OBJECTIVE To assess the effectiveness of interventions to reduce underage access to tobacco by deterring shopkeepers from making illegal sales.
METHOD Systematic literature review.
DATA SOURCES The Cochrane Tobacco Addiction group specialised register and Medline. Studies of interventions to alter retailer behaviour were identified. The terms used for searching combined terms for smoking and tobacco use with terms for minors, children or young people, and retailers, sales or commerce.
STUDY SELECTION Studies in which there was an intervention with retailers of tobacco, either through education about, or enforcement of, local ordinances. The outcomes were changes in retailer compliance with legislation (assessed by test purchasing), changes in young people's perceived ease of access to tobacco products, and changes in smoking behaviour. Controlled studies with or without random allocation of retail outlets or communities, and uncontrolled studies with pre- and post intervention assessment, were included.
DATA EXTRACTION Two reviewers assessed studies for inclusion. One extracted data with checking by the second.
DATA SYNTHESIS The results were synthesised qualitatively, with greater weight given to controlled studies. Thirteen of 27 included studies used controls.
RESULTS Giving retailers information was less effective in reducing illegal sales than active enforcement and/or multicomponent educational strategies. No strategy achieved complete, sustained compliance. In three controlled trials, there was little effect of intervention on youth perceptions of access or prevalence of smoking.
CONCLUSIONS Interventions with retailers can lead to large decreases in the number of outlets selling tobacco to youths. However, few of the communities studied in this review achieved sustained levels of high compliance. This may explain why there is limited evidence for an effect of intervention on youth perception of ease of access to tobacco, and on smoking behaviour.
- smoking prevention
- sales to minors
- young people
- systematic review
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
- Table 1 Details of studies included in the review
Study Participants Intervention Outcomes Alberta (Compliance for Kids), Abernathy.47
Pre-post study, no control group. Three communities, Canada
Stratified random sample of retail outlets: community A n=30; community B n=40; community C (all retail outlets) n=7. Different sample pre- and post-test. Legislation and education. Introduction of a city bylaw combined with a retailer education programme. In communities A&B materials were delivered by mail, in C they were delivered by a health unit employee. Illegal sales: single test purchase attempt by < 16 years at baseline and 3 months.
Retailers surveyed by telephone on knowledge of the Tobacco Restraint Act and stated willingness to sell to minors if accompanied by parental note.
Bristol, Naidoo and Platts.54
Pre-post study, no control group. City, UK.
100 tobacconists randomly selected from Yellow Pages at pre-test and 50 at post-test. Retailer education and publicity. Letter sent to all tobacconists clarifying law and offering publicity material. Press conference publicising results of purchase attempts. Illegal sales: single test purchase by one of four children "well under" 16 years at baseline and 12 months. Retailers interviewed about knowledge of law. Chicago, Jason et al.12
Controlled trial with stratified randomisation of stores within ethnic areas.
Urban communities (white, Latino and African American neighbourhoods), USA.
120 stores in 3 ethnic areas were identified. 7 stores went out of business during study and are excluded. Education and enforcement: Test of different schedules. Purchase attempts were made monthly, beginning in December 1993. After April 1994 assessment, merchants who sold cigarettes were notified and warned of programme of inspections. Other merchants given congratulatory note. All given educational materials. At enforcement visits fines of $200.
Interventions: Enforcement every 2, 4, and 6 months. Control: No enforcement.
Illegal sales: assessors judged to be 16- 17 years who gave correct age if challenged. Single attempt at each store each month. Generally unaccompanied. Different assessor each month.
Monthly compliance tests continued for stores not due for enforcement checks. Period of intervention 11 months.
Cook County, McDermott et al.49 Pre-post study, no control group. Suburban area, Illinois, USA. 347 retailers; 129 stores selling at baseline retested Education and warning. Copy of state law and warning letter sent to all vendors selling at baseline (compliers sent congratulatory letter, not revisited). Illegal sales: single purchase attempt by 14- 17 years accompanied by adult supervisor at baseline and 3 month follow up. Erie County 1987, Skretny et al.46Controlled trial with random allocation of stores to condition. Buffalo, New York, USA. Retail outlets (supermarkets, convenience stores, and pharmacies) Intervention n=60; control n=58. Retailer education. Stores mailed educational package with letter citing the New York state law and requesting assistance in observing the law. Warning signs and tip sheet for educating employees. Illegal sales: single test purchase by one of 7 minors, 14- 16, 2 weeks after intervention (no baseline assessment). Erie County 1995, Cummings et al.13
Controlled trial, 6 pairs of communities non-randomly assigned. Stores within communities randomly assigned to enforcement schedules. 12 socioeconomically varied communities, New York, USA.
