Cracking Down On Youth Tobacco May Influence Drug Use

Abstract

This study evaluated the influence of tobacco Possession-Use-Purchase (PUP) law enforcement and illicit drug use and offers. Twenty-four towns were randomly assigned into two conditions. Both conditions focused on reducing minors’ access to commercial sources of tobacco. The communities assigned to the experimental condition also increased their PUP law enforcement, whereas among communities in the control condition, PUP law enforcement remained at low levels. A Hierarchical Linear Modeling analytical approach was selected due to the multilevel data and nested design. The likelihood of a child currently using drugs, ever having used drugs, or illicit drug offers was lower in the experimental versus control conditions. These outcomes suggest that police efforts to reduce specific substance use behaviors (i.e., underage tobacco use) may have a positive spillover effect and help reduce teen drug use and illicit drug offers.

Reducing youth access to tobacco products has been advocated as one public health strategy to address the problem of youth tobacco use (Ma, Shive, & Tracy, 2001). Several investigators have argued that reductions in minors’ ability to purchase tobacco might have an effect on the level of cigarette smoking among youth (DiFranza et al., 1992; Jason et al., 1991; Jason et al., 1999). For example, Forster et al. (1998) found a 4.9% reduction in daily smoking for youth in cities exposed to enforcement of a comprehensive youth access ordinance compared to control cities. Altman et al. (1999) found that seventh graders in the treatment communities were significantly less likely to use tobacco than youth in control communities. In contrast, a study by Rigotti et al. (1997) found that sales law enforcement did not lead to reductions in youth tobacco use. A limitation in this study was that there were no significant self-report differences between experimental and control groups regarding the percentage of youth that purchased tobacco. This suggests that youth access to tobacco may not have been adequately restricted in the enforcement communities. In larger samples, Ross and Chaloupka (2004) and DiFranza, Savageau, and Fletcher (2009) found that youth access laws were related to lower probabilities of youth smoking in a nationally representative survey of high school students.

The fact that teen smokers are beginning to shift to social sources for tobacco (Jones, Sharp, Husten, & Crossett, 2002) suggests that for some youth smokers, the barriers to purchasing retail tobacco are strengthening. If merchants are less likely to sell tobacco to youth, minors increasingly will rely on social sources to obtain their cigarettes. Jones, Sharp, Husten, and Crossett (2002) found that there has been a general shift in youth cigarette acquisition from retail to social sources, but heavier smokers are more likely to purchase cigarettes than less frequent smokers. Shive, Ma, and Shive (2001) found that 30.8% of young adults provided tobacco products to minors, thus supporting the notion that minors are increasingly relying on social sources. More comprehensive programs (e.g., fining minors for tobacco possession, enforcing school smoking bans) might need to be combined with efforts to restrict merchants from illegally selling tobacco products to minors in order to decrease youth access to tobacco and possibly reduce youth smoking. For example, strongly enforcing school smoking bans is related to lower smoking prevalence rates (Wakefield & Chaloupka, 2000), and community attitudes supporting tobacco possession laws are associated with a lower likelihood of youth supplying tobacco to other minors (Pokorny, Jason, & Schoeny, 2006).

Tobacco Possession-Use-Purchase Laws (PUP) provide consequences for minors who illegally possess, use or purchase tobacco products. Studies evaluating the effectiveness of these laws to reduce youth tobacco use found promising results. For example, Langer and Warheit (2000) assessed adolescents cited for PUP law violations who appeared in a special tobacco court and then watched a video on the health effects of smoking. At the two-month follow-up, 28% of youth claimed not to have used tobacco since the time of their citation and an additional 29% said they used less tobacco. In school-based samples, Livingood et al. (2001) compared teen smoking attitudes and behaviors between two Florida counties with the highest levels of PUP law enforcement and two counties with the lowest levels of enforcement. Results indicated that youth in the high enforcement condition had a significantly reduced likelihood of 30-day tobacco smoking compared to those in the low enforcement counties. An eight-town randomized community-control trial found that white youth, who lived in communities with strict enforcement of tobacco sales and possession laws had significantly fewer increases in the prevalence of tobacco use over time than those living in communities with only moderate enforcement of tobacco sales laws (Jason, Pokorny, & Schoeny, 2003).

