Review Article

Impact of Tobacco Control Interventions on Smoking Initiation, Cessation, and Prevalence: A Systematic Review

Table 2

Overall summary of the impact of tobacco control interventions on smoking initiation, cessation, and prevalence.

InterventionSmoking behavior

Increasing the price through taxationOverall: high* evidence to estimate the independent impact on smoking behavior
Initiation: moderate evidence, 4 out of 5 longitudinal studies demonstrated some effectiveness; PPE of initiation ranged from −0.65 to −0.09
Cessation: moderate evidence, price elasticity of cessation ranged from 0.375 to 1.17
Prevalence: high evidence, suggesting effectiveness
PPEs ranged from −1.41 to −0.10 among youths and −0.45 to 0.10 among adults

Banning smoking in public placesOverall: moderate evidence to estimate the independent impact on smoking behavior
Initiation: low evidence, unable to make a conclusion due to equivocal results
Cessation: low evidence, 2 of 3 longitudinal studies with comparison groups did not find a significant change in cessation rates after implementation
Prevalence: moderate evidence, suggesting effectiveness;
Percentage change in prevalence ranged from −31.9% to −7.4% compared with control groups after 1 to 3.5 years

Banning advertising and sponsorship of tobacco productsOverall: insufficient evidence to estimate the independent impact on smoking behavior
Initiation: insufficient evidence, unable to make a conclusion because no studies were included
Cessation: insufficient evidence, unable to make a conclusion because no studies were included
Prevalence: low evidence, unable to make a conclusion due to low quality studies;
Two studies among adults showing no effectiveness, 2 studies among youths showing some effectiveness, and 1 found an increased prevalence with stronger laws

Educating people about the dangers of smoking through health warning labelsOverall: insufficient evidence to estimate the independent impact on smoking behavior
Initiation: insufficient evidence, unable to make a conclusion because no studies were included
Cessation: low evidence, 2 studies showing no effectiveness
Prevalence: low evidence, 2 studies showing no effectiveness

Educating people about the dangers of smoking through mass media campaignsOverall: moderate evidence to estimate the independent impact on smoking behavior
Initiation: moderate evidence, suggesting effectiveness
One cluster RCT demonstrated no effectiveness, but 4 longitudinal studies suggested a reduced initiation rate (odds of initiating smoking ranged from 0.67 to 0.8)
Cessation: low evidence, unable to make a conclusion due to equivocal results.
Seven studies with comparison groups showed equivocal results
Prevalence: moderate evidence, suggesting effectiveness.
Odds of being a smoker 1 to 6 years after start of intervention* ranged from 0.62 to 0.93§, but one cluster RCT showed no effect on smoking prevalence

*Grading classification: high strength of evidence indicates high confidence that the evidence reflects the true effect, and further research is very unlikely to change the result. Moderate strength of evidence indicates moderate confidence that the evidence reflects the true effect, and further research may change the result. Low strength of evidence indicates low confidence that the evidence reflects the true effect, and further research is likely to change the result. Insufficient indicates that no evidence was available.
One of these studies stratified results by gender and age (% impact on prevalence rate after 2 years for those under age 45 years = −7.4% and for those aged 45 years and older = −1.4%).
These studies had severe methodological flaws that limit our ability to make conclusions.
The strongest study methodologically showed a hazard ratio of 0.8 (95% CI: 0.71, 0.91; 𝑃 = 0 . 0 0 1 ) per 10,000 GRP cumulative exposure.
Two of the pre-/post- cross-sectional studies were methodologically stronger than the others. One study reported an odds ratio of cessation = 1.27 (95% CI: 0.77 to 2.08). The other reported a relative risk of quitting = 1.1 (95% CI: 0.98 to 1.24) per 5,000 GRPS.
§Additionally, a well-conducted time series analysis reported a decrease in percentage point prevalence two months later of −0.00077 per 1 GRP per month increase ( 𝑃 = 0 . 0 2 5 ). This is the equivalent of each person viewing an average of 4 ads per month to achieve a 0.30 percentage point decline in smoking prevalence.
CI: confidence intervals; GRP: gross rating point; PPE: price participation elasticity; RCT: randomized controlled trial.