All retail outlets selling tobacco over the counter in each community (excluding bars). 319 outlets at follow up. Enforcement compared to warning only. All licensed stores sent reminder of law and warning of random enforcement December 1994. Stores allocated to 1, 2 or 3 enforcement checks over 8 month period. Violators fined, compliers sent congratulatory letter. Illegal sales: test purchase attempt by one of 23 minors aged 15- 17 (some conducted enforcement as well as compliance checks) gave true age if asked. At follow up 98% of stores checked 3 times using different assessors. Same 3 minors used in paired communities. Smoking behaviour: not assessed. Everett, Hinds.39Pre-post study, no control group. One community, Snohomish, Washington, USA. Retailers (target of intervention)
10th grade high school students (pre-intervention n=221, post-intervention n=279). Mean age 15 years.
Legislation. New local ordinance required signs, restricted vending machines, required proof of age for purchases, required a local licence, introduced penalties for violations. Smoking behaviour: survey of students before and 1 year after ordinance. Perceived access: Purchasing behaviour and reports of retailers asking for proof of age. Gateshead, Bagott et al.19 20
Controlled study. Catchment areas for 2 schools, Gateshead, UK.
14- 15 year old students at 2 schools (117 at baseline in intervention school, 107 at control school). Retailers in intervention area: 13 shops surveyed, 70% of those close to intervention school. Enforcement (intended). Test purchasing was carried out around the intervention school, prosecution intended for offenders. The retailers around the control school were not approached. No illegal sales were made. Smoking behaviour: survey of year 10 students pre- and post-intervention. Self reported source of cigarettes and difficulty in purchasing. Illegal sales assessed only as part of the intervention, not as part of assessment. Trading standards officer was in shop during purchase attempt by child aged # 13 years. Harlem, Gemson et al.14
Controlled trial with random assignment of stores
Urban community, New York City, USA.
152 stores licensed to sell tobacco (excludes 29 no longer selling tobacco at end of study). Enforcement or education. Enforcement: stores violating regulations fined. Education: single visit; explanation of law, leaflets. Control: no intervention. There was media coverage of the result of the baseline survey of the enforcement stores. Illegal sales: test purchases by African American male or female aged 12- 14. Asked to purchase a single cigarette, if refused second student would ask for a packet. Assessment at baseline, 6 month (violators fined) and 1 year. Assessment and enforcement checks were combined. Leominster, DiFranza et al.38Pre-post study, no control group. Single community, Massachusetts, USA. Retailers (target). Students grade 7- 12, 501 at baseline, 633 in total in 2 post intervention surveys. Education plus enforcement. Merchant education followed by enforcement. Violators warned and repeat offenders fined. Compliers received letters of commendation. Smoking behaviour: change in self reported smoking prevalence in 3 cross sectional surveys. Illegal sales: attempted purchases on 3 occasions, 11, 15, and 19 months after regulations adopted, by males and females of different ages. No baseline before introduction of regulations. Manly, Junck et al.34Pre-post study, no control group. Single community, New South Wales, Australia. All 54 tobacco retailing outlets not licensed for alcohol in Manly. Assessed at baseline, 3 months, and 10 months. Community intervention including media coverage and public forum. Retailer education: Police delivered kit and verbally congratulated those not selling at baseline. Public Health Unit sent official warning letter to those selling at baseline. Illegal sales: test purchase by 14- 16 years visiting in pairs, truthful responses. At baseline and 3 months, outlets visited up to three times if they refused sale. At 10 months, outcome based on single attempt. Massachusetts, Rigotti et al.25Controlled trial. Three intervention communities planned active enforcement of similar regulations; three matched control communities not planning enforcement. USA. Every retail outlet in each community. All students in grades 9- 12 aged under 18 and resident in city or town where they attended school were surveyed at baseline and annually for 2 years. Enforcement. All six local health departments distributed written information. Health departments of intervention communities began testing and penalising violators with an escalating series of warning and fines. Encouraged to test 4 times/year, with more frequent testing for non-compliant retailers.
