- Email: [email protected]
The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use
YPMED-04519; No of Pages 6 Preventive Medicine xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use Michael F. Pesko ⁎, Jenna M. Hughes, Fatima S. Faisal Department of Healthcare Policy and Research, Weill Cornell Medical College, Cornell University, LA-216, 402 East 67th Street, New York, NY 10065, USA
a r t i c l e
i n f o
Article history: Received 9 August 2015 Received in revised form 25 January 2016 Accepted 1 February 2016 Available online xxxx Keywords: E-cigarettes Electronic nicotine delivery systems Tobacco control Minimum legal purchase age
a b s t r a c t Objective. In the United States, many states have established minimum legal purchase ages for electronic nicotine delivery systems (ENDS) to ban adolescent purchases, but these policies may also affect other related substance use. We explore whether ENDS are substitutes or complements for cigarettes, cigars, smokeless tobacco, and marijuana among adolescents by using variation in state-level implementation of ENDS age purchasing restrictions. Methods. We linked data on ENDS age purchasing restrictions to state- and year-specific rates of adolescent tobacco and marijuana use in 2007–2013 from the Youth Risk Behavior Surveillance System. This data provides a nationally representative sample of adolescents who attend public and private schools. We performed a fixed effect regression analysis exploring the influence of ENDS age purchasing restrictions on outcomes of tobacco use and marijuana use, controlling for state and year fixed characteristics, age-race cohorts, cigarette excise taxes, and cigarette indoor use restrictions. Results. For cigarette use, we separate our results into cigarette use frequency. We found causal evidence that ENDS age purchasing restrictions increased adolescent regular cigarette use by 0.8 percentage points. ENDS age purchasing restrictions were not associated with cigar use, smokeless tobacco use, or marijuana use. Conclusions. We document a concerning trend of cigarette smoking among adolescents increasing when ENDS become more difficult to purchase. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Tobacco use is the single largest cause of preventable disease and death (U.S. Department of Health and Human Services, 2014) in the United States and carries a high financial cost. In the USA alone, tobacco use is estimated to cost $289 billion a year, including $130 billion going to direct medical care for adults and more than $150 billion in lost productivity (U.S. Department of Health and Human Services, 2014). While the adult current cigarette use rate has declined from 40% since the release of the first Surgeon General Report on smoking in 1964, 18.1% of adults (Agaku et al., 2014) and 14% of high school students (Arrazola et al., 2013) still smoke. Each day, more than 3200 persons younger than 18 years of age (hereafter referred to as adolescents) smoke their first cigarette (U.S. Department of Health and Human Services, 2014). Since the majority of smoking begins in adolescence (U.S. Department of Health and Human Services, 2014), preventing cigarette use among this population remains an urgent public health priority.
⁎ Corresponding author. E-mail address: [email protected] (M.F. Pesko).
Electronic nicotine delivery systems (ENDS) are a new class of tobacco products that deliver aerosolized nicotine via a batterypowered device. ENDS products include e-cigarettes, atomizers, vapepens, vape pipes, hookah-pens, e-hookahs, e-vaporizers, e-cigars, and e-pipes. ENDS first entered the United States market in approximately 2007 and have subsequently rapidly increased in use (Riker et al., 2012). There is a high rate of dual-use between ENDS and cigarettes and 93%–96% of ENDS-using students have tried cigarettes in the past (Wills et al., 2015; Johnston et al., 2014). As of 2014, ENDS officially surpassed cigarettes as the most commonly used tobacco product among adolescents, with rates of current use at 13.4% vs 9.2% respectively (Arrazola et al., 2015; Johnston et al., 2015). While ENDS are not without health risks to adolescents, particularly in terms of nicotine's higher addiction potential (U.S. Department of Health and Human Services, 2012; Apelberg et al., 2014) and lasting brain effects (U.S. Department of Health and Human Services, 2014), studies suggest they are less harmful than cigarettes (Hampton, 2014; Abrams, 2014; Brandon et al., 2015; McNeill et al., 2015). A study by the British government suggests that ENDS use has 5% of the harm of cigarette use (McNeill et al., 2015). In nonsmoker adults, ENDS use can have multiple health effects similar to, but often less severe than, conventional tobacco smoking (Pisinger and Dossing, 2014; Vardavas
http://dx.doi.org/10.1016/j.ypmed.2016.02.001 0091-7435/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
2
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
et al., 2012). Smokers may experience fewer adverse health effects from ENDS than from cigarettes (Flouris et al., 2012; Farsalinos et al., 2014; Polosa et al., 2014). The chemical profile of ENDS has suggested that ENDS expose individuals to lower concentrations of tobacco-specific toxicants than cigarettes (Goniewicz et al., 2014; Cheng, 2014; Orr, 2014). A central issue in determining health implications of ENDS is understanding how they promote or discourage adolescent cigarette use. In isolation, the substitution of one potentially less harmful product (ENDS) for a known harmful product (cigarettes) would be a health-improving outcome, and the reverse would be a worse outcome. Substitution can be explored by increasing monetary costs (e.g. price increases) or increasing non-monetary costs (e.g. age purchasing restrictions) of one good and observing if this raises or lowers consumption of another good (Etter, 2015). Early published literature offers differing views on if ENDS and cigarettes are substitutes: one study found no substitution using prices and retail data (Huang et al., 2014), and a second study suggested cigarette price increases raise the likelihood of trying e-cigarettes using experimental data (Grace et al., 2015). Cigarettes have been found to be complements with marijuana (Chaloupka et al., 1999; Farrelly et al., 2001) and smokeless tobacco (Tauras et al., 2007), raising the question of whether ENDS are complements or substitutes for alternative tobacco products and marijuana as well. Another factor to consider in understanding the relationship between ENDS and marijuana is that some ENDS devices can be used to vaporize other substances, such as marijuana, opium, and crack cocaine (Young-Wolff et al., 2014; Marynak et al., 2014). Using ENDS to vaporize marijuana can decrease the odor and allows it to be used more discretely (Etter, 2015), which may promote a complement relationship between the two. ENDS minimum legal purchase ages are a recent policy, which if properly enforced, could make it more difficult for adolescents to obtain ENDS. For example, adolescents may now be forced to purchase ENDS devices through an intermediary or online (Williams et al., 2015). Therefore, the implementation of these restrictions can be used to explore if ENDS are substitutes or complements with cigarettes and marijuana because the non-monetary costs of purchasing ENDS products has increased (Williams et al., 2004). New Jersey implemented the first minimum legal purchase age in March, 2010, and as of November 2014, 40 states had age restrictions in place (Marynak et al., 2014). The minimum legal purchasing age among these states is 18, except for four states (Alabama, Alaska, New Jersey, and Utah) in which the age is 19 (Marynak et al., 2014). As of 2014, cigarettes had the same minimum legal purchase age in all states with ENDS age purchasing restrictions (Centers for Disease Control and Prevention, 2014a). The Food and Drug Administration, Center for Tobacco Products (FDA-CTP) has proposed a national ENDS minimum purchasing age of 18, but this has not yet been implemented. In this study, we explore possible substitution effects within a vulnerable adolescent population that is more likely to initiate cigarette use than adults. We use access restrictions from state-level variation in ENDS age purchasing restrictions to explore if ENDS are economic substitutes or complements for cigarettes (e.g. does cigarette consumption increase or decrease as ENDS become more difficult to access). ENDS age purchasing restrictions were enacted in different states at different times, providing policy variation akin to a natural experiment. We explore the effect that these policies have had on cigarette use and marijuana use. We do not explore the effect of these policies on ENDS use because state-level data on ENDS use is not yet available over multiple waves for adolescents. 2. Methods We use state-level aggregated data from the Youth Risk Behavior Surveillance System (YRBSS) (Centers for Disease Control and Prevention, 2014b). YRBSS is a biennial national school-based survey of adolescents conducted by Centers for Disease Control and Prevention (CDC) and state health