Other studies of PUP law enforcement suggest that this type of intervention may have important secondary effects. Jason, Katz, Pokorny, Engstrom, Tegart, and Curie (2000) examined data over an eight year period in 29 towns with and without laws dealing with minor access to tobacco and fining minors for possession of tobacco. Towns that developed and implemented polices toward the enforcement of minor access to tobacco laws had fewer overall crimes, both at the beginning and end of this time period. Early adopters of these youth access policies possibly have citizens and officials who are more oriented to responding quickly to potential youth problems, such as youth access to tobacco products, and such attitudes and behaviors might translate into policies that deter overall crime rates. In another study, a sample of high school youth living in two communities with regular enforcement and fines for tobacco possession engaged in significantly less smoking than those living in three communities without regular enforcement and fines for possession (8.1% vs. 15.5%, respectively) (Jason, Berk, Schnopp-Wyatt, & Talbot, 1999). A somewhat unexpected finding in the study by Jason et al. (1999) was that adolescents in communities with regular enforcement and fines for possession of tobacco, in comparison to youth in communities without these strong youth access laws, reported being approached significantly less frequently by someone trying to sell them illegal drugs. It is possible that it is more difficult to sell drugs in communities that are perceived to be actively protecting youth by strictly enforcing anti-smoking legislation. Perhaps fining youth for illegal possession of tobacco leads to fewer drug sales in those communities.

This idea can be linked to the “Broken Window” theory, which is an approach that claims the presence of disorder and incivility in a neighborhood frightens its inhabitants, which makes them withdraw from activities that will improve the conditions of the neighborhood (Wilson & Kelling, 1982). This fear and withdrawal only encourages offenders to engage in more illicit behaviors since there is a low level of social control. By having the police focus on “lower level” violations, they are so-called “fixing the windows”, and therefore order is maintained and offenders are less likely to engage in illegal behaviors.

Recently, Jason et al. (2008) randomly assigned 24 towns to two conditions. Both conditions focused on reducing minors’ access to commercial sources of tobacco, but the intervention communities also increased their PUP law enforcement, whereas in the control communities, PUP law enforcement remained at low levels. Findings indicated that rates of current smoking increased at a significantly slower rate for adolescents in towns where PUP law enforcement was increased. PUP law enforcement increases police contact with youth smoking in public and builds community awareness about youth tobacco use. Consequently, it is possible that these policies can have important secondary effects in communities that actively enforce them. For example, in the same study, Jason, Hunt, Adams, Pokorny, and Gadiraju (2007) found that 5 out of 12 communities (42%) adopted local 100% smoke-free ordinances after our research team had worked with them to increase PUP law enforcements; only 1 of the 12 (8%) communities that had not worked with us to increase PUP law enforcements adopted 100% smoke-free ordinances It is also possible that these types of PUP law enforcement might also affect other behaviors such as drug use or illicit drug offers.

The present study used the data set described in Jason et al.’s (2008) randomized study of 24 towns. It was hypothesized that towns exposed to an intervention designed to strengthen enforcement of PUP laws would have less drug use and illicit drug offers to minors compared to towns that did not actively increase their PUP enforcement efforts.

Method

Procedures

The Youth Tobacco Access Project involved 24 towns in Illinois, with four cohorts of data collected each spring from 2002 through 2005. Approximately 70 towns were contacted for recruitment into the study. Towns were identified from a list of municipalities participating in a state sponsored tobacco sales enforcement grant program. In each town, five gatekeepers were contacted for permission to participate and all five had to agree to participate for the community to be included in the study. Gatekeepers were contacted ion the following order: the Police Chief, the Superintendent of the Middle School, the Principal of the Middle School, the Superintendent of the High School, and finally the Principal of the High School. Elimination criteria included a history of high rates of PUP law citations (i.e., ≥ .15% of the population) or plans to discontinue participation in the state sponsored tobacco sales enforcement program. A final group of 24 towns agreed to be randomized into one of two interventions and cooperate with the research team during the four year project. The students in the study were followed over time.

The 24 participating towns were matched for population size and median income and then randomly assigned to the two conditions. Both conditions focused on reducing minors’ access to commercial sources of tobacco, but the intervention communities (E) also increased their PUP law enforcement, whereas in the control communities (C), PUP law enforcement remained at low levels (< .15% of the population). The C and E towns did not differ significantly at baseline on population size, median household income, racial distribution, PUP law citations, and rates of illegal tobacco sales to minors (See Jason et al., 2008 for more details). The goal was for all communities to have less than 20% illegal commercial sales of cigarettes to minors by the end of the study. Overall rates of illegal sales of tobacco to minors between the E and C conditions did not significantly differ over the course of the intervention.