In control communities no compliance testing was planned.
Illegal sales: test purchases (independent of compliance testing in intervention communities) made by girls aged 16. Truthful responses. Single attempt per store, at 6 month intervals. OTC and vending machine purchases attempted.
Smoking behaviour: prevalence of tobacco use in past 30 days, prevalence of any tobacco use, and of regular daily use. Access measured as difficulty reported by those who had tried to buy tobacco in previous 6 months: bought/tried to buy tobacco. Hardly ever refused/ refused at least half the time.
Monterey, Altman et al.24 Controlled trial with allocation of communities by coin toss. Two pairs of communities, California, USA. All retailers in each community assessed. All students in selected grades surveyed. Comprehensive community intervention with retailer education. Retail outlets contacted directly on multiple occasions. Aimed to raise awareness, educate and transmit community norms. Illegal sales: test purchase by 13- 17 years on multiple occasions over 34 months. Truthful responses. Smoking behaviour: initial survey of 7th, 9th, 11th graders. Grade cohorts surveyed at four time points. New South Wales, Schofield et al.15Controlled trial with random assignment of stores. New South Wales, Australia. Retailers within 50 km identified. 300 retail outlets randomised, 272 checked for compliance, 217 completed both surveys. Education or education and threat of enforcement. All retailers surveyed pre- and postintervention.. Minimal intervention (education): information from public health unit. Maximal intervention (education and threat of enforcement); plus warning of enforcement; plus visit. Illegal sales: single test purchase using 18 years assessed as looking younger, who wore school uniform, at baseline and 2 month postintervention. Each assessor did approx. 50 checks. Compliance was requiring proof of age. Knowledge and attitudes of retailers assessed by survey. Smoking behaviour: not assessed. Ontario, Dovell et al.29Pre-post study, no control group. Two neighbouring health units. Two interventions were implemented sequentially in one (KFL&A) with assessment over 21m and one in the other (H&PE) with assessment over 2 weeks. Canada. Stores in the health unit areas. Different numbers assessed at each time point. Retailer education. In KFL&A: 1. (between base and 6 months) - media events, public flyers, results of purchase attempts; 2. (between 6 months and 21 months) Targeted intervention sending information kits and advising retailers of compliance, also federal intervention.
In H&PE: Federal intervention - kits sent by Health Canada with letter saying inspector might call. Assessment 1 week later.
Illegal sales: single test purchase at baseline, 6 months and 21 months by pairs of 13- 14 years. One watched while other requested cigarettes. Oregon (Project SixTeen), Biglan et al.30 55 Multiple baseline time series. Eight communities in Oregon, USA. Retailers in eight communities. Programme first used in two while baseline assessment continued in two more, then replicated in a further four. Community intervention and education, positive reinforcement. Components: mobilising community support; merchants education; changing consequences to clerks for selling (reminder) or not selling (gift token reward); publicity; feedback. Implemented by a community coordinator. Illegal sales: test purchase attempts (underage possession illegal) by 16- 17 years at 2 week intervals over approx. 6 months. Perth, Mawkes et al.35 Pre-post study, no control group. Areas near schools in urban areas of Perth, Western Australia. Tobacco merchants within 2 km of 12 secondary schools surveyed at baseline in 1992 (230 outlets) and again in 1994 (284 outlets). Comprehensive, including education, enforcement and community action. State wide activities. Materials sent to retailers. Prosecutions, or warning letters if prosecution impossible. Illegal sales: single purchase attempts by pairs of 15- 16 years (same pair used for most areas at follow up). At baseline, but not follow up, children could say cigarettes were for parents if initially refused. San Diego Project TRUST, Keay, Wildey et al.26 27 Controlled trial with cluster allocation of stores. Six communities in California, USA. Retailers in six low income, ethnically diverse communities. Data from 260 stores at pre- and post-test and 236 at 6 month follow up. Education and community awareness. Face to face education of retailers combined with community and media strategies. Included video for training sales assistants. Managers of stores which sold at baseline also received 10 min discussion from Environmental Health Field inspectors. Duration of intervention 1 year. Control: no intervention. Illegal sales: single purchase or purchase attempt (in two communities ringing up of the cigarettes constituted a "sale") by 