departments to monitor health risk behaviors. It utilizes a two-stage, cluster sample design to produce a representative sample of students in grades 9–12 in its jurisdiction, and data are weighted to adjust for school and student nonresponse and to make the data representative. The data are typically collected between February and May of each year (Brener et al., 2013). We extract the data from 2007, 2009, 2011, and 2013 to correspond to a period before and during the implementation of ENDS age purchasing restrictions. ENDS first entered the US market in 2007 (Friedman, 2015). To explore how ENDS age purchasing restrictions are associated with the use of tobacco and marijuana, we use four different measures of cigarette smoking intensity, and measures of cigar use, smokeless tobacco use, and marijuana use. The smoking intensity measures that we use are: 1) recent cigarette use (at least 1 day over the past 30 days), 2) casual cigarette use (between 1 and 19 days over the past 30 days) 3) regular cigarette use (at least 20 days over the past 30 days), 4) and heavy cigarette use (every day over the past 30 days). We also use recent (past 30 days) cigar use, smokeless tobacco use, and marijuana use as outcome measures. We obtained data on ENDS age purchasing restrictions from a report by the CDC (Marynak et al., 2014). We matched this data using the February–May time window over which students were interviewed. Wyoming changed their policy halfway through the data collection in 2013, and we treat this state as if the policy had been in effect the full year. We exclude the state of Massachusetts from our analysis because age purchasing restrictions were implemented at the county level at different points in time (Global Advisors on Smokefree Policy (New Jersey GASP), 2015). In addition to ENDS age purchasing restrictions, we match the following onto the YRBSS data and use this as control variables in our regression analysis: racial/ethnic composition of adolescents, the combined state and federal cigarette excise taxes, and cigarette indoor use laws. We obtained state-level cigarette excise taxes from the State Tobacco Activities Tracking and Evaluation (STATE) System maintained by the CDC (Centers for Disease Control and Prevention, 2014a), and added the federal excise taxes, which were $0.39 in 2007, and $1.01 in 2009, 2011, and 2013. We adjust the tax levels for inflation. We also obtained state-level cigarette indoor use restrictions for restaurants, workplaces, and bars from the STATE system. For racial/ethnic composition of adolescents, we used Survey of Epidemiology and End Results (SEER) population data to construct three state-level ratios: White, other race, and Hispanic 10–19 year olds as a share of all adults (Roth, 2015). We used 19 as the upper bound because of how the data were grouped. For cigarette indoor use restrictions, the STATE system only provides data through 2012, so we impute 2013 data using 2012 data. To explore the association of ENDS age purchasing restrictions on these substance use measures, we estimate a state and year fixed effects regression model. The inclusion of state and year fixed effects allows us to exploit variation in ENDS age purchasing restrictions independent of state-level time-invariant unobservable characteristics, and unobservable characteristics unique to a particular year. For example, anti-smoking sentiment and non-changing population characteristics unique to a particular state from 2007 to 2013 is controlled by the inclusion of a state fixed effects. The national introduction of ENDS over time is controlled for by year fixed effects. We also include separate linear time trends for states enacting ENDS age purchasing restrictions and states not enacting age purchasing restrictions, which control for time-varying omitted variables in these two groups of states, such as changing antismoking sentiment. We perform the regression analysis using 4 annual observations for states with complete information from 2007 to 2013. Cigarette and marijuana use information is provided in all years for 33 states (and 34 for select outcomes), cigar use is provided in all years for 25 states, and smokeless tobacco is provided in all years for 28 states. In Table 1, we show the variation in ENDS age purchasing restrictions by state and year for the largest (34 states) and smallest (25 states) samples. In addition to measuring the association of ENDS age purchasing restrictions on tobacco and marijuana use measures during the period immediately after implementation, we also examine the association on outcomes 2 years before a restriction was in place. If significant associations are found before the policy was implemented, this may indicate that restrictions were endogenously enacted in states with changing cigarette use relative to non-implementing states, or some other omitted variable bias. If significant associations are not found in the prior period, then we argue any associations that we find in the period in which the policy was implemented exhibit evidence of causality. We also examine the association that restrictions have in the 2 years after implementation to observe any lagged effects.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
3
Table 1 Presence of ENDS age purchasing restriction between February and May of a given year. State
AL AKa AZ ARa CT DEa FL GAa HI IDa ILa KSa KYa LAa MEa MDa MIa MSa MTa NHa NMa NY NC NDa OKa RIa SCa SD TNa TXa UTa WVa WIa WY Total
Presence of age purchasing restriction 2007
2009
2011
2013
N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N 0
N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N 0
N N N N N N N N N N N N N N N N N N N Y N N N N N N N N N N Y N N N 2
N Y N N N N N N N Y N Y N N N Y N N N Y N Y N N N N N N Y N Y N Y Y 10
Fig. 1. Recent cigarette use in states with and without an ENDS age purchasing restriction before or during 2013. Note: Estimates are population-adjusted. The vertical line indicates the first ENDS age purchasing restriction in March, 2010 in New Jersey. By 2013, 24 of 50 states had an ENDS age purchasing restriction in place.
3. Results In Table 2, we show that the four adolescent cigarette use measures consistently declined from 2007 to 2013. Recent cigarette use declined from 19.0% to 13.7%, casual cigarette use declined from 10.9% to 9.1%, regular cigarette use declined from 8.0% to 4.5%, and heavy cigarette use declined from 6.1% to 3.4%. While regular and heavy cigarette use were roughly halved between 2007 and 2013, casual cigarette use declined by only 16.4%. In contrast, recent use of marijuana did not decline in 2013 (20.9%) compared to 2005 (18.9%). By 2013, adolescents were 52.3% more likely to recently use marijuana than cigarettes. Inflation-adjusted cigarette excise taxes rose by 59.6%, from $1.61 in 2007 to $2.57 in 2013. Table 2 also shows variation in states implementing ENDS age purchasing restrictions. Six states enacted this policy between March, 2010 (starting in New Jersey) and May, 2011. Fourteen additional states enacted this policy from June, 2012 to May, 2013 (Marynak et al., 2014). We find limited graphical evidence that ENDS may be economic substitutes for cigarettes and economic complements for marijuana. In Fig. 1, we plot population-adjusted recent cigarette use among adolescents for states that enacted an ENDS age purchasing restriction before or during 2013 against those states that did not. The vertical line indicates the first age purchasing restriction in New Jersey. We observe that between 2007 and 2009, recent cigarette use was declining faster in states that would later implement an ENDS age purchasing restriction. This trend reverses from 2009 to 2011, with the recent cigarette use rate remaining virtually identical in states adopting ENDS age purchasing restrictions, but falling in states that did not. From 2011 to 2013, the cigarette rate appears to decline approximately equally in both groups of
Note: We include only states in the 34 state sample for which YRBSS recent cigarette use information is available for all 4 years. a Indicates states also in the 25 state sample for which YRBSS cigar use information is available in all 4 years.
Our method of exploring how increasing the non-monetary costs of obtaining one type of tobacco product (ENDS) influences consumption of another tobacco product (cigarettes) frees us from needing data on both ENDS and cigarettes. We documented in our introduction that these products are often dual-used. However, there is no reason why increasing non-monetary costs of ENDS should increase ENDS use. This frees us to explore a complement/substitute relationship without data on ENDS, although we cannot determine if changes in ENDS consumption occur at the level of wholesale participation (extensive margin), or frequency of use (intensive margin). We performed all analyses in 2015 using Stata 13.1. Standard errors are clustered at the level of state.
Table 2 Descriptive statistics from the Youth Risk Behavior Surveillance Survey.
Recent cigarette use (at least 1 day over the past 30 days) (%) Casual cigarette use (between 1 and 19 days over the past 30 days) (%) Regular cigarette use (at least 20 days over the past 30 days) (%) Heavy cigarette use (every day over the past 30 days) (%) Current cigar use (%) Current smokeless tobacco use (%) Current marijuana use (%) Average real cigarette excise tax (state and federal) ($) Number of states with ENDS age purchasing restrictions by May of each year
States (N)
2007
2009
2011
2013
Change from 2007 to 2013
34 33 33 33 25 28 34 50 50
18.97 10.88 7.95 6.09 14.54 8.16 18.94 1.61 0
17.92 10.97 6.88 4.91 14.81 8.96 19.88 2.52 0
16.56 10.30 6.09 4.57 14.64 8.63 21.12 2.56 6
13.72 9.10 4.51 3.36 13.01 8.86 20.90 2.57 14
−27.68% −16.36% −43.27% −44.83% −10.52% 8.58% 10.35% 59.63% -
Notes: Population-weighted averages are reported. Cigarette excise taxes are in 2013 dollars.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
132 33 0.332 10.208 136 34 0.609 17.177 Observations States Adjusted R-Sq. Mean use rate
Notes: The regression results control for racial/ethnic adolescent composition, cigarette excise taxes, cigarette indoor use laws, state fixed effects, year fixed effects, and linear time trends for adopting and non-adopting states. 95% Confidence intervals indicated in brackets. Significance levels are as noted. ⁎ p b 0.10. ⁎⁎ p b 0.05.
