Youth access interventions do not affect youth smoking

Fichtenberg CM, Glantz SA.

Publication Details

Authors' objectives

To determine the effectiveness of laws restricting youth access to cigarettes on the prevalence of smoking among teenagers.

Searching

MEDLINE was searched from 1985 to 2001. The search terms used were reported. In addition, the references of reviews and located articles were checked.

Study selection

Study designs of evaluations included in the review

Studies that reported both compliance (with youth access laws) and prevalence data, or reported prevalence in the intervention and control communities, were eligible for inclusion. Cross-sectional studies based on aggregate data, as opposed to individual data, were excluded. A prospective cohort study and cross-sectional studies were included.

Specific interventions included in the review

Studies examining laws restricting youth access to cigarettes (or youth access programmes) were eligible for inclusion. The included studies reported the following types of interventions: community and retailer education with no enforcement; retailer education with enforcement (e.g. via warnings, fines, or suspension of tobacco selling licenses); and comprehensive interventions that include community intervention, enactment of laws and a variety of enforcement strategies. The intensity of the interventions varied between the studies. All of the included studies measured compliance with youth access laws using 'sting operations', where teenagers are sent into stores to try and buy cigarettes and record whether the merchants were willing to sell. Details of the youths used in these operations and the number of stores (and number of times) visited by the researchers were not reported for the individual studies.

Participants included in the review

The type of participants eligible for inclusion were not explicitly stated. The age of the participants within the included studies was reported either in terms of age or school grade. The categories of age (total range: 12 to 17 years) and grade (total range: 7 to 10) varied between the studies.

Outcomes assessed in the review

Studies reporting the prevalence of youth smoking were eligible for inclusion. Studies reporting smoking initiation were excluded, as were those that used process outcomes such as whether youths perceived that they could buy cigarettes. Smoking prevalence in the included studies was measured using school-based surveys. The smoking measures used were smoking at least once during the past 30 days, smoking at least once a week, self-reported 'smokers', daily smoking, frequent smokers (at least 20 in the past 30 days) and self-reported 'regular smokers'.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

Assessment of study quality

The authors did not state that they assessed validity.

Data extraction

The authors did not state how the data were extracted for the review, or how many reviewers performed the data extraction.

The different smoking measures used were pooled in two groups: 30-day smoking, which comprised 30-day smoking, weekly smoking and self-reported smoking; and regular smoking, which comprised daily smoking, frequent smoking and self-reported 'regular smokers'. For studies that used separate measurements for different age groups, arithmetic averages of the separate effects were used. Teenagers aged 18 years or older were excluded from the analysis.

Methods of synthesis

How were the studies combined?

To assess whether merchant compliance with access laws is related to youth smoking prevalence, the Pearson product moment correlation (r) between youth smoking prevalence (30-day and regular use) and percentage merchant compliance was calculated at both baseline and follow-up (the data from individual communities, as well as baseline and follow-up data, were treated as separate data points). To assess the impact of active enforced efforts, the authors correlated the changes in compliance (absolute changes from baseline to follow-up) with changes in youth smoking prevalence (30-day and regular use) (individual communities were treated as separate data points).

For controlled studies that reported longitudinal smoking data (n=5), differences between the intervention and control for changes in 30-day prevalence were pooled using a random-effects model with community as the unit of analysis. The standard errors were estimated for all but one of the studies. No pooling was performed for the outcome regular smoking as only 2 studies reported the relevant data.

How were differences between studies investigated?

Differences between the studies were discussed in the text; no formal statistical analysis of heterogeneity was conducted.

Results of the review

Eight studies (1 cohort study and 7 cross-sectional studies) met the inclusion criteria; these included at least 20 communities in total (number not stated for one study). Within the intervention communities, 15,446 participants were included at baseline and 16,586 at follow-up. Within the control communities, 9,401 participants were included at baseline and 10,431 at follow-up.

Relationship between merchant compliance and youth smoking.

There was no statistically-significant relationship (20 communities) between merchant compliance and 30-day (r=0.116, P=0.486) or regular (r=0.017, P=0.926) smoking prevalence. There was no evidence of a threshold effect after compliance reached a certain level (e.g. 90% in 3 studies). There was no evidence (18 communities) that an increase in merchant compliance was associated with a decrease in 30-day (r=0.294, P=0.237) or regular (r=0.274, P=0.287) smoking prevalence.

Effect of youth access programmes.

There was no significant difference in youth 30-day smoking prevalence in communities with youth access interventions, compared with control communities (5 studies; difference -1.5%, 95% confidence interval, CI: -6.0, 2.9). Four of the 5 studies reported the compliance rates; compliance exceeded 82% in the intervention community. For the outcome regular smoking, one study reported a 2.9% increase (P=0.08) in prevalence while another reported a -4.9% decrease (95% CI: -0.9, -0.7).

Authors' conclusions

Youth access interventions are not associated with a consistent positive effect on youth smoking prevalence. Furthermore, there was no evidence that increased compliance is associated with decreased prevalence.

CRD commentary

The review question was clear in terms of the interventions and outcomes of interest. The authors did not report explicit inclusion criteria in terms of the study design, but they did report some details of the type of data that eligible studies had to have reported. No explicit inclusion criteria were reported for participants either, other than youths. Only one electronic database was searched and no attempt was made to look for unpublished studies. This means that some important information may have been missed and publication bias cannot be ruled out. No information on the study selection and data extraction processes was given, and the authors do not appear to have assessed the quality of the included studies. This means that the reader cannot assess the potential for errors and reviewer bias.

All the outcome measures reported by the individual studies were pooled into two measures (30-day and regular smoking). There were insufficient details of the individual studies to check whether this approach was appropriate. For the correlation analysis, the data from individual communities, as well as baseline and follow-up data, were treated as separate data points. This means that the same participants would have been included twice for the cohort study. There were insufficient details of the methodology of the included studies to check whether this was the same for any remaining studies. The included studies appear to have varied quite considerably in terms of the interventions and participants, which may limit the meaningfulness of the pooled estimate of the meta-analysis. The authors did not investigate statistical heterogeneity. The authors also stated that their estimated standard errors were probably too small, which would mean that the results would be biased in favour of the intervention. The authors' conclusions appear to follow from the results presented.

Implications of the review for practice and research

Practice: The authors stated that given the limited resources available for tobacco control, as well as the expense of conducting youth access programmes, tobacco control advocates should abandon this strategy and devote the limited resources that are available for tobacco control toward other interventions with proven effectiveness.

Research: The authors did not state any implications for further research.

Funding

National Cancer Institute, grant number CA-61021.

Bibliographic details

Fichtenberg C M, Glantz S A. Youth access interventions do not affect youth smoking. Pediatrics 2002; 109(6): 1088-1092. [PubMed: 12042547]

Indexing Status

Subject indexing assigned by NLM

MeSH

Adolescent; Adolescent Behavior /psychology; Child; Commerce /legislation & jurisprudence; Confidence Intervals; Health Policy /legislation & jurisprudence; Humans; Legislation as Topic /standards; Outcome Assessment (Health Care); Prevalence; Program Evaluation /statistics & numerical data; Smoking /epidemiology /legislation & jurisprudence /prevention & control; Social Control, Formal /methods; Tobacco Industry /legislation & jurisprudence; United States /epidemiology

AccessionNumber

12002001384

Database entry date

29/02/2004

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.