Are They Smoking Premium, Discount Or Native Brands? S T Leatherdale,1,2,3 R Ahmed,4 a Barisic,1 D Murnaghan,5,6 S Manske2,7

Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from Cigarette brand preference as a function of price among smoking youths in Canada: are they smoking premium, discount or native brands? S T Leatherdale,1,2,3 R Ahmed,4 A Barisic,1 D Murnaghan,5,6 S Manske2,7

1 Department of Population ABSTRACT Historically, virtually all cigarette brands in Studies and Surveillance, Cancer Introduction: Given that little is known about the price- Canada were sold at the same price (although Care Ontario, Canada; 2 Department of Health Studies related cigarette brand preferences of youths, the current there were some minor variations across provinces and Gerontology, University of study seeks to characterise cigarette brand preferences owing to provincial difference in tobacco taxes). Waterloo, Canada; 3 Dalla Lana and examine factors associated with smoking discount or More recently, tobacco manufactures have School of Public Health, native cigarette brands among Canadian youths who are launched a variety of ‘‘discount’’ brands, that sell University of Toronto, Canada; current smokers. for approximately $C10 to $C20 less per carton 4 Population Health Research Group, University of Waterloo, Methods: This study used nationally representative data (200 cigarettes) compared to premium brands Canada; 5 School of Nursing, collected from 71 003 grade 5–12 students as part of the despite being taxed at the same rate as premium University Prince Edward Island, cigarette brands.56 For instance, according to an 6 2006–7 Canadian Youth Smoking Survey (YSS). Using Canada; PEI Health Research data from current smokers, logistic regression models Imperial Tobacco Canada report, their average Institute, University of Prince Edward Island, Canada; 7 Centre were used to examine factors associated with smoking retail price for a carton of 200 cigarettes in 2006 for Behavioural Research and discount or native cigarette brands relative to premium was $C76.58 (£43; J50) (for premium brands and Evaluation, Canadian Cancer cigarette brands. $C55.97 for discount brands.5 Considering the Society/National Cancer Institute Results: In 2006, premium cigarettes were the most higher cost of purchasing premium cigarette of Canada, Canada prevalent brand of cigarette youths report usually smoking brands, it is not surprising that Canadian smokers (49.4%); a substantial number of youths do report usually are turning to cheaper cigarette options. The

Correspondence to: Protected by copyright. Dr Scott T Leatherdale, smoking either discount (12.9%) or native (9.3%) market share for discount cigarette brands has Department of Population cigarette brands. Occasional smokers were more likely to increased from 10% in 2003 to 40% in 2005.7 Studies and Surveillance, Cancer Care Ontario, 620 University report usually smoking premium cigarettes whereas daily According to the 2005 Canadian Tobacco Use Avenue, Toronto, ON, Canada smokers were more likely to report smoking either Monitoring Survey (CTUMS), 36% of current M5G 2L7; scott.leatherdale@ discount or native cigarettes. In particular, discount and smokers aged 15 and older reported purchasing a cancercare.on.ca native brands appear to be appealing among smoking discount cigarette brand in the previous six http://tobaccocontrol.bmj.com/ months.1 This is consistent with historical data Received 26 January 2009 youths with less spending money or those who are Accepted 10 August 2009 heavier smokers compared to youths smoking premium from the United States, which demonstrated that brands. the prevalence of use of discount cigarettes among Conclusion: Discount and native cigarette brands are a sample of adults increased from 6.2% in 1988 to commonly used by a substantial number of smoking 23.4% in 1993 as the price of cigarettes increased youths in Canada. Additional research is required to better over that period of time, especially among lower understand the reasons behind different cigarette brand income smokers and heavier smokers.8 Research preferences and how youths are able to access premium, has yet to examine if youth populations are using discount and illicit native cigarettes. Moreover, ongoing discount cigarette brands. surveillance of the cigarette brand preferences of youths Another affordable cigarette brand option avail- is required for guiding future tobacco control policy and able to Canadian smokers is contraband tobacco. on 20 August 2019 at University of Cape Town Libraries. programming activities. According to the Royal Canadian Mounted Police (RCMP), there are currently four major types of contraband tobacco available in the Canadian National surveillance data suggest that there has market: (1) unlawfully/lawfully manufactured been a 50% reduction in the prevalence of current Canadian products or US products smuggled into smoking among Canadian youths over the last Canada; (2) diverted tax-exempt tobacco products; decade.1 However, considering the link between (3) counterfeit tobacco products; and (4) tobacco the price of cigarettes and youths smoking products from other criminal activities (that is, prevalence (that is, higher tobacco prices result in cargo thefts, store thefts, etc).9 At the present time, lower cigarette consumption rates),2–4 the recent two of the larger sources of contraband tobacco in emergence of tobacco manufactures providing Canada are from native manufactures that either cheaper cigarette options for smokers is cause for unlawfully/lawfully manufacture products in concern. If price sensitive youth smokers are able Canada or manufacture products in the US and to access the more affordable brands of cigarettes smuggle them into Canada (these sources are that tobacco manufactures are providing smokers, primarily located in Ontario and Quebec), and there is the possibility that youth smoking rates from tax-exempt cigarettes designated for may plateau or even potentially rise in the future. Aboriginals to be purchased on First Nations Using nationally representative data, we examine Reserves that are illegally diverted to the general if current youth smokers in Canada are smoking population.9 These types of native cigarettes these more affordable cigarettes. products represent an affordable option for youth