136 34 0.365 20.038 112 28 0.259 9.853 100 25 0.546 13.615 132 33 0.766 5.082
[−0.13, 1.42] 0.76 [−0.67, 2.20] [0.02,1.59] 1.17 [−0.82,3.16] [−0.22, 4.85] 0.77 [−0.71, 2.24] [0.20, 5.95] 1.88⁎ [−0.20, 3.96]
132 33 0.801 6.838
0.15 [−1.64, 1.93] −0.75 [−3.09, 1.59] −1.39 [−3.31, 0.53] −0.23 [−1.80, 1.35] 0.55 [−1.17, 2.26] 0.15 [−1.16, 1.46] −0.12 [−0.77, 0.53] 0.65⁎ 1.96⁎⁎ [0.02, 3.89] 2.31⁎ 1.97⁎ [−0.10, 4.05] 3.08⁎⁎
Previous period ENDS age purchasing restriction Current period ENDS age purchasing restriction Next period ENDS age purchasing restriction
0.16 [−0.50,0.81] 0.80⁎⁎
Recent smokeless tobacco use (at least 1 day over the past 30 days) Heavy cigarette use (every day over the past 30 days) Regular cigarette use (at least 20 days over the past 30 days) Casual cigarette use (between 1 and 19 days over the past 30 days)
We matched policy variation akin to a natural experiment to longitudinal data on tobacco use and marijuana use, and found that cigarette use increased as ENDS became harder to legally purchase due to age purchasing restrictions. However, we failed to detect statistically
Recent cigarette use (at least 1 day over the past 30 days)
4. Discussion
Table 3 Fixed effects regression analysis of cigarette and marijuana use with previous and next period restrictions on ENDS sales to adolescents, 2007–2013.
states. In Fig. 2, we plot recent marijuana use among adolescents. This graphical evidence suggests that marijuana use remained virtually the same after 2009 in states that adopted ENDS age purchasing restrictions, but increased in non-adopting states. In Table 3, we explore if the relationships we observe in the figures are present after controlling for racial/ethnic adolescent composition, cigarette excise taxes, cigarette indoor use laws, state fixed effects, year fixed effects, and linear time trends for adopting and nonadopting states. We find evidence that the ENDS age purchasing restrictions are associated with a 3.1 percentage point (17.9% of the mean) increase in adolescent cigarette use (p b 0.05) in the period of implementation. Most of this effect is accounted for within casual cigarette using adolescents (2.3 percentage point increase, 22.6% of the mean) (p b 0.10). However, for both of these outcomes we observe that cigarette use was increasing in the two-year period prior to the ENDS age purchasing restriction being implemented, which prevents us from making claims of causality due to the presence of potential policy endogeneity. For regular cigarette use (at least 20 days over the past 30 days) and heavy cigarette use (every day over the past 30 days), we do not observe the same increase in the two-year period prior to the ENDS age purchasing restriction being implemented. In the current period, we observe that ENDS age purchasing restrictions are associated with an increase in regular cigarette use of 0.8 percentage points (11.7% of the mean) (p b 0.05), and an increase in heavy cigarette use of 0.7 percentage points (12.8% of the mean) (p b 0.10). While the coefficients remain of similar magnitude in the two-year period after the policy implementation, these associations were not statistically significant. In sum, we find associational evidence of ENDS age purchasing restrictions increasing cigarette use at all levels of intensity, but we only find causal evidence that these restrictions increased cigarette use among heavier cigarette users. These results suggest that ENDS may be substitutes for cigarettes because limiting ENDS purchasing opportunities is associated with increases in cigarette use. In the last three columns on Table 3, we do not find evidence of ENDS age purchasing restrictions being associated with recent cigar use, recent smokeless tobacco use, or recent marijuana use.
Recent cigar use (at least 1 day over the past 30 days)
Fig. 2. Recent marijuana use in states with and without an ENDS age purchasing restriction before or during 2013. Note: Estimates are population-adjusted. The vertical line indicates the first ENDS age purchasing restriction in March, 2010 in New Jersey. By 2013, 24 of 50 states had an ENDS age purchasing restriction in place.
0.25 [−1.11, 1.61] 0.50 [−1.50, 2.49] −1.29 [−3.21, 0.62]
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx Recent marijuana use (at least 1 day over the past 30 days)
4
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
significant increases in cigar use, smokeless tobacco use, or marijuana use. Our results suggest that adolescents are willing to substitute ENDS for cigarettes depending on legal purchasing opportunities of ENDS. We do not suspect that this substitution was complete in the sense that individuals consuming ENDS now only consume cigarettes. It is possible, and likely given statistics on dual-use (Johnston et al., 2014), that adolescents continue to use ENDS after minimum legal purchase ages are implemented, but with less frequency. Our finding of a 0.8 percentage point increase in cigarette use among regular cigarette users in response to ENDS age purchasing restrictions is consistent with a study by Friedman (2015) that found a 0.9 percentage point increase in cigarette use, using data from the National Survey on Drug Use and Health and a difference-in-difference empirical approach similar to our own. Both studies found evidence of substitution effects. One key difference between these studies is that Friedman's data provided all states, whereas our study used at most 34 states due to data availability in the YRBSS. Another difference is that in our study we explored the effect of ENDS age purchasing restrictions for a wider array of outcomes, including cigarette use at different levels of intensity, cigar use, smokeless tobacco use, and marijuana use. There are at least four limitations of our study. First, we are unable to explore the effect of ENDS age purchasing restrictions on ENDS use itself because no state-level panel data on adolescent ENDS use exists currently. While this limitation does not prevent us from exploring a substitution/complement relationship between ENDS and cigarettes, it does prevent us from understanding if ENDS consumption was reduced on the participation (extensive) or frequency of use (intensive) margins. Second, our data are limited by having at most 34 states in our sample. Third, we do not consider local ENDS age purchasing restrictions and the effectiveness of these restrictions. Along these same lines, if adolescents are able to obtain ENDS online or elsewhere with minimal additional difficulty (non-monetary costs) following the implementation of ENDS age purchasing restrictions, this will bias toward the null the measured influence of these restrictions on our outcomes (e.g. underestimate the effect size compared to if this policy were fully enforced). Fourth, this investigation does not explore more nuanced relationships between ENDS availability and cigarette use, such as renormalizing smoking. 5. Conclusion All policymaking bodies should be aware of a potential increase in cigarette use following ENDS age purchasing restrictions. The FDA-CTP is currently evaluating the implementation of a national ENDS age purchasing restriction. In the meantime, the 10 state governments that do not have an age purchasing restriction in place as of January 2015 are free to implement this policy. Given that there are known health risks for ENDS that justify states enacting age purchasing restrictions, these policymaking bodies may wish to pair ENDS age purchasing restrictions with policies that raise the price of cigarettes for adolescents, such as increasing enforcement efforts against venders that sell cigarettes to adolescents. State governments also have the power to concurrently increase cigarette excise taxes to prevent increases in cigarette use in response to ENDS age purchasing restrictions. Another option, in the spirit of regulating ENDS proportionate to their risk relative to combustible tobacco (Chaloupka et al., 2015), may be to establish an ENDS minimum legal purchase age lower than the cigarette minimum legal purchase (Friedman, 2015). Conflict of interest and funding The authors declare that there are no conflicts of interest.
