Effective Youth Smoking Prevention: Myths, Realities and Uncertainties

1
Effective Youth Smoking Prevention Myths,
Realities and Uncertainties

• Paul McDonald, Ph.D.
• University of Waterloo
• Ontario Tobacco Research Unit

2
Do you know the answer to these questions?

• Are youth smoking rates on the rise in Canada?
• When do youth start experimenting with smoking?
• Is peer pressure an important determinant of
  youth smoking?
• Are school-based educational interventions
  effective in reducing the prevalence of youth
  smoking?
• Which of the following public polices are
  effective in reducing youth smoking rates?
• Restrictions on smoking in public places
• Bans on smoking on school property
• Restrictions on access to tobacco products
• Restrictions on tobacco advertising

3
Do you know the answers to these questions?

• What proportion of adolescent smokers develop
  nicotine dependence?
• What types of smoking cessation programs help
  young smokers to quit?
• What are some of the new frontiers in smoking
  prevention?
• What should communities do to prevent youth from
  smoking?

4
Effective Youth Smoking Prevention Myth,
Reality and Uncertainty

• Paul McDonald, Ph.D.
• University of Waterloo
• Ontario Tobacco Research Unit

5
Is youth smoking on the rise in Canada?

• Yes
• No
• It depends

6
Is youth smoking on the rise in Canada?

• Yes
• No (it may actually be starting to decline among
  younger age groups)
• It depends

7
Youth smoking rates in Canada, 1979 to 2000
who currently smoke
Year
8
So What?

• Its important to know what the secular trends
  are, in order to properly evaluate our efforts
• Rates started to increase before a series of
  important setbacks (e.g. federal/prov. Tax
  rollbacks)
• Rates in Canada mirrored those in out developed
  countries (Australia, US, Norway, UK, Denmark,
  etc)
• Ecomonics may play a very large role in youth
  smoking rates, since youth are most sensitive to
  stress, unemployment rates, prospects for the
  future, etc.

9
Comparison of Poverty and Smoking Rates in
Canadian Youth
children in low income families
youth smoking
child poverty
Year
Source Schissel, 1997
10
Pauls Prediction for Youth Smoking Rates in
Canada, 1979 to 2005
who currently smoke
Year
11
What percentage of youth have tried smoking at
least once, by the end of high school?

• 40
• 50
• 60
• 70
• 80
• 90

12
What percentage of youth have tried smoking at
least once, by the end of high school?

• 40
• 50
• 60
• Approx. 70
• 80
• 90

13
When do youth first start to experiment with
smoking?

• A. Mostly before grade 6
• B. Mostly during grade 6 and 7
• C. Mostly during grade 8 and 9
• D. Mostly during grade 10 and 11
• E. There is fairly consistent uptake from grade
  6 to 12

14
When do youth first start to experiment with
smoking?

• A. Mostly before grade 6
• B. Mostly during grade 6 and 7
• C. Mostly during grade 8 and 9
• D. Mostly during grade 10 and 11
• E. There is fairly consistent uptake from grade
  6 to 12

15
When do youth first start to experiment?
Cumulative Per cent of Youth Who Try Smoking, by
the End of Each Grade
Cumulative who have tried smoking
Grade
Driezen et al, submitted
16
When are experimenters most likely to become
regular users?

• Before grade 6
• During grade 6 and 7
• During grade 8 and 9
• During grade 10 and 11
• During grade 12
• Fairly evenly from grade 6 to 12

17
When are experimenters most likely to become
regular users?

• Before grade 6
• During grade 6 and 7
• During grade 8 and 9
• During grade 10 and 11
• During grade 12
• Fairly evenly from grade 6 to 12

18
When do youth become regular smokers?
of regular grade 12 smokers who began
smoking
Grade
Driezen et al, submitted
19
So What?