466 Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from smokers that is substantially cheaper than both premium and ‘‘Have you ever smoked at least 100 cigarettes in your life?’’ discount brands. (These products are referred to as native (yes, no), and ‘‘On how many of the last 30 days did you smoke brands for the remainder of this paper.) one or more cigarettes’’ (none, 1 day, 2–3 days, 4–5 days, 6– For instance, native cigarettes can be purchased for approxi- 10 days, 11–20 days, 21–29 days, 30 days). Consistent with the mately $C8 to $C10 for a bag of 200 cigarettes (equivalent of a derived variables for classifying smoking using YSS data,13 14 carton) according to both research studies5 and personal current smokers were defined as those who have smoked at testimonials.10 At the present time, the majority of these least 100 cigarettes in his/her lifetime and have smoked in the products are consumed in Ontario and Quebec; however, there 30 days preceding the survey. Daily smokers were defined as is evidence that the proliferation of these products is increasing current smokers who reported smoking at least one cigarette per in parts of Atlantic Canada.5 10 Considering the extremely low day for each of the 30 days preceding the survey. Occasional price, and the apparent ease of non-Aboriginal smokers smokers were defined as current smokers who reported smoking accessing these products,59 native brands of cigarettes may be at least one cigarette during the 30 days preceding the survey attractive products for price sensitive smoking youths. For but have not smoked every day. Respondents were asked to example, research among a sample of adults in the US identified indicate what brand of cigarette they usually smoke from a list that smokers who live within 40 miles (,64 km) of an Indian of popular Canadian cigarette brands consumed by youths. reservation were more likely to use a price avoidance strategy Using the response options provided and the list of discount (for example, buying untaxed cigarettes) when buying cigar- brands provided by the Non-Smokers Rights Association,7 we ettes.11 A recent study identified that over 20% of the cigarettes grouped the respondents’ usual brand of cigarette into one of consumed by a non-random sample (n = 300) of youth smokers five categories based on the brand price (premium, discount, in Toronto (Ontario) were native brands.12 native, other, no usual brand), where ‘‘premium’’ represents any Given that little is known about the price-related cigarette brand of cigarettes sold for the traditional price per carton brand preferences among Canadian youths, the current study (DuMaurier, Players, Export A), ‘‘discount’’ represents any seeks to characterise price-related cigarette brand preferences brand of cigarettes sold for $C10–$C12 less per carton than a and examine factors associated with smoking discount or native premium brand (No 7, Peter Jackson), ‘‘native’’ represents tax cigarette brands among Canadian youths who are current exempt cigarettes sold to non-Aboriginals and cigarettes smokers. illegally/illegally manufactured and sold as native brands, and