5
Contributor statement MF Pesko performed the analysis and drafted the article. All authors conceptualized the study, interpreted the data, and critically revised the article. Transparency document The Transparency document associated with this article can be found in the online version. Acknowledgments We thank individuals at New York State Pediatric Advocacy Conference for helpful comments. References U.S. Department of Health and Human Services, 2014. The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Agaku, I.T., King, B.A., Dube, S.R., 2014. Current cigarette smoking among adults—United States, 2005–2012. MMWR Morb. Mortal. Wkly Rep. 63, 29–34. Arrazola, R.A., Dube, S.R., King, B.A., 2013. Tobacco product use among middle and high school students—United States, 2011 and 2012. MMWR Morb. Mortal. Wkly Rep. 62, 893–897. Riker, C.A., Lee, K., Darville, A., Hahn, E.J., 2012. E-cigarettes: promise or peril? Nurs. Clin. North Am. 47, 159–171. Wills, T.A., Knight, R., Williams, R.J., Pagano, I., Sargent, J.D., 2015. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics 135, e43–e51. Johnston, L.D., O'Malley, P.M., Miech, R.A., Bachman, J.G., Schulenberg, J.E., 2014. Monitoring the Future National Survey Results on Drug Use: 1975–2014: Overview, Key Findings on Adolescent Drug Use. Institute for Social Research, The University of Michigan, Ann Arbor, MI. Arrazola, R.A., Singh, T., Corey, C.G., et al., 2015. Tobacco use among middle and high school students—United States, 2011–2014. MMWR Morb. Mortal. Wkly Rep. 64, 381–385 (Apr. 17). Johnston, L.D., O'Malley, P.M., Miech, R.A., Bachman, J.G., Schulenberg, J.E., 2015. Monitoring the Future National Results on Adolescent Drug Use: Overview of Key Findings, 2014. Institute for Social Research, the University of Michigan, Ann Arbor, Michigan. U.S. Department of Health and Human Services, 2012. Preventing Tobacco Use Among Youth and Young Adults: A Report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Apelberg, B.J., Corey, C.G., Hoffman, A.C., et al., 2014. Symptoms of tobacco dependence among middle and high school tobacco users results from the 2012 National Youth Tobacco Survey. Am. J. Prev. Med. 47, S4–S14. Hampton, T., 2014. Experts call for research plus regulation of e-cigarettes. JAMA 311, 123–124. Abrams, D.B., 2014. Promise and peril of e-cigarettes: can disruptive technology make cigarettes obsolete? JAMA 311, 135–136. Brandon, T.H., Goniewicz, M.L., Hanna, N.H., et al., 2015. Electronic nicotine delivery systems: a policy statement from the american association for cancer research and the american society of clinical oncology. Clin. Cancer Res. 21, 514–525. McNeill, A., Brose, L.S., Calder, R., Hitchman, S.C., Hajek, P., McRobbie, H., 2015. E-Cigarettes: An Evidence Update. Public Health England (Aug). Pisinger, C., Dossing, M., 2014. A systematic review of health effects of electronic cigarettes. Prev. Med. 69, 248–260. Vardavas, C.I., Anagnostopoulos, N., Kougias, M., Evangelopoulou, V., Connolly, G.N., Behrakis, P.K., 2012. Short-term pulmonary effects of using an electronic cigarette: impact on respiratory flow resistance, impedance, and exhaled nitric oxide. Chest 141, 1400–1406. Flouris, A.D., Poulianiti, K.P., Chorti, M.S., et al., 2012. Acute Effects of Electronic and Tobacco Cigarette Smoking on Complete Blood Count. Food and chemical toxicology: an international journal published for the British Industrial Biological Research Association 50 pp. 3600–3603. Farsalinos, K.E., Tsiapras, D., Kyrzopoulos, S., Savvopoulou, M., Voudris, V., 2014. Acute effects of using an electronic nicotine-delivery device (electronic cigarette) on myocardial function: comparison with the effects of regular cigarettes. BMC Cardiovasc. Disord. 14, 78. Polosa, R., Morjaria, J., Caponnetto, P., et al., 2014. Effect of smoking abstinence and reduction in asthmatic smokers switching to electronic cigarettes: evidence for harm reversal. Int. J. Environ. Res. Public Health 11, 4965–4977. Goniewicz, M.L., Knysak, J., Gawron, M., et al., 2014. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob. Control. 23, 133–139. Cheng, T., 2014. Chemical evaluation of electronic cigarettes. Tob. Control. 23 Suppl 2, ii11–ii17.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
6
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
Orr, M.S., 2014. Electronic cigarettes in the USA: a summary of available toxicology data and suggestions for the future. Tob. Control. 23 Suppl 2, ii18–ii22. Etter, J.F., 2015. Electronic cigarettes and cannabis: an exploratory study. Eur. Addict. Res. 21, 124–130. Huang, J., Tauras, J., Chaloupka, F.J., 2014. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob. Control. 23 Suppl 3, iii41–iii47. Grace, R.C., Kivell, B.M., Laugesen, M., 2015. Estimating cross-price elasticity of ecigarettes using a simulated demand procedure. Nicotine Tob. Res. (in press:ntu268). Chaloupka, F.J., Pacula, R.L., Farrelly, M.C., Johnston, L.D., O'Malley, P.M., 1999. Do higher cigarette prices encourage youth to use marijuana? NBER Working Paper No 6939. National Bureau of Economic Research, Cambridge, MA Farrelly, M.C., Bray, J.W., Zarkin, G.A., Wendling, B.W., 2001. The joint demand for cigarettes and marijuana: evidence from the National Household Surveys on Drug Abuse. J. Health Econ. 20, 51–68. Tauras, J., Powell, L., Chaloupka, F., Ross, H., 2007. The demand for smokeless tobacco among male high school students in the United States: the impact of taxes, prices and policies. Appl. Econ. 39, 31–41. Young-Wolff, K.C., Karan, L.D., Prochaska, J.J., 2014. Electronic cigarettes in jails: a panacea or public health problem? JAMA Psychiatry. Marynak, K., Holmes, C.B., King, B.A., Promoff, G., Bunnell, R., McAfee, T., 2014. State laws prohibiting sales to minors and indoor use of electronic nicotine delivery systems—United States. MMWR Morb. Mortal. Wkly Rep. 63, 1145–1150. Williams, R.S., Derrick, J., Ribisl, K.M., 2015. Electronic cigarette sales to minors via the Internet. JAMA Pediatrics 169, e1563.
Williams, J., Pacula, R.L., Chaloupka, F.J., Wechsler, H., 2004. Alcohol and marijuana use among college students: economic complements or substitutes? Health Econ. 13, 825–843. Centers for Disease Control and Prevention, 2014a. State Tobacco Activities Tracking and Evaluation (STATE) System. (Accessed Jun 12 2014, at http://apps.nccd.cdc.gov/ statesystem/Default/Default.aspx. Centers for Disease Control and Prevention, 2014b. Youth Risk Behavior Surveillance System (YRBSS). (Accessed Jun 12 2014, at http://www.cdc.gov/HealthyYouth/yrbs/. Brener, N.D., Kann, L., Shanklin, S., et al., 2013. Methodology of the Youth Risk Behavior Surveillance System—2013. MMWR Recommendations and Reports: Morbidity and Mortality Weekly Report Recommendations and Reports/Centers for Disease Control. 62, pp. 1–20. Friedman, A.S., 2015. How does electronic cigarette access affect adolescent smoking? J. Health Econ. 44, 300–308. Global Advisors on Smokefree Policy (New Jersey GASP), 2015. Electronic smoking devices. (Accessed Dec 21 2015, at http://www.njgasp.org/wp-content/uploads/ 2014/05/E-Cigs_White_Paper.pdf. Roth, J., 2015. Survey of Epidemiology and End Results (SEER) U.S. state and county population data. Natl. Bur. Econ. Res. Accessed Jan 30 2015 at http://www.nber.org/ data/seer_u.s._county_population_data.html. Chaloupka, F.J., Sweanor, D., Warner, K.E., 2015. Differential taxes for differential risks— toward reduced harm from nicotine-yielding products. N. Engl. J. Med. 373, 594–597.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use Michael F. Pesko ⁎, Jenna M. Hughes, Fatima S. Faisal Department of Healthcare Policy and Research, Weill Cornell Medical College, Cornell University, LA-216, 402 East 67th Street, New York, NY 10065, USA
a r t i c l e
i n f o
Article history: Received 9 August 2015 Received in revised form 25 January 2016 Accepted 1 February 2016 Available online xxxx Keywords: E-cigarettes Electronic nicotine delivery systems Tobacco control Minimum legal purchase age
a b s t r a c t Objective. In the United States, many states have established minimum legal purchase ages for electronic nicotine delivery systems (ENDS) to ban adolescent purchases, but these policies may also affect other related substance use. We explore whether ENDS are substitutes or complements for cigarettes, cigars, smokeless tobacco, and marijuana among adolescents by using variation in state-level implementation of ENDS age purchasing restrictions. Methods. We linked data on ENDS age purchasing restrictions to state- and year-specific rates of adolescent tobacco and marijuana use in 2007–2013 from the Youth Risk Behavior Surveillance System. This data provides a nationally representative sample of adolescents who attend public and private schools. We performed a fixed effect regression analysis exploring the influence of ENDS age purchasing restrictions on outcomes of tobacco use and marijuana use, controlling for state and year fixed characteristics, age-race cohorts, cigarette excise taxes, and cigarette indoor use restrictions. Results. For cigarette use, we separate our results into cigarette use frequency. We found causal evidence that ENDS age purchasing restrictions increased adolescent regular cigarette use by 0.8 percentage points. ENDS age purchasing restrictions were not associated with cigar use, smokeless tobacco use, or marijuana use. Conclusions. We document a concerning trend of cigarette smoking among adolescents increasing when ENDS become more difficult to purchase. © 2016 Elsevier Inc. All rights reserved.