The twelve E communities agreed to increase PUP law enforcement practices, whereas the 12 C communities received instructions to maintain their current low levels of PUP law enforcement. Over a four year period, the average yearly number of PUP law citations issued to minors within the E communities was significantly higher than those within the C communities (Ms = 16.54 versus 6.31 citations; t(22) = −2.30, p = .03), indicating that PUP enforcement was, in fact, stronger in E compared to C towns. There were no significant citation differences between conditions at baseline.

PUP law enforcement

Towns in the E condition had, or implemented, a PUP law specifying that civic fines of approximately $75 could be given to minors caught using or possessing tobacco in any public setting. In most towns, we worked with one or two police officers to ensure that they implemented these procedures. Prior to beginning the enforcement, our project staff had meetings with these police officers and discussed ways to successfully implement these fining measures. Our project staff monitored police efforts closely, via phone calls and person-to-person meetings, to assess the police department’s progress in locating and issuing citations to minors who violated the tobacco PUP law. Project staff conducted town observations to identify areas in which violations occurred when police departments requested additional assistance. We also obtained records of all of the citations issued. The key idea was to convey the message to community youth that purchase, use, and possession of tobacco was illegal, and we felt that periodic fining might be an effective way to communicate this message.

Merchant Sales Enforcements

Merchants in the United States are prohibited from selling tobacco products to minors under the age of 18. Stores that sell cigarettes are also required to post signs, indicating the law against selling cigarettes to minors. Each year, police officials in all 24 towns conducted annual merchant education programs and three compliance checks of all tobacco merchants as part of their participation in the state sponsored tobacco sales enforcement grant program.. Violations of the law were treated as a civil offense resulting in a fine or tobacco sales license suspension for repeat offenders. In other words, all towns were participating in tobacco control activities on the supply side, with regular merchant enforcements to reduce illegal sales of tobacco. The research team independently assessed the proportion of illegal tobacco sales to minors during the study. The procedures for doing these enforcements and compliance checks are described in detail elsewhere (see Jason et al., 2008).

Student Participants

A standardized self-report confidential survey was administered to students in grades seven to ten during 2002, grades seven to eleven in 2003, and grades seven to twelve in 2004 and 2005. In this manner, students were followed over time. In other words, there were four sequential cross-sectional surveys in which some of the same students were included at each wave and many new ones were also added. We were required to get consent from parents during the first time their child participated in the survey and a second time if their child transitioned from middle school to high school during the study. We obtained child assent at each survey point. Students were tracked from year to year based on name from the student assent form data from the survey.

Based on the decision of each school’s administrator, either all students enrolled in the targeted grades or only students who lived in the participating community who were also enrolled in these targeted grades were sampled. Across the four waves of data collection for the present study, a total of 52,550 students were eligible to be surveyed (i.e., enrolled in a target grade at a participating school) at one or more waves. In 11 of the 41 participating schools, school administrators selected only students who lived in the target towns to be eligible for surveys. Of the eligible students, parental consent forms were obtained for 33,991 (65%) students. A total of 29,851 (57%) eligible students completed the survey during at least one wave of data collection. Over the course of four waves, a total of 59,160 surveys were completed, representing an average of two waves of data for each participating student. Of the 59,160 surveys, 482 (0.8%) were excluded from the analyses because of inconsistent or invalid responding across survey items. Three criteria were used to eliminate participants from the data set: 1) inconsistent responding (e.g., responding that they never smoked in one question and responding that they smoked every day in another question); 2) missing data on 70% or more of the items; and 3) invalid responses or unrealistically high reports of tobacco, alcohol, or other drug use (e.g., used all drug types every day during the past 30 days). Because the analyses included a town-level covariate, 7,953 (13%) surveys (i.e., from 4,630 students) were excluded from analyses because the students lived outside of the participating towns and, therefore, were not directly exposed to the intervention. The final sample for the present analyses included 25,404 students and 50,725 assessments.

Measures

Student Survey

The Youth Tobacco Access Project’s Student Survey is a 74 item self-report survey developed to assess students’ demographic variables (i.e., gender, race, grade), as well as their attitudes and behaviors toward tobacco and other drugs (Jason et al., 2003). Members of our research staff administered these questionnaires to the students. The participants were asked whether they had ever tried a variety of tobacco products, including cigarettes, whether there were adults in their home that smoked, whether their friends smoked, whether they used a variety of other drugs, and how many times someone tried to give or sell them illegal drugs over the past year.

Level-1 Variables

All level-1 variables were derived from self-report data obtained from the student survey. Only variables expected to change from wave to wave were selected as level-1 time-varying covariates (i.e., friends who use tobacco).

