hilary graham university of york, uk · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70...

37
Tobacco control Hilary Graham University of York, UK WHO Global Health Histories Seminar March 23 2011

Upload: others

Post on 24-Aug-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Tobacco control

Hilary Graham University of York, UK

WHO Global Health Histories Seminar March 23 2011

Page 2: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

1.

global health history of tobacco use2.

…with an inequalities slant (how cig smoking follows the contours of socioeconomic inequalities within countries)

3.

implications for tobacco control policy

•Draw on UK examples that highlight issues for tobacco control globally

•Every society has its own unique history of tobacco use

Page 3: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

1. global health history of tobacco use

Page 4: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

• backed by advertising, manufactured cigarettes displace traditional patterns of tobacco use, fuelling rapid rise in consumption -

evident first in high-income countries, then

globally

• common pattern is for existing tobacco users to switch & increase consumption, and new users to take up smoking

• lagged health effects of smoking: decades later a sharp rise in smoking-related disease & death to become ‘the single most preventable cause of death in the world’

WHO 2008

• smoking epidemic tackled through TC policies which drive down prevalence, particularly when integrated into a comprehensive strategy (FCTC)

• run through these stages…

Page 5: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

In early industrialising countries (US, UK,Netherlands..)

• late 19th

century, new tobacco product (‘milder’

& more addictive) & new technology (machine- made cigs -

larger scale, lower cost: 200 cigs per

min by machine/200 per hr by hand)• created a market commodity: a cheap

standardised product looking for a mass market to sell to…

Page 6: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

backed by advertising, the introduction of manufactured cigarettes displaces traditional patterns of tobacco use…

Advert for Pall Mall, 1916

Page 7: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

20

40

60

80

100

120

1870 80 90 1900 10 20 30 40 50 60 70 80 90

'000

s of

tonn

es

other products manufactured cigarettes

Annual sales of manufactured cigarettes + all other tobacco products (in thousand tonnes of manufactured weight) UK 1870-

1990

introduction of manufactured cigarettes displaces traditional patterns of tobacco use, fuelling rapid rise in consumption

Source: Wald and Nicolaides-Bouman, 1991

Page 8: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Annual sales of manufactured cigarettes to men & women (in thousand tonnes of manufactured weight) UK 1870-1990

Source: Wald and Nicolaides-Bouman, 1991

0

10

20

30

40

50

60

70

80

1870 80 90 1900 10 20 30 40 50 60 70 80 90

'000

s of

tonn

es

men - cigs women -cigs

typical pattern is for existing tobacco users to switch & increase consumption, and new users to take up smoking

1928

Page 9: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Annual death rates per 100,000 from lung cancer, 55-74yrs, UK 1950-2000

Men

Women

Source: Peto et al. 2000

lagged health effects: decades later, a sharp rise in smoking-related disease & death

Page 10: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

40

50

1980

1984

1988

1992

1996

2000

2004

2008

men women

Prevalence of cigarette smoking among men and women, England, 1980-2008

Source: Robinson and Bulger, 2010

smoking epidemic tackled through tobacco control policies which drive down prevalence,particularly when integrated into a comprehensive strategy

Page 11: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

2. global health history of tobacco use …with an inequalities slant (how cig

smoking follows the contours of socioeconomic inequalities within

countries)

part of the global history but harder to evidence & more challenging toaddress –

how over time & across place, cigarette smoking becomesentrenched as a habit of the poor

Page 12: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Lopez et al: model of ‘the smoking epidemic’

Younger adults in advantaged groups (men, city-living, well-off) take up manufactured cigs first (symbol of power & prestige)

Then taken up by younger women & poorer groups →

overall prevalence increases (no longer mark of social distinction)

Prevalence falls first in advantaged groups

As overall rates fall, smoking is increasingly concentrated in disadvantaged groups

Page 13: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

• a model only! • based on high-income market economies in 20th

century

(companies competing to ‘grow the market’

for cigarettes in ways that increases their

market share)• & not middle and low-income countries with

cultural & religious practices, undergoing rapid urbanisation & opening up economies to global companies, including tobacco companies

• run through stages of ‘smoking epidemic’

Page 14: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Advantaged groups (men, city-living, well-off) take up manufactured cigs first

….illustrate with patterns of smoking (traditional products and manufactured cigarettes) among men by educational level in Mumbai in early 1990s

Page 15: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Cigarette advertising in Mumbai in 1990s was ‘ubiquitous and associated smoking with aspiration ..’

Bansal et al 2005

Traditional –bidis

New-

manufactured cigarettes

Page 16: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

bidi smokers cig smokers

%

none primary middle secondary college

Prevalence of smoking among men by educational level, Mumbai, 1992-4

Source: Sorensen et al, 2005

Page 17: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Then taken up by younger women & poorer groups → overall prevalence

increases (loses its association with social distinction)

….illustrate with patterns of cigarette smoking among women by educational level in Spain in 1980s

Page 18: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

• Strong cultural prohibitions against female tobacco use in Spain until 1970s

• When widening opportunities for women post Franco regime coincided with transnational tobacco companies gaining entry into the Spanish cigarette market

• Spending on cig advertising increased 8fold from 1982-88, predominantly through TV advertising

Page 19: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

40

50

60

70

16-24 25-44 45-64 65+

% s

mok

ers

In Spain by late 1980s, 55% of men and 23% of women smoked cigarettes -

among women, marked age

differences…

Source: De Onis & Villar, 1991

Cigarette smoking among women by age Spain 1987

It was younger women (<45yrs) who led the way –

rates very low among older women

Page 20: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

40

50

60

70

lowest 2nd 3rd highest

%

0

10

20

30

40

50

60

70

lowest 2nd 3rd highest

%

0

10

20

30

40

50

60

70

lowest 2nd 3rd highest

% s

mok

ers 16-24 yrs

25-44yrs

45-64yrs

Cigarette smoking among women by age & educational level Spain 1987

Source: De Onis & Villar, 1991

….younger women from advantaged backgrounds led the way….but prevalence quickly rose among women from less advantaged backgrounds (by mid-1990s, social gradient had reversed)