1. Introduction Tobacco use is the single largest cause of preventable disease and death (U.S. Department of Health and Human Services, 2014) in the United States and carries a high financial cost. In the USA alone, tobacco use is estimated to cost $289 billion a year, including $130 billion going to direct medical care for adults and more than $150 billion in lost productivity (U.S. Department of Health and Human Services, 2014). While the adult current cigarette use rate has declined from 40% since the release of the first Surgeon General Report on smoking in 1964, 18.1% of adults (Agaku et al., 2014) and 14% of high school students (Arrazola et al., 2013) still smoke. Each day, more than 3200 persons younger than 18 years of age (hereafter referred to as adolescents) smoke their first cigarette (U.S. Department of Health and Human Services, 2014). Since the majority of smoking begins in adolescence (U.S. Department of Health and Human Services, 2014), preventing cigarette use among this population remains an urgent public health priority.
⁎ Corresponding author. E-mail address: [email protected] (M.F. Pesko).
Electronic nicotine delivery systems (ENDS) are a new class of tobacco products that deliver aerosolized nicotine via a batterypowered device. ENDS products include e-cigarettes, atomizers, vapepens, vape pipes, hookah-pens, e-hookahs, e-vaporizers, e-cigars, and e-pipes. ENDS first entered the United States market in approximately 2007 and have subsequently rapidly increased in use (Riker et al., 2012). There is a high rate of dual-use between ENDS and cigarettes and 93%–96% of ENDS-using students have tried cigarettes in the past (Wills et al., 2015; Johnston et al., 2014). As of 2014, ENDS officially surpassed cigarettes as the most commonly used tobacco product among adolescents, with rates of current use at 13.4% vs 9.2% respectively (Arrazola et al., 2015; Johnston et al., 2015). While ENDS are not without health risks to adolescents, particularly in terms of nicotine's higher addiction potential (U.S. Department of Health and Human Services, 2012; Apelberg et al., 2014) and lasting brain effects (U.S. Department of Health and Human Services, 2014), studies suggest they are less harmful than cigarettes (Hampton, 2014; Abrams, 2014; Brandon et al., 2015; McNeill et al., 2015). A study by the British government suggests that ENDS use has 5% of the harm of cigarette use (McNeill et al., 2015). In nonsmoker adults, ENDS use can have multiple health effects similar to, but often less severe than, conventional tobacco smoking (Pisinger and Dossing, 2014; Vardavas
http://dx.doi.org/10.1016/j.ypmed.2016.02.001 0091-7435/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
2
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
et al., 2012). Smokers may experience fewer adverse health effects from ENDS than from cigarettes (Flouris et al., 2012; Farsalinos et al., 2014; Polosa et al., 2014). The chemical profile of ENDS has suggested that ENDS expose individuals to lower concentrations of tobacco-specific toxicants than cigarettes (Goniewicz et al., 2014; Cheng, 2014; Orr, 2014). A central issue in determining health implications of ENDS is understanding how they promote or discourage adolescent cigarette use. In isolation, the substitution of one potentially less harmful product (ENDS) for a known harmful product (cigarettes) would be a health-improving outcome, and the reverse would be a worse outcome. Substitution can be explored by increasing monetary costs (e.g. price increases) or increasing non-monetary costs (e.g. age purchasing restrictions) of one good and observing if this raises or lowers consumption of another good (Etter, 2015). Early published literature offers differing views on if ENDS and cigarettes are substitutes: one study found no substitution using prices and retail data (Huang et al., 2014), and a second study suggested cigarette price increases raise the likelihood of trying e-cigarettes using experimental data (Grace et al., 2015). Cigarettes have been found to be complements with marijuana (Chaloupka et al., 1999; Farrelly et al., 2001) and smokeless tobacco (Tauras et al., 2007), raising the question of whether ENDS are complements or substitutes for alternative tobacco products and marijuana as well. Another factor to consider in understanding the relationship between ENDS and marijuana is that some ENDS devices can be used to vaporize other substances, such as marijuana, opium, and crack cocaine (Young-Wolff et al., 2014; Marynak et al., 2014). Using ENDS to vaporize marijuana can decrease the odor and allows it to be used more discretely (Etter, 2015), which may promote a complement relationship between the two. ENDS minimum legal purchase ages are a recent policy, which if properly enforced, could make it more difficult for adolescents to obtain ENDS. For example, adolescents may now be forced to purchase ENDS devices through an intermediary or online (Williams et al., 2015). Therefore, the implementation of these restrictions can be used to explore if ENDS are substitutes or complements with cigarettes and marijuana because the non-monetary costs of purchasing ENDS products has increased (Williams et al., 2004). New Jersey implemented the first minimum legal purchase age in March, 2010, and as of November 2014, 40 states had age restrictions in place (Marynak et al., 2014). The minimum legal purchasing age among these states is 18, except for four states (Alabama, Alaska, New Jersey, and Utah) in which the age is 19 (Marynak et al., 2014). As of 2014, cigarettes had the same minimum legal purchase age in all states with ENDS age purchasing restrictions (Centers for Disease Control and Prevention, 2014a). The Food and Drug Administration, Center for Tobacco Products (FDA-CTP) has proposed a national ENDS minimum purchasing age of 18, but this has not yet been implemented. In this study, we explore possible substitution effects within a vulnerable adolescent population that is more likely to initiate cigarette use than adults. We use access restrictions from state-level variation in ENDS age purchasing restrictions to explore if ENDS are economic substitutes or complements for cigarettes (e.g. does cigarette consumption increase or decrease as ENDS become more difficult to access). ENDS age purchasing restrictions were enacted in different states at different times, providing policy variation akin to a natural experiment. We explore the effect that these policies have had on cigarette use and marijuana use. We do not explore the effect of these policies on ENDS use because state-level data on ENDS use is not yet available over multiple waves for adolescents. 2. Methods We use state-level aggregated data from the Youth Risk Behavior Surveillance System (YRBSS) (Centers for Disease Control and Prevention, 2014b). YRBSS is a biennial national school-based survey of adolescents conducted by Centers for Disease Control and Prevention (CDC) and state health
departments to monitor health risk behaviors. It utilizes a two-stage, cluster sample design to produce a representative sample of students in grades 9–12 in its jurisdiction, and data are weighted to adjust for school and student nonresponse and to make the data representative. The data are typically collected between February and May of each year (Brener et al., 2013). We extract the data from 2007, 2009, 2011, and 2013 to correspond to a period before and during the implementation of ENDS age purchasing restrictions. ENDS first entered the US market in 2007 (Friedman, 2015). To explore how ENDS age purchasing restrictions are associated with the use of tobacco and marijuana, we use four different measures of cigarette smoking intensity, and measures of cigar use, smokeless tobacco use, and marijuana use. The smoking intensity measures that we use are: 1) recent cigarette use (at least 1 day over the past 30 days), 2) casual cigarette use (between 1 and 19 days over the past 30 days) 3) regular cigarette use (at least 20 days over the past 30 days), 4) and heavy cigarette use (every day over the past 30 days). We also use recent (past 30 days) cigar use, smokeless tobacco use, and marijuana use as outcome measures. We obtained data on ENDS age purchasing restrictions from a report by the CDC (Marynak et al., 2014). We matched this data using the February–May time window over which students were interviewed. Wyoming changed their policy halfway through the data collection in 2013, and we treat this state as if the policy had been in effect the full year. We exclude the state of Massachusetts from our analysis because age purchasing restrictions were implemented at the county level at different