Outcome variables

For the purposes of the current analyses, we wanted to determine the total current and ever use of drugs. This information was obtained from the following questions: Have you ever used (name of the drug)? Yes, but not in the past 30 days; Yes, in the past 30 days; No. This question was repeated for the following drugs: alcohol, inhalant, marijuana, cocaine, LSD, methamphetamine, ecstasy, prescription drugs, and other drugs (not mentioned). Current use was defined by having used the drug in the past 30 days (0= not currently using, 1= currently using). Ever use was defined as having used the drug either in the past 30 days or had used but not in the past 30 days (0 = never used, 1 = used). For each student, a total score for ever use was tabulated by adding up the nine drug types (a total score could range from 0 to 9). A total score for current use was also tabulated by adding up the nine drug types for current use (scores ranging from 0 to 9).

The final outcome measure assessed illicit drug offers. Students were asked: Over the past year, how many times has someone tried to give or sell you illegal drugs? 0 times; 1 time; 2 to 5 times; 6 to 10 times; 11 to 25 times; 26 to 50 times; and 51 or more times. Students were given a score that could range from 0 (meaning that no one had tried to give or sell them illegal drugs) to 6 (meaning that someone had tried to give or sell them illegal drugs 51 or more times).

The means and SDs of these variables over time are reported in Table 1.

Table 1.

Means and SDs for the Drug and Illicit Drug Offer Outcomes for Es and CS

	Wave 1 (Es)		Wave 1 (Cs)		Wave 4 (Es)		Wave 4 (Cs)
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Alcohol
Current	0.19	0.39	0.19	0.39	0.22	0.41	0.27	0.44
Ever	0.5	0.5	0.51	0.5	0.49	0.5	0.56	0.5
Inhalant
Current	0.02	0.13	0.02	0.15	0.02	0.12	0.02	0.14
Ever	0.07	0.25	0.07	0.25	0.06	0.23	0.07	0.25
Marijuana
Current	0.08	0.27	0.07	0.25	0.09	0.29	0.1	0.29
Ever	0.17	0.38	0.16	0.36	0.21	0.41	0.22	0.41
Cocaine
Current	0.01	0.09	0.01	0.08	0.01	0.1	0.01	0.12
Ever	0.02	0.15	0.02	0.15	0.03	0.16	0.03	0.18
LSD
Current	0.01	0.07	0.01	0.07	0.01	0.07	0.01	0.09
Ever	0.02	0.14	0.02	0.13	0.01	0.14	0.02	0.14
Meth
Current	0.01	0.1	0.01	0.08	0.01	0.07	0.01	0.09
Ever	0.03	0.16	0.02	0.13	0.02	0.14	0.02	0.13
Ecstasy
Current	0.01	0.09	0.01	0.07	0.01	0.09	0.01	0.09
Ever	0.03	0.17	0.02	0.15	0.02	0.14	0.02	0.15
Prescription
Current	0.02	0.13	0.01	0.11	0.02	0.15	0.02	0.15
Ever	0.05	0.21	0.03	0.19	0.06	0.24	0.06	0.23
Other Drugs
Current	0.02	0.14	0.02	0.14	0.04	0.28	0.04	0.29
Ever	0.05	0.22	0.05	0.22	0.05	0.22	0.05	0.22
Total
Current	0.35	0.85	0.33	0.82	0.41	0.97	0.48	1.03
Ever	0.92	1.33	0.89	1.26	0.95	1.36	1.03	1.34
Drug Offers	0.94	1.63	0.9	1.59	0.94	1.72	1.04	1.75

^*Current use was defined by having used the drug in the past 30 days (0= not currently using, 1= currently using).

^**Ever use was defined as having used the drug either in the past 30 days or had used but not in the past 30 days (0 = never used, 1 = used).

^***Total was defined by adding up the 9 drug types, and the sum could range from 0 to 9.

^****Drug offers could range from 0, meaning that no one had tried to give or sell them illegal drugs, to 6, meaning that someone had tried to give or sell them illegal drugs 51 or more times

Time

Time was modeled as a level-1 variable and represents the wave of assessment.

Friend tobacco users

The presence of friend tobacco users in the student’s life was calculated as a continuous variable based on the response to the question: “How many of your four closest friends use tobacco? (None, 1, 2, 3, and 4).”

Level-2 Variables

All level-2 variables represents stable student-level characteristics and were also derived from self-report data obtained from the student survey.

Grade

Grade was determined from the grade the student was in at the start of the study in 2002. Grade was grand mean centered.

