 26, 534-41.

Brenner H. (1993). A birth cohort analysis of the smoking epidemic in West Germany. J Epidemiol Community Health, 47, 54-8.

Burns D.M., Lee L., Shen L.Z., Gilpin E., et al (1997). Cigarette smoking behavior in the United States. In: Burns D., Garfinkel L. and Samet J.M. (eds) Changes in cigarette-related disease risks and their implications for prevention and control. USDHHS, NIH, NCI; 13-112. (Smoking and Tobacco Control, Monograph 8.) NIH Publication Number 97-4213.

Degenhardt L., Lynskey M. and Hall W. (2000). Cohort trends in the age of initiation of drug use in Australia. Australian and New Zealand Journal of Public Health, 24, 421-6.

Durston B. and Jamrozik K. (eds) (1990). Tobacco & health 1990. The global war. Proceedings of the Seventh World Conference on Tobacco and Health, Perth, Western Australia, 1-5 April 1990. Organising Committee of the Seventh World Conference on Tobacco and Health, 189 Royal Street, East Perth, Western Australia 6004.

Escobedo L.G. and Peddicord J.P. (1996). Smoking prevalence in US birth cohorts: the influence of gender and education. Am J Public Health, 86, 231-6.

Escobedo L.G. and Peddicord J.P. (1997). Long-term trends in cigarette smoking among young U.S. adults. Addict Behav, 22, 427-30.

Escobedo L.G. and Remington P.L. (1989). Birth cohort analysis of prevalence of cigarette smoking among Hispanics in the United States. JAMA, 261, 66-9.

(p.801)

Ferrence R.G. (1988). Sex differences in cigarette smoking in Canada, 1900-1978: a reconstructed cohort study. Can J Public Health 79, 160-5.

Gilpin EA, Lee L., Evans N. and Pierce J.P. (1994). Smoking initiation rates in adults and minors: United States, 1944-1988. Am J Epidemiol, 140, 535-43.

Harris J.E. (1983). Cigarette smoking among successive birth cohorts of men and women in the United States during 1900-80. J Natl Cancer Inst, 71, 473-9.

Heuer C. and Becker N. (1999). Smoking prevalence and lung cancer mortality in Germany. J Epidemiol Biostat, 4, 45-52.

Hyndman J., Hobbs M., Jamrozik K., Hockey R. and Parsons R. (1990). A retrospective cohort study of smoking habits in Australia. In: Durston B. and Jamrozik K. (eds) Tobacco and health 1990. The global war. Proceedings of the 7th world conference on tobacco and health Perth, Western Australia, 1st-5th April 1990. Organising committee of the Seventh World Conference on Tobacco and Health, 189 Royal Street, East Perth, Western Australia 6004; 264-7.

Jensen O.M. (1975). Lung cancer and smoking in Danish women. Int J Cancer, 15, 954-61.

Johnson R.A. and Gerstein D.R (1998). Initiation of use of alcohol, cigarettes, marijuana, cocaine, and other substances in US birth cohorts since 1919. Am J Public Health, 88, 27-33.

La Vecchia C., Decarli A. and Pagano R. (1986). Prevalence of cigarette smoking among subsequent cohorts of Italian males and females. Prev Med, 15, 606-13.

Laaksonen M., Uutela A., Vartiainen E.and Jousilahti P. (1997). Development of smoking by birth cohort in Eastern Finland 1972-92. In: 10th World Conference on Tobacco or Health Beijing China, 24-28 August 1997. 46 (OS 54)

Launoy G., Milan Cl., Milan Ch., Coudray B., et al (1995). Evolution of tobacco consumption by men in France between 1930 and 1990. In: Slama K. (ed) Tobacco and Health, Ninth World Conference on Tobacco and Health, Paris, Oct 10-14 1994. Plenum Press, New York and London; 711-3.

Nordlund L.A (1998). Trends in smoking habits and lung cancer in Sweden. Eur J Cancer Prev, 7, 109-16.

Pelletier F., Marcil-Gratton N. and Légaré J. (1996). A cohort approach to tobacco use and mortality: the case of Quebec. Prev Med, 25, 730-40.

Rønneberg A, Hafstad A. and Lund K.E. (1994a). Røykevaner siden 1910 blant norske menu og kvinner fodt etter 1890 (Smoking habits since 1910 among Norwegian men and women born after 1890). Tidsskr Nor Laegeforen, 114, 1623-6.