70 teens aged 12- 17 years. Assessment at baseline and 1 year (1 month after conclusion of intervention), and 6 months later. Santa Clara, Altman et al.16 17 Controlled trial with random allocation of stores to different contact. California, USA. 412 stores selling cigarettes over the counter and 30 with vending machines. At 6 months post-test n=408, at 1 year n=97 from a stratified random sample. Retailer education (as supplement to community intervention). Components: 1. Community education through mass media and presentations to community groups; 2. Direct education with merchants who sold tobacco - stores were randomly allocated to one of 3 variants of this component: (i) no personal contact, (ii) mailed information, (iii) visits from project staff with education kit; 3. Contact with CEOs of major chains and franchises. Illegal sales: single test purchases by 14- 16 year olds (18 assessors). Attempt to use same assessor for pre- and 6 month post-test. Solana County, Feighery et al.31 Pre-post study, no control group. Four communities in one county, California, USA. The initial intervention was modelled on that used in Santa Clara. Random 50% sample of stores in three cities and all in a fourth city surveyed at pretest (n=169). 104 revisited at post-test 2, and a further 41 visited for the first time. Education only, followed by enforcement
Education packages mailed to retailers. Pretest results widely publicised. Second phase; local police departments requested to carry out stings - underage police cadets attempted purchases and citations issued. Results (34% sales) were publicised in local media. Local judges did not always willing to sentence or fine those issued with citations.
Illegal sales. Test purchases by 14- 16 year olds. Single attempt at pre-test and two post-tests. St Paul, Forster et al.37Pre-post study, no control group. Minnesota, USA. Random sample of businesses with vending machines. Baseline data for 95 of a possible 237 machines; complete data for 77. Vending machine locks. New ordinance which required locking devices on all cigarette vending machines in city. Illegal sales: test purchase by females 15 years before implementation and 3 and 12 months postimplementation. Stirling, Campbell.56
Pre-post study, no control
Convenience sample of 41 outlets surveyed at baseline. Intervention delivered to those selling. Resurvey of these and a further 26 outlets not initially visited 7 months later. Warning: immediately after a successful purchase at baseline, verbal warning given by Trading Standards Officer. Followed up by letter. Illegal sales: test purchase by 11- 15 year olds. Sydney 1992, Chapman et al.18Controlled trial with randomisation of outlets to warning letter. Six suburban areas, Australia. Pre-intervention (T1) "stings" at a sample of outlets (n=255). Postintervention (T2) stings carried out at 2 months (n=244). Retailers found selling cigarettes at T1 were randomly allocated to receive or not receive a warning letter (intervention n=50, control n=49). Threat of enforcement. Letter warning of another "sting" and threatening prosecution sent to 50% of retailers selling at T1. Also media publicity about an undercover buying operation. Illegal sales: single test purchase by 12- 13 year olds who visited stores in pairs and gave truthful answers at T1 and T2, approx. 2 months after warning letter. Sydney 1995, Staff et al.28 Controlled study. Two distinct areas of city, Australia All retailers targeted, 357 education kits distributed. Public and schools also targeted. Students at 13 public secondary schools surveyed. Retailer education and community awareness. Intervention area: "beat police" delivering education kits to retailers. Media articles, information in school newsletters. "Informers" line to identify non compliant retailers. Control area: no intervention. Illegal sales: not assessed. Smoking behaviour: baseline survey of students in years 7- 11. Follow up 6 months later. Smoking prevalence. Ease of purchase. TPOP (Tobacco Policy Options for Prevention), Forster et al.21- 23 Controlled trial with community randomisation. Fourteen rural communities, Minnesota, USA. All retail outlets, and students in grades 8- 10 in all communities. Comprehensive community intervention including new ordinances, community awareness, media campaigns, and compliance checks. All intervention communities introduced new ordinances, with a variety of provisions. Three control communities introduced new ordinances but these were weaker and less comprehensive. Illegal sales: test purchases, two attempts on successive days by 15 year old females. Smoking behaviour: in school survey of grades 8- 10 in 1993 and 1996. Never/monthly/weekly/daily smokers. Perceived access: ease of purchase; usual source; number of purchase attempts. Wisconsin, Schensky et al.48
Pre-post study, no control group. Dane County, Wisconsin, USA.