• Experimentation with smoking is normal
• Most kids who try smoking dont become regular
  smokers
• Transition from trying to experimenting to to
  regular smoking is slow and occurs across grade 6
  to 12
• Quitting is difficult (but not impossible) once
  youth become regular smokers
• Therefore
• Prevention efforts dont all need to be focused
  at grade 6
• Perhaps more effort should be focused on
  preventing transition to regular smoking rather
  than preventing experimentation

20
Is peer pressure a major reason why youth start
to smoke?

• Yes
• No
• It depends

21
Is peer pressure a major reason why youth start
to smoke?

• Yes
• No
• It depends

22
Peer Pressure Revisited

• Peers model smoking behaviour
• Kids more likely to smoke if their peers smoke or
  if they go to a school with high prevalence rates
  
• Equivocal evidence that peer groups put pressure
  on kids to smoke
• Kids may be more likely to smoke if they have
  smoking peers because tobacco is more accessible
• Direction of the influence may be backwards,
  especially in high school
• Different identifiable groups have different
  smoking rates (skaters, jocks, arties, nerds,
  misfits, metal heads, etc.)
• Kids may smoke to be accepted in a desired group
  that has a perceived norm around smoking

23
Think in Terms of Social Influences, Not Peer
Pressure

• Positive reinforcers to smoke
• Adult and peer role models
• Misplaced social attributions
• Tobacco advertising (better lifestyle,
  independence, etc.)
• Smoking cues (e.g., smell of smoke)
• Negative reinforcers to smoke
• Difficulty getting access to tobacco
• Limitations where you can smoke (with penalties
  for violation)
• Clear anti-smoking norms
• Messages about the costs of smoking (health,
  social, economic)

24
Are school-based education interventions
effective in reducing the prevalence of youth
smoking?

• Yes
• No
• It depends

25
Are school-based education interventions
effective in reducing the prevalence of youth
smoking?

• Yes
• No
• It depends

26
Are School Education Programs Effective?

• Approaches based on information deficit or
  affective education do not reduce prevalence
• Info on health risks
• Enhancing self esteem
• Influence beliefs and intentions about tobacco
• Stress management

27
Are School Education Programs Effective?

• Social influences approach involves understanding
  and developing skills to counter social
  influences such as family, friends, media,
  tobacco companies.
• Social influences approaches can reduce smoking
  by 5 to 30 for up to two years
• Booster sessions can extend delay to 4 years
  (e.., grade 6 to grade 10)
• Education approaches may be particularly
  effective with some schools
• This may explain why the recent Hutchinson study
  failed to find significant effects

28
Interaction of Education and School Environment
Odds that a grade 6 never smoker will become a
Smoker by the End of grade 8
Prevalence of grade 8 smoking
Cameron et al, 1999
29
What Else Improves Education-Based Outcomes?

• Interactive rather than didactic presentation
• Involvement of older peers
• Minimum of 4 hours, but up to 10 hours is better
• Doesnt matter if adult leader is a teacher or a
  nurse or how training is provided
• Programs should aim for flexibility, not fidelity
  with rigid protocols (I.e., encourage local
  adaptation)

30
Are School Education Programs Cost Effective?

• Wang et al. (2001) spent 16,400 to reach 11200
  grade 7 students with 10 lessons
• Cost effectiveness was US 13,316/life year saved
  or US 8482/QALY
• Cost effectiveness of other interventions
• PAP screening or women 15-74 US 22,000/life
  year saved
• Breast screening for women 50-69 US
  46,000/life yr saved
• Nicotine patch brief phys. Counsel 965-
  2360/life yr saved

31
Summary

• School programs that include a social influences
  approach can reduce smoking in high risk schools
  and delay smoking in most types of schools
• Reductions are greater when curriculum is
  supplemented by no smoking school policies and
  mass media campaigns
• Interventions should include
• Information on social norms and influences and
  how to counter them
• Information on the short and long term effects of
  smoking
• Training for teachers and peers

32
Which of the following smoking restriction
policies have been shown to reduce youth smoking
rates?

• Smoking bans in public places
• Smoking bans in schools
• Smoking bans in homes
• Both A and B but not C
• Both B and C but not A
• A, B and C
• None of the above

33
Which of the following smoking restriction
policies have been show to reduce youth smoking
rates?

• Smoking bans in public places
• Smoking bans in schools
• Smoking bans in homes
• Both A and B but not C
• Both B and C but not A
• A, B and C
• None of the above

34
Effect of No-Smoking Policies on Youth Smoking

• Wakefield et al (2000) surveyed 17,287 students
  in 202 US high schools
• Odds of 30 day prevalence in locations with
  restrictions compared to no restrictions
• Total home ban .79 (.67 - .91)
• Partial home restrictions .85 (.74 - .95)
• Public places .91 (.83 - .99)
• Enforced school ban .86 (.77 - .94)
• School ban (no enforce) .99 (.85, 1.13)