Rønneberg A., Lund K.E. and Hafstad A. (1994b). Lifetime smoking habits among Norwegian men and women born between 1890 and 1974. Int J Epidemiol, 23, 267-76.

Tolley H.D., Crane L. and Shipley N. (1991). Smoking prevalence and lung cancer death rates. In: Strategies to control tobacco use in the United States: a blueprint for public health action in the 1990’s. US Dept Health & Human Services, Public Health Service, NIH, Bethesda, Md;73-126. Smoking and Tobacco Control Monograph 1. NIH publication 92-3316.

Weinkam J.J.and Sterling T.D. (1990). Age related changes in age of starting to smoke. J Clin Epidemiol, 43, 133-40.

(p.802)

Table A5.1 Prevalence of smoking selected years from different sources

1950

1960

1970

Average difference1 1948-1975

Sex

Age group2

TAC

HALS

AHIP

TAC

HALS

AHIP

TAC

HALS

AHIP

HALS-TAC

AHIP-TAC

Males

16-19

%C

51

44

49

65

56

49

55

53

-1.9

-4.13

%A

52

45

52

65

56

52

56

53

-3.2

-2.73

20-24

%C

68

73

71

67

60

63

58

61

2.1

2.44

%A

71

74

76

70

61

69

65

62

-1.1

3.24

25-34

%C

70

77

72

64

69

68

60

64

60

6.5

4

%A

79

81

81

73

72

75

70

67

65

0.5

1.6

35-59

%C

66

70

64

64

67

62

55

58

57

4.2

0.6

%A

81

78

76

78

74

76

71

65

70

-4.3

-1.7

60+

%C

46

45

46

47

39

0.65

-4.26

%A

69

64

68

60

57

-8.55

-7.66

Females

16-19

%C

36

26

29

45

34

36

52

42

-5.7

-1.83

20-24

%C

48

46

55

48

45

45

54

51

-1.9

0.14

25-34

%C

53

47

53

53

50

53

51

48

51

-2.4

0.6

35-59

%C

38

34

38

47

41

46

50

45

48

-6.5

-2

60+

%C

22

16

26

21

27

-5.85

0.66

%C - Prevalence of MC smoking (TAC and AHIP) or cigarette smoking (HALS)

%A - Prevalence of smoking any product

(1) Calculated from estimates as whole numbers, no greater precision being available for TAC

(2) For the indirect estimates, the base is all subjects falling within the age group at that year, but these subjects may come from a more limited age range. Estimates based on less than 20 subjects have been omitted. For instance, in 1970, the AHIP subjects were aged 24-68, so that the group headed 60+ actually contains subjects 60-68, and the age group 20-24 has been omitted as it contained only 4 male and 2 female 24-years olds.

(3) 1948-1963

(4) 1948-1968

(5) 1960-1975

(6) 1965-1975

Table A5.2 Average number of MC smoked per smoker at selected years from different sources

1950

1960

1970

Sex

Age group1

TAC

AHIP

TAC

AHIP

TAC

AHIP

Males

16-19

9.7

14.6

13

15.7

20-24

14.1

19.1

17.9

18.1

25-29

19.9

17.7

19.6

19.6

19.5

24.9

30-34

18.6

23.2

19.9

22.4

35-49

21.2

21.9

20.8

24.3

50-59

21

21.3

19.9

22.9

60+

15.8

21.1

Females

16-19

5.2

10

7.3

10.3

20-24

7.8

10.6

9.6

12.3

25-29

7.4

11.8

9.5

14

14.5

15.3

30-34

11.9

13

13.7

15.3

35-49

12.2

13.9

14.7

15.7

50-59

10.7

12.4

14.2

15.3

60+

10.1

13.9

(1) See footnote 2 to table A5.1

(p.803)

Table A5.3 HALS Study. Prevalence (%) of smoking at selected years, by sex and birth cohort