Page 21: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Smoking prevalence among female rural-to-urban migrants in Beijing, China, by educational level, 2002 (average age: 22yrs)

0

5

10

15

20

% s

mok

ers

primary middle high post-school

Source: Chen et al, 2004

China –

low rates of female smoking (≤4%) but will it avoid history repeating itself? Rural-to-urban migrants may capture trends that took decades to unfold in high-income countries

(rates among men over 50% -

with highest rates in lowest educational group)

Page 22: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Prevalence tends to fall first in advantaged groups -

as overall rates fall,

smoking is increasingly concentrated in disadvantaged groups

….illustrate with patterns of cigarette smoking in UK among women from 1940s (longest run of historical data)

Page 23: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Prevalence of cigarette smoking among women in highest & lowest socio-economic groups, 1948-68

0

10

20

30

40

50

1948 1958 1968 1978 1988 1998

% s

mok

ers

highest lowest

1961

1961

Source: Wald and Nicolaides-Bouman, 1991

Page 24: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Prevalence of cig smoking among women in the lowest socioeconomic group, 1948-2004

0

10

20

30

40

50

1948 1958 1968 1978 1988 1998

% s

mok

ers

highest lowest

Source: Wald and Nicolaides-Bouman, 1991

Page 25: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Prevalence of cig smoking among women in highest & lowest socioeconomic groups, 1948-2004

0

10

20

30

40

50

1948 1958 1968 1978 1988 1998

% s

mok

ers

highest lowest

2009

Source: Wald and Nicolaides-Bouman, 1991

Page 26: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

3.

implications for tobacco control policy

Page 27: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

WHO Framework Convention on Tobacco Control

‘The Parties to this convention

determined to give priority to their right to protect health, recognising that the spread of the tobacco epidemic is a global problem (and)….seriously concerned about the impact on the poor’

(Preamble)

The global poor –

and poorer groups within countries

Page 28: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

• Limited evidence on impact of policies on social inequalities in smoking

• What does the evidence suggest?

Page 29: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Weak policies -

widening inequalities in smoking?

• Evidence from patterns across high-income societies (Europe, North America etc)

• UK as example -

prevalence of cig smoking among men in highest & lowest socioeconomic groups, 1948-2004

Page 30: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

40

50

60

70

80

1948 1958 1968 1978 1988 1998

% s

mok

ers

highest lowest

Source: Wald and Nicolaides-Bouman, 1991; ONS, 2001; Berridge, 1997; Cairney, 2007

1980-90s: policies slowly strengthened

1950: Doll & Hill, Smoking and carcinoma of the lung, BMJ

1965: ban on tv cig adverts before 9.0pm

1969: first major antismoking campaign

1970s: begingings..1st tv campaign;1st smoking in pregnancycampaign; 1st health warningson packs

Cigarette smoking among men in highest & lowest socioeconomic groups, 1948-2004

Page 31: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Suggests…

• socioeconomic inequalities will widen in the absence of strong and co-ordinated TC policies –

• strong, comprehensive policies are needed both to reduce overall rates and to narrow inequalities in smoking

• such TCPs now in place in UK -

ranked 1st

in

Europe in 2007 for its TCP

Page 32: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

UK policies over last decade:

• Cig tax increases (1997-2010)• Bans on smoking in enclosed public places (introduced 2006-7)• Smoking Cessation Services (from 1999; by 2003, across UK)• Consumer information (picture warnings from 2008)• Ad bans extended to branded goods & sport sponsorship (from 2005)• Picture pack warnings introduced (2008 -

graphic pictures of rotting lungs, throat cancer, bodies in morgue)

Strong policies –

narrowing inequalities in smoking?

(we might expect prevalence to continue to fall in all socioeconomic groups –

but to fall faster in poorer groups?)

Page 33: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

0

10

20

30

40

2003 2008 2003 2008

highest 2nd 3rd 4th lowest

Prevalence of cigarette smoking among men and women by income quintile, England 2003 and 2008

Women

Men

Source: Craig et al 2009

inequalities in smoking widened across a period of stronger TCP

rates fell – except for poorest groups, where they rose

Page 34: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Suggests that strong, comprehensive policies may be necessary but not sufficient to narrow inequalities in smoking –

it may be important to address wider inequalities in

people’s lives

In UK, inequalities in the living standards of poorer and richer households have widened dramatically since the 1980s –

and are continuing to widen. Similar patterns in the US and in many low and middle income countries, including China and India

Page 35: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

‘Tobacco use disproportionately affects lower socioeconomic groups in developed and developing countries…Efforts to prevent and control tobacco consumption among disadvantaged groups are not likely to succeed other than through an integrated approach that seeks to reduce underlying social inequities’

WHO 2010 ¹

¹

Blas & Sivasankarakarup (eds) Equity, Social Determinants & Public Health Programmes

Page 36: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Conclusions• global history of social inequalities in tobacco use is

important area to investigate, with insights for TCP • declining prevalence has historically been associated with

widening inequalities in smoking -

and with weak tobacco control policies

• but, on their own, strong policies may not be enough to reduce inequalities in smoking

• suggests that TCP need to be embedded in wider policies to reduce inequalities in life chances & living standards

Page 37: Hilary Graham University of York, UK · 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % 0 10 20 30 40 50 60 70 lowest 2nd 3rd highest % smoker s

Thank you