Race

Race was determined from responses to the questions “Are you of Latino or Hispanic origin?” (Yes or No) and “How do you describe yourself? Mark all that apply: Asian, Black/African American, Middle Eastern, Native American/Alaskan Native, Native Hawaiian/Other Pacific Islander, White/Caucasian, Other. Because the majority of students were White, African American, or Latino, this variable was reduced to four categories (i.e., White, African American, Latino, and Other). If a student indicated that he/she was of Latino or Hispanic origin, then the person would be placed in the Latino category. For the present analyses, this variable was indicator (i.e., dummy) coded by creating dichotomous variables, indicating African American, Latino, and Other. Therefore, in all analyses, White youth are the reference group for each of the three dummy coded variables.

Gender

Gender was coded as a dichotomous variable determined from responses to the question: “What is your gender? (Female or Male).” Females were coded as 0 and males as 1.

Adult tobacco users

The presence of an adult tobacco user in the home was calculated as a dichotomous variable determined by the response to the question: “Is there an adult (someone over 18 years old) living in your home who uses tobacco? (Yes or No).” No was scored as 0 and yes as 1.

Level-3 Variables

The level-3 variables represent community-level constructs.

Experimental versus Control condition

Twenty-four towns were randomly assigned into two conditions. Both conditions focused on reducing minors’ access to commercial sources of tobacco, but the E communities also increased their PUP law enforcement, whereas in the C communities, PUP law enforcement remained at low levels. The 12 E towns were assigned a score of 1, whereas the 12 C towns were assigned a score of 0.

Proportion of Commercial Tobacco Sales to Youth

Assessments of the proportion of commercial tobacco sales to youth occurred in year 2 and in year 4. A standardized procedure using female adolescents aged 15 or 16 to make cigarette purchases was used to estimate commercial tobacco availability to youth (see Jason et al., 2008). The average of the two assessments carried out by our research team represents the proportion of commercial tobacco sales to youth variable used in the final analyses.

Household Income

The median household income in thousands of dollars for each town was coded as a continuous variable based on the 2000 Census data. This variable was grand mean centered (M = $59,725.50; SD = $20,785.06) to represent the mean household income across the towns.

Statistical Analysis

A random coefficient multilevel analysis was performed using HLM 6.03 (Raudenbush, Bryk, & Congdon, 2006). This analytical approach was selected due to the multilevel data (i.e., observations within individuals within towns). Because outcomes were skewed, we selected a poisson distributed model, with constant exposure.

Our interpretation focused on the population-average model as it tests for an intervention effect averaging across towns. Centering decisions impact the estimation and interpretation of coefficients and the stability of models (Raudenbush & Bryk, 2002). The goals of the current study were largely in accordance with the incremental approach to centering. Grand mean centering controls for these lower level variables, so we settled on a grand mean centering approach, except in the case of dummy variables, which are entered in an original, raw score metric. Treatment condition (E vs. C) at level-3 was uncentered, but the other level-3 variables were grand mean centered.

Because friends who use tobacco might change over time, we placed friends as a level-1, time-varying covariate. At levels-2 and -3, the intercept was allowed to randomly vary, accounting for random variability in the outcome measures across individuals and towns. The wave slope was also modeled as random at level-2, based on our prediction that individuals would vary in the likelihood that they would smoke over time. At level-2 (i.e., person-level), we included grade, race, gender, and adult tobacco users as covariates. At level-3 (i.e., town-level), we included experimental condition, town household income, and commercial availability of tobacco to minors.

Results

The E and C towns did not differ significantly at baseline with respect to community-level data (population size, median household income, PUP law citation rates, commercial illegal sales of tobacco to minors), and student-level data (gender, race, grade level, the presence of adult smokers in the home, or friends who use tobacco).

Those students who stayed in the study over time tended to be at less risk of being a smoker, and were more likely to be in lower grade levels and had fewer friends who smoked. It is certainly understandable that those who stay in a study tend to be less risky, but that occurred equally for both E and C conditions. Even though we had attrition, and this occurred at higher rates for some of our participants (e.g., those who were smokers), the rates of attrition occurred similarly across conditions. In addition, we controlled for a number of demographic variables in the analyses (see level 2 variables), and therefore, our HLM analyses controlled for these potential biases.

Current Drug Use

Results from an unconditional model revealed significant between-town variation in current drug use (p < .01). Variation between towns remained even after controlling for whether a town was in the treatment condition, indicating clustering and confirming the need for a multilevel analytic strategy. In the next step all student-level (i.e., level-2) and the town-level (i.e., level-3) variables were added to the model.