60 retailers surveyed at baseline and follow up. Retailer education and feedback. Retailers notified of the results of compliance test, given signage and offered training. Illegal sales: single test purchases by 12- 15 year olds in groups of 2- 3 who reported age honestly, at baseline and 12 months. Woodridge, Jason et al.32 33 50 Pre-post study, no control. Single community, Illinois, USA. All stores in Woodridge (19- 20 at baseline, 22- 30 at follow up assessments). Local students (680 at pretest, 639 at post-test). Legislation and enforcement. New ordinances, including licensing, enforcement, possession of cigarettes an offence. Education and media coverage. Legislation introduced May 1989. Quarterly compliance checking, compliers congratulated. Illegal sales: test purchases by 12- 17 year olds (age range changed over period). Assessments at 6 month intervals from August 1988 to December 1994. Smoking behaviour: surveys of local 7th and 8th graders in March 1989 and April 1991. In 1991 they were asked about perceived availability. Further survey in 1996.Illegal sales: test purchase attempt indicates that assessment procedure did not involve completion of purchase.
Table 2 MEDLINE search strategy used to identify interventions to reduce tobacco sales to minors
Smoking cessation OR tobacco OR cigar* OR SMOKING-CESSATION OR TOBACCO-USE-DISORDER OR TOBACCO OR NICOTINE OR TOBACCO-SMOKELESS OR SMOKING/ prevention-and-control , therapy , legislation-and-jurisprudence OR ((quit* OR stop* OR ceas* OR giv*) near smoking)
adolescen* OR minors OR under()age OR student* OR (young near2 people) OR children OR juveniles OR girls OR boys OR teenager* OR teens OR child
sale OR sales OR retail OR retailer* OR store OR stores OR sell OR selling OR shop OR shops OR tobacconist* OR vending OR vendor* OR merchant* OR COMMERCE/ legislation-and-jurisprudence
Terms in capitals are MeSH headings. Last issue of Medline Express on Silverplatter searched was 1999/7. Additional searches of other databases did not retrieve any references to studies not retrieved from MEDLINE.
Table 3 Relevant studies not meeting full inclusion criteria
Study Description Reason for exclusion DiFranza and Brown 1992.40 41 Survey of tobacco retailers, undertaken in Massachusetts in 1991, to evaluate the efficacy of the Tobacco Institute�s "Its the Law" programme. Of the retailers surveyed only 4.5% (7) were participating in the programme. Most of these retailers were found to be willing to sell cigarettes to minors. Most of the retailers not participating in the programme (131/149) were also willing to sell cigarettes to minors. Post intervention only, comparing retailers participating with those not. DiFranza et al 1996. 40 41 Survey of tobacco retailers to evaluate the efficacy of the Tobacco Institute's "Its the Law" programme, undertaken in Massachusetts in 1994
Of 480 purchase attempts, 240 were made from participating retailers. Participation did not lead to a significantly lower rate of illegal sales.
Post intervention only, comparing retailers participating with those not. Lewis et al 1996.4242- 44 Project SCAN (Stop Children�s Addiction to Nicotine) included three broad activities: 1. a public information campaign utilising the media, newsletters, and a speaker�s bureau; 2. distribution of sale to minor cards to all citizens to report on stores that illegally sell cigarettes to minors; 3. work with community leaders, voluntary organisations, businesses, and law enforcement officials to support and implement activities. Police officers delivered the package to merchants. This study did not include any preprogramme measurement. Lewis et al 1996. 42- 44 Intervention was citation or commendation of clerks by citizens in a Wichita community coalition. Both alcohol and tobacco sales targeted. No data on number of stores selling tobacco, or number in which intervention was delivered. Woodruff et al 1995. 42- 44 Intervention to reduce sales of single cigarettes. Test purchasers were 19- 32, not underage youths.