points in time (Global Advisors on Smokefree Policy (New Jersey GASP), 2015). In addition to ENDS age purchasing restrictions, we match the following onto the YRBSS data and use this as control variables in our regression analysis: racial/ethnic composition of adolescents, the combined state and federal cigarette excise taxes, and cigarette indoor use laws. We obtained state-level cigarette excise taxes from the State Tobacco Activities Tracking and Evaluation (STATE) System maintained by the CDC (Centers for Disease Control and Prevention, 2014a), and added the federal excise taxes, which were $0.39 in 2007, and $1.01 in 2009, 2011, and 2013. We adjust the tax levels for inflation. We also obtained state-level cigarette indoor use restrictions for restaurants, workplaces, and bars from the STATE system. For racial/ethnic composition of adolescents, we used Survey of Epidemiology and End Results (SEER) population data to construct three state-level ratios: White, other race, and Hispanic 10–19 year olds as a share of all adults (Roth, 2015). We used 19 as the upper bound because of how the data were grouped. For cigarette indoor use restrictions, the STATE system only provides data through 2012, so we impute 2013 data using 2012 data. To explore the association of ENDS age purchasing restrictions on these substance use measures, we estimate a state and year fixed effects regression model. The inclusion of state and year fixed effects allows us to exploit variation in ENDS age purchasing restrictions independent of state-level time-invariant unobservable characteristics, and unobservable characteristics unique to a particular year. For example, anti-smoking sentiment and non-changing population characteristics unique to a particular state from 2007 to 2013 is controlled by the inclusion of a state fixed effects. The national introduction of ENDS over time is controlled for by year fixed effects. We also include separate linear time trends for states enacting ENDS age purchasing restrictions and states not enacting age purchasing restrictions, which control for time-varying omitted variables in these two groups of states, such as changing antismoking sentiment. We perform the regression analysis using 4 annual observations for states with complete information from 2007 to 2013. Cigarette and marijuana use information is provided in all years for 33 states (and 34 for select outcomes), cigar use is provided in all years for 25 states, and smokeless tobacco is provided in all years for 28 states. In Table 1, we show the variation in ENDS age purchasing restrictions by state and year for the largest (34 states) and smallest (25 states) samples. In addition to measuring the association of ENDS age purchasing restrictions on tobacco and marijuana use measures during the period immediately after implementation, we also examine the association on outcomes 2 years before a restriction was in place. If significant associations are found before the policy was implemented, this may indicate that restrictions were endogenously enacted in states with changing cigarette use relative to non-implementing states, or some other omitted variable bias. If significant associations are not found in the prior period, then we argue any associations that we find in the period in which the policy was implemented exhibit evidence of causality. We also examine the association that restrictions have in the 2 years after implementation to observe any lagged effects.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
3
Table 1 Presence of ENDS age purchasing restriction between February and May of a given year. State
AL AKa AZ ARa CT DEa FL GAa HI IDa ILa KSa KYa LAa MEa MDa MIa MSa MTa NHa NMa NY NC NDa OKa RIa SCa SD TNa TXa UTa WVa WIa WY Total
Presence of age purchasing restriction 2007
2009
2011
2013
N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N 0
N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N 0
N N N N N N N N N N N N N N N N N N N Y N N N N N N N N N N Y N N N 2
N Y N N N N N N N Y N Y N N N Y N N N Y N Y N N N N N N Y N Y N Y Y 10
Fig. 1. Recent cigarette use in states with and without an ENDS age purchasing restriction before or during 2013. Note: Estimates are population-adjusted. The vertical line indicates the first ENDS age purchasing restriction in March, 2010 in New Jersey. By 2013, 24 of 50 states had an ENDS age purchasing restriction in place.
3. Results In Table 2, we show that the four adolescent cigarette use measures consistently declined from 2007 to 2013. Recent cigarette use declined from 19.0% to 13.7%, casual cigarette use declined from 10.9% to 9.1%, regular cigarette use declined from 8.0% to 4.5%, and heavy cigarette use declined from 6.1% to 3.4%. While regular and heavy cigarette use were roughly halved between 2007 and 2013, casual cigarette use declined by only 16.4%. In contrast, recent use of marijuana did not decline in 2013 (20.9%) compared to 2005 (18.9%). By 2013, adolescents were 52.3% more likely to recently use marijuana than cigarettes. Inflation-adjusted cigarette excise taxes rose by 59.6%, from $1.61 in 2007 to $2.57 in 2013. Table 2 also shows variation in states implementing ENDS age purchasing restrictions. Six states enacted this policy between March, 2010 (starting in New Jersey) and May, 2011. Fourteen additional states enacted this policy from June, 2012 to May, 2013 (Marynak et al., 2014). We find limited graphical evidence that ENDS may be economic substitutes for cigarettes and economic complements for marijuana. In Fig. 1, we plot population-adjusted recent cigarette use among adolescents for states that enacted an ENDS age purchasing restriction before or during 2013 against those states that did not. The vertical line indicates the first age purchasing restriction in New Jersey. We observe that between 2007 and 2009, recent cigarette use was declining faster in states that would later implement an ENDS age purchasing restriction. This trend reverses from 2009 to 2011, with the recent cigarette use rate remaining virtually identical in states adopting ENDS age purchasing restrictions, but falling in states that did not. From 2011 to 2013, the cigarette rate appears to decline approximately equally in both groups of
Note: We include only states in the 34 state sample for which YRBSS recent cigarette use information is available for all 4 years. a Indicates states also in the 25 state sample for which YRBSS cigar use information is available in all 4 years.
Our method of exploring how increasing the non-monetary costs of obtaining one type of tobacco product (ENDS) influences consumption of another tobacco product (cigarettes) frees us from needing data on both ENDS and cigarettes. We documented in our introduction that these products are often dual-used. However, there is no reason why increasing non-monetary costs of ENDS should increase ENDS use. This frees us to explore a complement/substitute relationship without data on ENDS, although we cannot determine if changes in ENDS consumption occur at the level of wholesale participation (extensive margin), or frequency of use (intensive margin). We performed all analyses in 2015 using Stata 13.1. Standard errors are clustered at the level of state.
Table 2 Descriptive statistics from the Youth Risk Behavior Surveillance Survey.
Recent cigarette use (at least 1 day over the past 30 days) (%) Casual cigarette use (between 1 and 19 days over the past 30 days) (%) Regular cigarette use (at least 20 days over the past 30 days) (%) Heavy cigarette use (every day over the past 30 days) (%) Current cigar use (%) Current smokeless tobacco use (%) Current marijuana use (%) Average real cigarette excise tax (state and federal) ($) Number of states with ENDS age purchasing restrictions by May of each year
States (N)
2007
2009
2011
2013
Change from 2007 to 2013
34 33 33 33 25 28 34 50 50
18.97 10.88 7.95 6.09 14.54 8.16 18.94 1.61 0
17.92 10.97 6.88 4.91 14.81 8.96 19.88 2.52 0
16.56 10.30 6.09 4.57 14.64 8.63 21.12 2.56 6
13.72 9.10 4.51 3.36 13.01 8.86 20.90 2.57 14
−27.68% −16.36% −43.27% −44.83% −10.52% 8.58% 10.35% 59.63% -
Notes: Population-weighted averages are reported. Cigarette excise taxes are in 2013 dollars.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
132 33 0.332 10.208 136 34 0.609 17.177 Observations States Adjusted R-Sq. Mean use rate
Notes: The regression results control for racial/ethnic adolescent composition, cigarette excise taxes, cigarette indoor use laws, state fixed effects, year fixed effects, and linear time trends for adopting and non-adopting states. 95% Confidence intervals indicated in brackets. Significance levels are as noted. ⁎ p b 0.10. ⁎⁎ p b 0.05.