Controlling for a variety of individual-level variables (described below), the odds that a student was currently using drugs at the start of the study did not vary by treatment condition. The degrees of freedom for the t statistic change due to whether the error terms at upper levels are modeled at random or fixed effects. A number of individual factors increased the likelihood of current drug use at baseline: more friends who used tobacco (OR = 1.58; t(48,473) = 79.14, p < .01), students in higher grades (OR = 1.24; t(25,294) = 34.13, p < .01), an adult tobacco user in the home (OR = 1.42; t(25,294) = 17.11, p < .01), and for girls versus boys (OR = 1.10; t(25,294) = 5.40, p < .01). As compared to Caucasians, more current drug use occurred if the person was Latino (OR = 1.21; t(25,294) = 7.02, p < .01), and less current drug use if Other (OR = .80; t(25,294) = −5.79, p < .01). At baseline, higher rates of current drug use occurred for youth in towns with lower household income (OR = .99; t(20) = −2.08, p < .01). ¹

Over time, the odds that a student had currently used drugs significantly increased (OR = 1.21, t (25,300) = 18.89, p < .01). Figure 1 shows that the effect of time was significantly moderated by treatment condition, such that the odds of currently using drugs over time decreased by 0.96 for students in the E condition versus those in the C condition, t (25,300) = −2.95, p < .01.

Figure 1.

Current Drug Use Over Time

The dependent variable involved adding up the 9 drug types, and could range from 0 to 9

Ever Drug Use

Results from an unconditional model revealed significant between-town variation in ever drug use (p < .01). Variation between towns remained even after controlling for whether a town was in the experimental condition, indicating clustering and confirming the need for a multilevel analytic strategy. In the next step, all student-level (i.e., level-2) and town-level (i.e., level-3) variables were added to the model.

Controlling for a variety of individual-level variables (described below), the odds that a student had ever used drugs at the start of the study did not vary by treatment condition. A number of individual factors increased the likelihood of ever drug use at baseline: more friends who used tobacco (OR = 1.37; t(48,473) = 78.63, p < .01), students in higher grades (OR = 1.16; t(25,294) = 36.79, p < .01), an adult tobacco user in the home (OR = 1.43; t(25,294) = 26.05, p < .01), and for girls versus boys (OR = 1.05; t(25,294) = 3.70, p < .01). As compared to Caucasians, more ever drug use occurred if the person was African-American (OR = 1.11; t(25,294) = 3.97, p < .01) or Latino (OR = 1.27; t(25,294) = 12.84, p < .01), and less ever drug use if Other (OR = .83; t(25,294) = −7.56, p < .01). At baseline, higher rates of ever drug use occurred for youth in towns with lower household income (OR = .99; t(20) = −3.71, p = .05).

Over time, the odds that a student had ever used drugs significantly increased (OR = 1.13, t (25,300) = 18.11, p < .01). Figure 2 shows that the effect of time was significantly moderated by treatment condition, such that the odds of ever using drugs over time decreased by 0.98 for students in the E condition versus those in the C condition, t (25,300) = −2.22, p < .01.

Figure 2.

Ever Used Drugs Over Time

The dependent variable involved adding up the 9 drug types, and could range from 0 to 9

Illicit Drug Offers

Results from an unconditional model revealed significant between-town variation in illicit drug offers (p < .01). Variation between towns remained even after controlling for whether a town was in the treatment condition, indicating clustering and confirming the need for a multilevel analytic strategy. In the next step, all student-level (i.e., level-2) and the town-level (i.e., level-3) variables were added to the model.

Controlling for a variety of individual-level variables (described below), the odds of illicit drug offers at the start of the study did not vary by treatment condition. A number of individual factors increased the likelihood of illicit drug offers at the baseline: more friends who used tobacco (OR = 1.38; t(47,922) = 63.11, p < .01), students in higher grades (OR = 1.25; t(25,294) = 41.07, p < .01, an adult tobacco user in the home (OR = 1.55; t(25,294) = 23.24, p < .01), and for girls versus boys (OR = 1.34; t(25,294) = 17.57, p < .01). As compared to Caucasians, higher rates of illicit drug offers occurred if the person was African-American (OR = 1.43; t(25,294) = 9.82, p < .01) or Latino (OR = 1.35; t(25,294) = 11.83, p < .01), and less if Other (OR = .85; t(25,294) = −4.84, p < .01).

Over time, the odds of illicit drug offers significantly increased (OR = 1.17, t (25,300) = 19.53, p < .01. Figure 3 shows that the effect of time was significantly moderated by treatment condition, such that the odds of illicit drug offers over time decreased by 0.96 for students in the E condition versus those in the C condition, t (25,300) = −3.81, p < .01.