35

• Effect of restrictions on stages of smoking
  transition (stage compared to prev. stage)
• Early exper. Regular smoker
• Public places .93 (.84 1.02) .90 (.81 -.98)
• Total home ban .69 (.59 - .79) .78 (.67 -
  .90)
• Some home ban .83 (.74 - .92) .83 (.74 - .92)
• School ban .98 (.85 1.10) 1.22
  (1.07 1.37)
• Enforced school ban .89 (.85 - .99) .89 (.85
  - .99)
• Signficantly inhibits transition to next stage
  of smoking
• Facilitates transition to regular smoking

Wakefield et al, 2000
36
Do Smoking Bans Reduce Prevalence or Does Reduced
Prevalence Make it Easy to Get a Smoking Ban?

• Probably both are true, but to be of value, we
  must demonstrate that smoking bans reduce
  prevalence of smoking
• No longitudinal studies to date
• Moore et al (2001) did multi level cross
  sectional study with 1375 grade 11 students and
  55 teachers from 55 high schools in Wales

37
Prevalence of Smoking, by Level of School Smoking
Policy
f grade 11 students who smoked daily

• Differences were statistically significant after
  adjusting for parental and
• peers smoking, student attitudes, parental
  expectations, school achievement
• - Rates of occasional smoking also differed, but
  not as much as with daily smokers

Moore et al, 2001
38
Ways to Increase the Effect of Smoking
Restrictions

• Let people know that home, school and public
  place restrictions work
• Enforce the policies
• Get broad support from stakeholders
• Make sure stakeholders know precisely what the
  policies and penalties are
• With youth, school suspension may only increase
  rebelliousness and be worn as a badge of pride

39
Does restricting youth access to retail tobacco
products reduce youth smoking prevalence?

• Yes
• No
• It depends

40
Does restricting youth access to retail tobacco
products reduce youth smoking prevalence?

• Yes
• No
• It depends

41
Does Reducing Access to Tobacco Reduce Youth
Smoking?

• Programs may reduce retail sales to minors but
  generally do not smoking prevalence because of
  social distribution of tobacco products (friends,
  family, strangers, theft)
• Most youth do not believe restrictions impede
  their ability to obtain tobacco products
• Trials are more likely to significantly reduce
  youth prevalence if
• There is regular and sustained enforcement (e.g.,
  min. of 2 checks per vendor per year)
• Penalties are graduated (e.g., lower for first
  time offenders)
• There is broad community support for restrictions
• There is a simultaneous campaign to reduce
  distribution of cigarettes through social
  channels (adults, peers)
• Adjacent communities also have similar
  restrictions that are enforced

42
Vendor Compliance with Sales Restrictions to
Minors In Ontario
who complied
Ontario Tobacco Control Act Introduced
OTRU, 2001
43
Have Tobacco Industry Programs to Restrict Sales
to Minors Worked?

• Tobacco Institutes Its the Law Campaign
• Only 5 of tobacco retailers participated in the
  program
• No difference between participants and
  non-participants in their willingness to sell to
  minors 86 vs. 88 (DiFranza et al. 1992)
• Tobacco companies have found new ways to
  distribute their product (internet)

44
Does restricting tobacco advertising reduce the
prevalence of youth smoking?

• Yes
• No
• It depends

45
Effects of Tobacco Advertising and Restrictions

• Youth smoking is related to advertising
• Smoking rates increased as industry advertising
  increased (particularly those aimed at young
  women) (Pierce et al, 1995)
• Youth smoke the most heavily advertised brands
• Youth who own a tobacco promotional item are gt 2
  x more likely to start smoking in next three
  years (Pierce et al, 1998)
• Results of ad restrictions have been equivocal
• May be explained because part and complete bans
  have diferential effects
• Complete bans may reduce youth smoking by up to
  5 (Safer and Chaloupka, 1999)

46

• Total bans would eliminate point of purchase
  displays, sponsorships, promotional give-aways,
  internet sales, etc.
• Do any of these things happen in your community?