136 34 0.365 20.038 112 28 0.259 9.853 100 25 0.546 13.615 132 33 0.766 5.082
[−0.13, 1.42] 0.76 [−0.67, 2.20] [0.02,1.59] 1.17 [−0.82,3.16] [−0.22, 4.85] 0.77 [−0.71, 2.24] [0.20, 5.95] 1.88⁎ [−0.20, 3.96]
132 33 0.801 6.838
0.15 [−1.64, 1.93] −0.75 [−3.09, 1.59] −1.39 [−3.31, 0.53] −0.23 [−1.80, 1.35] 0.55 [−1.17, 2.26] 0.15 [−1.16, 1.46] −0.12 [−0.77, 0.53] 0.65⁎ 1.96⁎⁎ [0.02, 3.89] 2.31⁎ 1.97⁎ [−0.10, 4.05] 3.08⁎⁎
Previous period ENDS age purchasing restriction Current period ENDS age purchasing restriction Next period ENDS age purchasing restriction
0.16 [−0.50,0.81] 0.80⁎⁎
Recent smokeless tobacco use (at least 1 day over the past 30 days) Heavy cigarette use (every day over the past 30 days) Regular cigarette use (at least 20 days over the past 30 days) Casual cigarette use (between 1 and 19 days over the past 30 days)
We matched policy variation akin to a natural experiment to longitudinal data on tobacco use and marijuana use, and found that cigarette use increased as ENDS became harder to legally purchase due to age purchasing restrictions. However, we failed to detect statistically
Recent cigarette use (at least 1 day over the past 30 days)
4. Discussion
Table 3 Fixed effects regression analysis of cigarette and marijuana use with previous and next period restrictions on ENDS sales to adolescents, 2007–2013.
states. In Fig. 2, we plot recent marijuana use among adolescents. This graphical evidence suggests that marijuana use remained virtually the same after 2009 in states that adopted ENDS age purchasing restrictions, but increased in non-adopting states. In Table 3, we explore if the relationships we observe in the figures are present after controlling for racial/ethnic adolescent composition, cigarette excise taxes, cigarette indoor use laws, state fixed effects, year fixed effects, and linear time trends for adopting and nonadopting states. We find evidence that the ENDS age purchasing restrictions are associated with a 3.1 percentage point (17.9% of the mean) increase in adolescent cigarette use (p b 0.05) in the period of implementation. Most of this effect is accounted for within casual cigarette using adolescents (2.3 percentage point increase, 22.6% of the mean) (p b 0.10). However, for both of these outcomes we observe that cigarette use was increasing in the two-year period prior to the ENDS age purchasing restriction being implemented, which prevents us from making claims of causality due to the presence of potential policy endogeneity. For regular cigarette use (at least 20 days over the past 30 days) and heavy cigarette use (every day over the past 30 days), we do not observe the same increase in the two-year period prior to the ENDS age purchasing restriction being implemented. In the current period, we observe that ENDS age purchasing restrictions are associated with an increase in regular cigarette use of 0.8 percentage points (11.7% of the mean) (p b 0.05), and an increase in heavy cigarette use of 0.7 percentage points (12.8% of the mean) (p b 0.10). While the coefficients remain of similar magnitude in the two-year period after the policy implementation, these associations were not statistically significant. In sum, we find associational evidence of ENDS age purchasing restrictions increasing cigarette use at all levels of intensity, but we only find causal evidence that these restrictions increased cigarette use among heavier cigarette users. These results suggest that ENDS may be substitutes for cigarettes because limiting ENDS purchasing opportunities is associated with increases in cigarette use. In the last three columns on Table 3, we do not find evidence of ENDS age purchasing restrictions being associated with recent cigar use, recent smokeless tobacco use, or recent marijuana use.
Recent cigar use (at least 1 day over the past 30 days)
Fig. 2. Recent marijuana use in states with and without an ENDS age purchasing restriction before or during 2013. Note: Estimates are population-adjusted. The vertical line indicates the first ENDS age purchasing restriction in March, 2010 in New Jersey. By 2013, 24 of 50 states had an ENDS age purchasing restriction in place.
0.25 [−1.11, 1.61] 0.50 [−1.50, 2.49] −1.29 [−3.21, 0.62]
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx Recent marijuana use (at least 1 day over the past 30 days)
4
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
significant increases in cigar use, smokeless tobacco use, or marijuana use. Our results suggest that adolescents are willing to substitute ENDS for cigarettes depending on legal purchasing opportunities of ENDS. We do not suspect that this substitution was complete in the sense that individuals consuming ENDS now only consume cigarettes. It is possible, and likely given statistics on dual-use (Johnston et al., 2014), that adolescents continue to use ENDS after minimum legal purchase ages are implemented, but with less frequency. Our finding of a 0.8 percentage point increase in cigarette use among regular cigarette users in response to ENDS age purchasing restrictions is consistent with a study by Friedman (2015) that found a 0.9 percentage point increase in cigarette use, using data from the National Survey on Drug Use and Health and a difference-in-difference empirical approach similar to our own. Both studies found evidence of substitution effects. One key difference between these studies is that Friedman's data provided all states, whereas our study used at most 34 states due to data availability in the YRBSS. Another difference is that in our study we explored the effect of ENDS age purchasing restrictions for a wider array of outcomes, including cigarette use at different levels of intensity, cigar use, smokeless tobacco use, and marijuana use. There are at least four limitations of our study. First, we are unable to explore the effect of ENDS age purchasing restrictions on ENDS use itself because no state-level panel data on adolescent ENDS use exists currently. While this limitation does not prevent us from exploring a substitution/complement relationship between ENDS and cigarettes, it does prevent us from understanding if ENDS consumption was reduced on the participation (extensive) or frequency of use (intensive) margins. Second, our data are limited by having at most 34 states in our sample. Third, we do not consider local ENDS age purchasing restrictions and the effectiveness of these restrictions. Along these same lines, if adolescents are able to obtain ENDS online or elsewhere with minimal additional difficulty (non-monetary costs) following the implementation of ENDS age purchasing restrictions, this will bias toward the null the measured influence of these restrictions on our outcomes (e.g. underestimate the effect size compared to if this policy were fully enforced). Fourth, this investigation does not explore more nuanced relationships between ENDS availability and cigarette use, such as renormalizing smoking. 5. Conclusion All policymaking bodies should be aware of a potential increase in cigarette use following ENDS age purchasing restrictions. The FDA-CTP is currently evaluating the implementation of a national ENDS age purchasing restriction. In the meantime, the 10 state governments that do not have an age purchasing restriction in place as of January 2015 are free to implement this policy. Given that there are known health risks for ENDS that justify states enacting age purchasing restrictions, these policymaking bodies may wish to pair ENDS age purchasing restrictions with policies that raise the price of cigarettes for adolescents, such as increasing enforcement efforts against venders that sell cigarettes to adolescents. State governments also have the power to concurrently increase cigarette excise taxes to prevent increases in cigarette use in response to ENDS age purchasing restrictions. Another option, in the spirit of regulating ENDS proportionate to their risk relative to combustible tobacco (Chaloupka et al., 2015), may be to establish an ENDS minimum legal purchase age lower than the cigarette minimum legal purchase (Friedman, 2015). Conflict of interest and funding The authors declare that there are no conflicts of interest.