Figure 3.

Illicit Drug Offers Over Time

Illicit drug offers could range from 0, meaning that no one had tried to give or sell them illegal drugs, to 6, meaning that someone had tried to give or sell them illegal drugs 51 or more times

Discussion

The present study found that the change in the likelihood of a student reporting current drug use, ever drug use, or illicit drug offers differed over time by condition, with greater increases in the C towns than in the E towns. From the baseline wave to the end of the study, there were higher increases in current drug use for students in the C towns (M = .15, from .33 to .48) than those in the E towns (M = .06, from .35 to .41). For ever drug use, there were higher increases in ever drug use for those students in the C (M =.14; from .89 to 1.03) than E (M = .02, from.93 to .95) towns. Finally, reports of illicit drug offers increased in the C towns (M = .14, from .90 to 1.04), whereas reports in the E towns did not increase (M = 0, from .94 to .94). While these results are modest, the current study contributes to the growing literature supporting the potential effectiveness of enforcing PUP laws (Jason et al., 2003, 2008; Langer & Warheit, 2000).

Clearly, the figures indicate that drug use and illicit drug offers rose in both groups for the first three waves and then dropped in the fourth wave. The increases were smaller for the E condition in contrast to the C condition, and the drop over Wave 4 was higher for those in the E condition. The fact that drug use increased over the first 3 waves is in part due to the fact that this was a multiple cohort design, and proportion of older students increase across the first three waves, as Wave 1 had 7^th through 10^th grade students, Wave 2 had 7^th through 11^th grade students, and Waves 3 and 4 had 7^th through 12^th grade students. In other words, the grade levels surveyed changed over the first 3 waves, and only the last two surveys (Waves 3 and 4) included all grades 7–12. The change in grades could explain the increase in rates of drug use over the first 3 waves of surveys, as there were older children in each of these waves, and drug use increases with age. The changes in grade composition is might also help explain the drop in drug use patterns from Wave 3 to Wave 4, as both of these waves were comprised of students in grades 7–12, and it is possible that E condition had a larger influence on actually reducing drug use. It is also important to note that both groups did have an intervention, and there was no pure control group, so it is possible that the intervention effects for both conditions took several years to have their affects. In addition, in Figure 2, the ever drug use decline was especially steep from Wave 3 to 4, which was counter to expectations. Because the sampling strategy for the multiple cohort design new participants were recruited each year, there were new students in the study during Wave 4, and this might have also contributed to the reductions that are noted in the Figure 2.

Examining Table 1, alcohol is the most frequently used drug, and it could be argued that our analyses should have involved current and ever use of alcohol alone. We did such an analysis, and again found the slopes for the E and C condition were significantly different over Waves 1 through 4 (because the outcome was dichotomous, e.g., whether or not someone was currently using alcohol or had ever used alcohol, a Bernoulli model was selected, which specifies a binomial distribution and a logit-link function).

There are several possible explanations for the observed lower levels of drug use and illicit drug offers to minors in towns with enforcement of tobacco PUP laws. The act of being punished for tobacco-related crimes might deter individual youth from engaging in more delinquent behaviors such as possessing and using other drugs. Results from other studies have highlighted the utility of targeting multiple substances to prevent escalation into harder drugs (e.g., alcohol, tobacco, marijuana; Scheier, Botvin &, Griffin, 2001). Also, the awareness that police officers approach youth and provide punishments for illegal behaviors like public smoking may deter youth or even adults from selling drugs in these communities.

As studies on “broken window” enforcement tactics have shown, the enforcement of lower-level crimes can have a deterrent effect on more serious offenses. Investigators have found that strict enforcement of “quality of life” offenses and increases of misdemeanor arrests were negatively associated with gun-related homicides (Messner et al., 2007). If possession of cigarettes is prohibited, it might discourage youth from possessing other types of illicit drugs. According to this theory, creating an environment where youth cigarette use is not tolerated might create an unfavorable environment for drug use. Similar to explanations for why police tactics may have reduced gun-related violence in other cities in part by “taking guns of the streets” (Messner et al., 2007), PUP polices might also increase the contact between youth and police, offering them the opportunity to search and confiscate illegal drugs that might otherwise have been given or sold to youth in the community.