47
What proportion of high school smokers report
they have quit smoking in the past year?

• 3
• 5
• 10
• 15
• 20
• 25
• It depends

48
What proportion of high school smokers report
they have quit smoking in the past year?

• 3
• 5
• 10
• 15
• 20
• 25
• It depends

49
Comparison of Adolescent Quit Rates Using Various
CriteriaWaterloo Smoking Prevention Projects,
Study 4
of Quitters
(15.4 - 17.8)
(2.7-3.3)
(2.3-2.8)
(1.9-2.3)
Column 12 No smk in past 7 days.
Column 12 No smk in past 30 days
Ever smoked - y Usually smk every wk No, I quit
Column 1 smked 100 cigs. lifetime
McDonald Driezen, unpublished
50
What percentage of young smokers will develop
nicotine dependency?

• 0
• 15
• 30
• 50
• 75
• gt90

51
What percentage of young smokers will develop
nicotine dependency?

• 0
• 15
• 30
• 50
• 75
• gt90

52
Prevalence of Nicotine Dependence in the US, by
Age and Sex
of last month smokers
Age
Kandel Chen, 2000 data from the 1993 US
National Household Survey on Drug Abuse
53
Which of the following interventions have been
proven to help adolescents quit smoking?

• Nicotine patches brief physician counselling
  self help materials
• Brief physician interventions
• Group self help
• All of the above
• None of the above
• We dont know

54
Which of the following interventions have been
proven to help adolescents quit smoking?

• Nicotine patches brief physician counselling
  self help materials
• Brief physician interventions
• Self help programs (e.g., Quit 4 life)
• All of the above
• None of the above

55
Youth Collaborative Best Practices Guideline for
Youth Cessation

• Joint initiative of CDC, CTCRI, RWJ, NCI, etc.
• Expert panel still meeting
• Preliminary conclusions are that it is possible
  to help adolescents quit
• Best methods, theories, sites of intervention are
  undetermined at the present time
• Effective interventions tend to be intensive and
  have multiple components
• Confidentiality is essential

56
What Actions Can Communities Take to Prevent
Tobacco Use?

• What are your suggestions based on what you heard
  today?

57
What Actions Can Communities Take to Prevent
Tobacco Use?

• Implement and support school education in high
  risk schools using social influence model
• Create supportive environments by advocating for
  smoking bans on school property (and while on
  school business), smoke free public places
• Create broad base public support
• Develop a campaign to encourage smoke free homes
  for kids
• Provide adequate cessation programs for adults
  who want to be good role models
• Enforce youth access laws and build public
  support limiting distribution through non-retail
  channels

58
What Can Communities Do?

• Advocate for complete advertising restrictions
  (e.g., point of purchase displays)
• Advocate for tax increases and for tax revenue
  from sales to minors to be re-invested for youth
  prevention
• Closely monitor local youth smoking rates in
  schools and communities and provide feedback to
  all stakeholders
• Develop partnerships with researchers to shape
  the research agenda, collect and share
  standardized data, and enhance evaluation

59
New Frontiers in Prevention

• Post secondary students
• Smoking among university students has increased
  from 22 in 1993 to28 in 1999
• 10 of university students started smoking after
  leaving high school
• 11 of university smokers started smoking
  regularly after leaving high school
• Few interventions to date have been designed for
  post secondary students
• The effects of smoking restrictions are unknown

60
New Frontiers in Prevention

• Harm reduction
• Hamilton et al (2000) interviewed 250 smokers in
  grades 8, 9 and 10 from Australia
• 50 had tried to quit while another 20 had tried
  to reduce their smoking
• Half belief that adolescent smokers try to reduce
  their smoking
• 40 of daily and occasional smokers said that
  messages to stop smoking made them feel
  rebellious and produced a desire to smoke more
• positive toward two types of messages

Reg smk Occas
smk Exp Dont smoke 12 24 30 Smoke
less 57 63 71
61
New Frontiers in Prevention

• Local community monitoring and feedback
• Computer-based interventions
• Low cost intervention can be tailored youth are
  receptive
• Focus on preventing the transition to regular
  smoking, not experimentation
• Focus on the determinants of smoking (e.g.,
  marginalization, social cohesion, rebelliousness)
• Smoking is highly correlated with other risk
  behaviours (drinking, drug use, risky sex)

62
New Frontiers for Prevention

• Family based interventions
• No success to date
• Penalties for youth possession and use
• No evidence of effectiveness to date
• May lead to further marginalization
• Key may be make penalty meaningful to youth
• E.g., In Florida, student has drivers license
  suspended