Year of birth

Year

1920

1925

1930

1935

1940

1945

1950

1955

1960

1965

1970

1975

1980

1985

Male - cigarette smoking

1890-1899

47

47

47

47

47

40

40

40

40

33

13

13

13

7

1900-1909

33

62

75

76

75

73

68

65

61

55

48

38

29

19

1910-1919

1*

13

41

63

75

76

74

69

64

60

53

48

38

28

1920-1929

1

11

40

71

76

76

73

70

61

53

46

32

1930-1939

1

10

31

58

63

63

57

52

46

36

1940-1949

1

9

36

63

66

58

49

40

1950-1959

1

10

36

53

50

39

1960-1969

1

11

35

36

Male - smoking of any product

1890-1899

73

67

67

67

67

60

60

60

53

47

33

33

33

20

1900-1909

35

67

80

81

82

81

77

75

71

65

60

50

42

31

1910-1919

1

13

42

65

77

81

79

75

71

69

64

60

49

37

1920-1929

1

12

41

73

79

79

77

74

67

59

54

40

1930-1939

2

11

32

60

65

66

62

58

53

44

1940-1949

1

9

36

64

69

63

55

46

1950-1959

1

10

36

54

51

41

1960-1969

1

11

35

36

Female - cigarette smoking

1890-1899

15

19

19

19

19

23

21

21

17

15

13

13

13

8

1900-909

3

10

19

23

28

29

30

28

26

24

20

17

14

10

1910-1919

0

1

8

23

34

38

39

39

39

39

36

32

26

21

1920-1929

0

2

19

46

51

53

52

52

50

49

44

36

1930-1939

0

3

16

36

47

48

48

46

43

34

1940-1949

0

2

19

44

49

45

39

33

1950-1959

0

4

28

46

45

35

1960-1969

0

9

35

36

(*) This group of smokers comprises 2 subjects born in 1910 and starting at ages 8 and 9, 1 subject born in 1911 starting at age 7 and 2 subjects born in 1912 starting at ages 6 and 8.

Table A5.4 AHIP study. prevalence (%) of smoking at selected years, by sex and birth cohort

Year of birth

Year

1920

1925

1930

1935

1940

1945

1950

1955

1960

1965

1970

1975

19801

Male-MC smoking

1900-1909

13

48

63

66

67

67

64

59

53

45

38

30

16

1910-1919

0

7

33

55

65

68

64

63

60

55

50

43

28

1920-1929

1

8

39

69

74

72

69

65

60

51

39

1930-1939

1

8

36

60

66

64

61

53

43

1940-1949

2

13

54

64

63

48

39

Male-smoking of any product

1900-1909

14

53

72

77

78

80

77

77

72

56

47

47

32

1910-1919

0

7

36

61

74

77

75

75

73

70

65

57

40

1920-1929

1

9

43

76

82

81

79

76

72

67

51

1930-1939

1

8

38

64

72

71

69

64

53

1940-1949

2

14

57

69

69

57

50

Females - MC smoking

1900-1909

1

9

17

20

25

30

31

30

30

29

24

19

13

1910-1919

0

1

8

26

37

45

45

45

45

45

41

35

23

1920-1929

0

3

22

47

55

57

57

56

54

49

37

1930-1939

0

4

22

43

47

46

49

45

39

1940-1949

1

1

34

50

52

47

35

(1) Estimates for 1980 are generally based on fewer subjects as they exclude subjects hospitalized before 1980

(p.804)

Table A5.5 AHIP Study. Average number of MC per smoker at selected years, by sex and birta cohort

Year of birth

Year

1920

1925

1930

1935

1940

1945

1950

1955

1960

1965

1970

1975

1980

Male

1900-1909

9.2

12.0

16.4

17.6

21.2

20.7

21.5

20.3

17.7

1910-1919

7.9

11.7

15.1

17.7

18.7

22.1

22.2

22.2

21.1

18.8

1920-1929

10.4

12.9

17.1

20.0

20.4

22.1

23.5

24.1

25.1

23

1930-1939

12.1

13.9

17.5

19.7

23.5

24.4

25.5

24.6

1940-1949

10.4

15.1

19.2

23.6

25.0

24.1

Female

1900-1909

5.9

7.0

9.2

10.0

12.8

14.1

14.4

13.3

12.4s

1910-1919

5.3

7.3

8.6

10.5

12.1

13.8

14.7

15.0

15.0

14.1

1920-1929

7.8

9.5

10.9

11.3

11.1

13.5

14.8

15.3

16.3

15.9

1930-1939

6.8

9.6

12.0

13.4

15.5

16.2

17.9

16.8

1940-1949

7.6

9.8

13.8

15.7

18.7

17.5