It is also possible that those who chose to approach minors to sell them drugs do not select potential buyers at random. For example, we have been informed by our police contacts that one signal to a drug dealer that a youth might be wanting and willing to buy drugs is if they are publicly smoking cigarettes. If this were to occur, enforcement of PUP laws might reduce the amount of publicly visible youth smoking in a community and therefore a community may appear to a drug dealer to have fewer potential customers. This reduced visibility may decrease the effects of modeling and minimize the perception of illegal behavior as normal and acceptable behavior within the community (Alesci, Forster, & Blaine, 2003). If this interpretation is accurate, not only would some individual smoking youths be approached less (because the dealers can’t tell that they are smokers because they aren’t smoking in public), but also a community that has few youth smoking publicly may get passed over by dealers looking for adolescent buyers, and thereby, many adolescents would have a lesser probability of being solicited.

Alternatively, it is possible that the findings reflect reduced offers of alcohol or other drugs from friends rather than drug dealers. The finding that girls reported such offers more than boys seems consistent with this explanation. This could happen because reductions in use of tobacco spread to other substances, especially alcohol. In fact, it is possible that the results are driven by alcohol, which could also have been affected by increased citations for possession of tobacco. Use of these substances tends to co-occur and if the police crack down on tobacco, they might also discourage use of alcohol. Findings in Table 1 would be somewhat supportive of such an explanation.

A number of other characteristics relate to the substance use behaviors of youth in this sample. It was understandable that the likelihood of current and ever drug use as well as illicit drug offers at baseline was higher for students in higher grades, those with more friends who used tobacco, and those who reported an adult tobacco user in the home. For example, children who are older and exposed to more smokers, either through their family members or friends, are more likely to use more tobacco (Pokorny, Jason, & Schoeny, 2006) and other drugs. Racial/ethnic differences were also observed. As compared to Caucasians, more current and ever drug use, as well as more illicit drug offers occurred if the person was Latino; less drug use was observed if they reported being Other. In addition, as compared to Caucasians, African-Americans reported higher levels of ever drug use and more illicit drug offers. Additionally, rates for all three variables were higher for girls than boys.

There are several limitations in this study. Because we needed to obtain active consent, we were only able to recruit about 50% of the available youth. The utilization of active parental consent might have resulted in a lower risk sample (Rojas, Sherrit, Harris, & Knight, 2008), and may limit the generalizability of the findings. It is also important to recognize that the results are only generalizable to towns, but this is the appropriate unit for the community-level tobacco policy intervention. Additionally, losses to follow-up were high and we did not obtain any biochemical confirmation of self-reported tobacco abstinence. In the same way that it is possible that the intervention may have lead youth to feel that drug use is less acceptable and deter them from using, it might have also deterred them from honestly reporting use. It is possible that there was more underreporting over time in the E communities. Finally, it is unclear what the longer term influence of PUP laws on drug use might be after youth finish high school.

This study provides preliminary evidence that police efforts to reduce specific substance use behaviors (i.e. underage tobacco use) might have a positive spillover effect into other high risk activities. Given the co-occurrence of substance use, strategies that strengthen community norms against youth tobacco use might work synergistically to help reduce teen drug use and illicit drug offers in communities. Interventions to reduce illegal tobacco use should monitor other substance use behaviors, not only to prevent iatrogenic effects, but also to identify whether the strategies can have a positive influence on other illegal activity. In order to better understand the relationship between strict tobacco enforcement and drug use, it would be useful for future studies to observe this relationship in various settings, communities, and environments.

Biography

Leonard A. Jason is a Professor of Psychology at DePaul University, and the Director of the Center for Community Research. He was the PI of the NCI grant that provided the financial support for this study. His research interests are in tobacco control, recovery homes, and chronic fatigue syndrome. Ljason@depaul.edu

Steven B. Pokorny is an Assistant Professor of Psychology at the University of Florida. He served as the Project Director of the study reported upon in this article. His interests include tobacco control, passive smoking, and cardiovascular risk issues. spokorny@hhp.ufl.edu

Monica Adams is an advanced doctoral graduate student in the Community Psychology program at DePaul University. She served as the community outreach worker on the grant described in this study. Her interests include health psychology, community participatory methods, and tobacco control. madams8@depaul.edu

Annie Nihls has a master’s degree and currently works at the School for New Learning at DePaul University dealing with chronic illnesses. She worked as a research associate on the study reported upon in this publication. Anihls@depaul.edu

Hyo Yeon Kim is an undergraduate at DePaul University. She worked as a volunteer research assistant on this study. Her interests are in social psychology.

Yvonne Hunt currently works at the National Cancer Institute. She worked as a project director for the study reported on in this paper. She has interests in tobacco control and other areas of addictions. yvonnehunt@gmail.com