5
Contributor statement MF Pesko performed the analysis and drafted the article. All authors conceptualized the study, interpreted the data, and critically revised the article. Transparency document The Transparency document associated with this article can be found in the online version. Acknowledgments We thank individuals at New York State Pediatric Advocacy Conference for helpful comments. References U.S. Department of Health and Human Services, 2014. The Health Consequences of Smoking—50 Years of Progress: A Report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Agaku, I.T., King, B.A., Dube, S.R., 2014. Current cigarette smoking among adults—United States, 2005–2012. MMWR Morb. Mortal. Wkly Rep. 63, 29–34. Arrazola, R.A., Dube, S.R., King, B.A., 2013. Tobacco product use among middle and high school students—United States, 2011 and 2012. MMWR Morb. Mortal. Wkly Rep. 62, 893–897. Riker, C.A., Lee, K., Darville, A., Hahn, E.J., 2012. E-cigarettes: promise or peril? Nurs. Clin. North Am. 47, 159–171. Wills, T.A., Knight, R., Williams, R.J., Pagano, I., Sargent, J.D., 2015. Risk factors for exclusive e-cigarette use and dual e-cigarette use and tobacco use in adolescents. Pediatrics 135, e43–e51. Johnston, L.D., O'Malley, P.M., Miech, R.A., Bachman, J.G., Schulenberg, J.E., 2014. Monitoring the Future National Survey Results on Drug Use: 1975–2014: Overview, Key Findings on Adolescent Drug Use. Institute for Social Research, The University of Michigan, Ann Arbor, MI. Arrazola, R.A., Singh, T., Corey, C.G., et al., 2015. Tobacco use among middle and high school students—United States, 2011–2014. MMWR Morb. Mortal. Wkly Rep. 64, 381–385 (Apr. 17). Johnston, L.D., O'Malley, P.M., Miech, R.A., Bachman, J.G., Schulenberg, J.E., 2015. Monitoring the Future National Results on Adolescent Drug Use: Overview of Key Findings, 2014. Institute for Social Research, the University of Michigan, Ann Arbor, Michigan. U.S. Department of Health and Human Services, 2012. Preventing Tobacco Use Among Youth and Young Adults: A Report of the Surgeon General. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health. Apelberg, B.J., Corey, C.G., Hoffman, A.C., et al., 2014. Symptoms of tobacco dependence among middle and high school tobacco users results from the 2012 National Youth Tobacco Survey. Am. J. Prev. Med. 47, S4–S14. Hampton, T., 2014. Experts call for research plus regulation of e-cigarettes. JAMA 311, 123–124. Abrams, D.B., 2014. Promise and peril of e-cigarettes: can disruptive technology make cigarettes obsolete? JAMA 311, 135–136. Brandon, T.H., Goniewicz, M.L., Hanna, N.H., et al., 2015. Electronic nicotine delivery systems: a policy statement from the american association for cancer research and the american society of clinical oncology. Clin. Cancer Res. 21, 514–525. McNeill, A., Brose, L.S., Calder, R., Hitchman, S.C., Hajek, P., McRobbie, H., 2015. E-Cigarettes: An Evidence Update. Public Health England (Aug). Pisinger, C., Dossing, M., 2014. A systematic review of health effects of electronic cigarettes. Prev. Med. 69, 248–260. Vardavas, C.I., Anagnostopoulos, N., Kougias, M., Evangelopoulou, V., Connolly, G.N., Behrakis, P.K., 2012. Short-term pulmonary effects of using an electronic cigarette: impact on respiratory flow resistance, impedance, and exhaled nitric oxide. Chest 141, 1400–1406. Flouris, A.D., Poulianiti, K.P., Chorti, M.S., et al., 2012. Acute Effects of Electronic and Tobacco Cigarette Smoking on Complete Blood Count. Food and chemical toxicology: an international journal published for the British Industrial Biological Research Association 50 pp. 3600–3603. Farsalinos, K.E., Tsiapras, D., Kyrzopoulos, S., Savvopoulou, M., Voudris, V., 2014. Acute effects of using an electronic nicotine-delivery device (electronic cigarette) on myocardial function: comparison with the effects of regular cigarettes. BMC Cardiovasc. Disord. 14, 78. Polosa, R., Morjaria, J., Caponnetto, P., et al., 2014. Effect of smoking abstinence and reduction in asthmatic smokers switching to electronic cigarettes: evidence for harm reversal. Int. J. Environ. Res. Public Health 11, 4965–4977. Goniewicz, M.L., Knysak, J., Gawron, M., et al., 2014. Levels of selected carcinogens and toxicants in vapour from electronic cigarettes. Tob. Control. 23, 133–139. Cheng, T., 2014. Chemical evaluation of electronic cigarettes. Tob. Control. 23 Suppl 2, ii11–ii17.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001
6
M.F. Pesko et al. / Preventive Medicine xxx (2016) xxx–xxx
Orr, M.S., 2014. Electronic cigarettes in the USA: a summary of available toxicology data and suggestions for the future. Tob. Control. 23 Suppl 2, ii18–ii22. Etter, J.F., 2015. Electronic cigarettes and cannabis: an exploratory study. Eur. Addict. Res. 21, 124–130. Huang, J., Tauras, J., Chaloupka, F.J., 2014. The impact of price and tobacco control policies on the demand for electronic nicotine delivery systems. Tob. Control. 23 Suppl 3, iii41–iii47. Grace, R.C., Kivell, B.M., Laugesen, M., 2015. Estimating cross-price elasticity of ecigarettes using a simulated demand procedure. Nicotine Tob. Res. (in press:ntu268). Chaloupka, F.J., Pacula, R.L., Farrelly, M.C., Johnston, L.D., O'Malley, P.M., 1999. Do higher cigarette prices encourage youth to use marijuana? NBER Working Paper No 6939. National Bureau of Economic Research, Cambridge, MA Farrelly, M.C., Bray, J.W., Zarkin, G.A., Wendling, B.W., 2001. The joint demand for cigarettes and marijuana: evidence from the National Household Surveys on Drug Abuse. J. Health Econ. 20, 51–68. Tauras, J., Powell, L., Chaloupka, F., Ross, H., 2007. The demand for smokeless tobacco among male high school students in the United States: the impact of taxes, prices and policies. Appl. Econ. 39, 31–41. Young-Wolff, K.C., Karan, L.D., Prochaska, J.J., 2014. Electronic cigarettes in jails: a panacea or public health problem? JAMA Psychiatry. Marynak, K., Holmes, C.B., King, B.A., Promoff, G., Bunnell, R., McAfee, T., 2014. State laws prohibiting sales to minors and indoor use of electronic nicotine delivery systems—United States. MMWR Morb. Mortal. Wkly Rep. 63, 1145–1150. Williams, R.S., Derrick, J., Ribisl, K.M., 2015. Electronic cigarette sales to minors via the Internet. JAMA Pediatrics 169, e1563.
Williams, J., Pacula, R.L., Chaloupka, F.J., Wechsler, H., 2004. Alcohol and marijuana use among college students: economic complements or substitutes? Health Econ. 13, 825–843. Centers for Disease Control and Prevention, 2014a. State Tobacco Activities Tracking and Evaluation (STATE) System. (Accessed Jun 12 2014, at http://apps.nccd.cdc.gov/ statesystem/Default/Default.aspx. Centers for Disease Control and Prevention, 2014b. Youth Risk Behavior Surveillance System (YRBSS). (Accessed Jun 12 2014, at http://www.cdc.gov/HealthyYouth/yrbs/. Brener, N.D., Kann, L., Shanklin, S., et al., 2013. Methodology of the Youth Risk Behavior Surveillance System—2013. MMWR Recommendations and Reports: Morbidity and Mortality Weekly Report Recommendations and Reports/Centers for Disease Control. 62, pp. 1–20. Friedman, A.S., 2015. How does electronic cigarette access affect adolescent smoking? J. Health Econ. 44, 300–308. Global Advisors on Smokefree Policy (New Jersey GASP), 2015. Electronic smoking devices. (Accessed Dec 21 2015, at http://www.njgasp.org/wp-content/uploads/ 2014/05/E-Cigs_White_Paper.pdf. Roth, J., 2015. Survey of Epidemiology and End Results (SEER) U.S. state and county population data. Natl. Bur. Econ. Res. Accessed Jan 30 2015 at http://www.nber.org/ data/seer_u.s._county_population_data.html. Chaloupka, F.J., Sweanor, D., Warner, K.E., 2015. Differential taxes for differential risks— toward reduced harm from nicotine-yielding products. N. Engl. J. Med. 373, 594–597.
Please cite this article as: Pesko, M.F., et al., The influence of electronic cigarette age purchasing restrictions on adolescent tobacco and marijuana use, Prev. Med. (2016), http://dx.doi.org/10.1016/j.ypmed.2016.02.001