‘‘other’’ represented that their usual brand was not in the list Protected by copyright. provided (that is, could include other brands of Canadian METHODOLOGY cigarettes, American brands of cigarettes, or roll-your-own Design (RYO) tobacco products). Respondents were also asked to This study used nationally representative data collected from report ‘‘When you smoke, how often do you share a cigarette 71 003 grade 5–12 students as part of the 2006–7 Canadian with others?’’ (never, sometimes, usually, always), and 13 14 Youth Smoking Survey (YSS). In brief, the target population ‘‘Thinking back over the last 30 days, on the days that you http://tobaccocontrol.bmj.com/ for the YSS consisted of all young Canadian residents in grades smoked, how many cigarettes did you usually smoke each day?’’ 5–12 attending public and private elementary and secondary (a few puffs to 1, 2–3, 4–5, 6–10, 11–20, 21–29, 30 or more). schools in 10 Canadian provinces; youths residing in the Yukon, Based on the response distribution the response categories for Nunavut and the Northwest Territories were excluded from the average cigarettes per day were collapsed (few puffs to 1, 2–3, 4– target population, as were youths living in institutions or on 10, 11 or more). Respondents were also asked to report their sex First Nation Reserves, and youths attending special schools or (boy, girl), grade (5–12), and ‘‘Are you an Aboriginal person?’’ schools on military bases. The sample design consisted of a two- (yes First Nations, yes Me´tis, yes Inuit, no I am not an stage stratified clustered design with schools as primary Aboriginal person). Responses were collapsed (Aboriginal, non- sampling units and classes as secondary sampling units. All of aboriginal). Respondents were also asked ‘‘How much money the students in the selected classes were surveyed. The sample do you usually get each week to spend on yourself or to save?’’ design featured three levels of stratification: province, health ($C0, $C1–5, $C6–10, $C11–20, $C21–40, $C41–100, more than on 20 August 2019 at University of Cape Town Libraries. region and school level (elementary vs secondary). The sample $C100), Based on the response distribution, categories for of schools was selected systematically with probability propor- spending money were collapsed ($C0, $C1–10, $C11 or more). tional to school size (73.7% school response rate). The selection Respondents who reported that they had a usual brand of of the secondary sampling units (classes) was conducted by field cigarette were to asked a series of questions related to the staff who randomly selected one class in the desired grade question ‘‘Why do you smoke the brand of cigarettes that you (61.1% student response rate). (Owing to the healthy New do?’’ Respondents were asked to report ‘‘yes’’ for each Brunswick en sante´ initiative, YSS data were collected from an appropriate reason provided (my friends smoke the same brand, additional 171 schools beyond the original sample target. This is my parents smoke the same brand, I like the packaging, this described in more detail in references 13 and 14.) brand costs less than other brands, I like the image, I like the The survey design and sample weights allow us to produce taste, they are the only ones I can get, they have less tar, for the population-based estimates within this paper. Detailed informa- nicotine buzz). tion on the sample design, methods and survey rates for the 13 14 2006–7 YSS are available in print or online (https://www. Analyses yss.uwaterloo.ca/ysssite_app/controller/index.cfm). Using data from current smokers, descriptive analyses examining smoking status and cigarette brand preference were Measures performed according to sex, grade, their reasons for smoking a The YSS collected demographic information and data on youth particular brand, weekly spending money, frequency of sharing smoking behaviour. Smoking status was defined based on how cigarettes, average cigarettes per day, Aboriginal status and the respondents answered, ‘‘Have you ever smoked a whole province. We then conducted two logistic regression models to cigarette?’’ (yes, no). Those that answered ‘‘yes’’, were asked, examine characteristics that differentiate current smokers with

Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 467 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from

Table 1 Logistic regression analyses examining factors associated with cigarette brand preference among youths who are current smokers{ in grades 5–12, 2006–2007, Canada Adjusted odds ratio{ (95% CI) Model 1 Model 2 Discount cigarettes vs Premium Native cigarettes vs Premium Parameters cigarettes cigarettes

Sex Girl 1.00 1.00 Boy 0.97 (0.73 to 1.29) 1.72 (1.24 to 2.40)** Smoking status Occasional smoker 1.00 1.00 Daily smoker 1.55 (1.14 to 2.10)** 1.45 (1.02 to 2.06)* Aboriginal status Non-aboriginal 1.00 1.00 Aboriginal 0.28 (0.19 to 0.43)*** 0.56 (0.37 to 0.84)** Weekly spending money $C0 1.00 1.00 $C1 to $C10 0.39 (0.19 to 0.79)** 0.72 (0.39 to 1.32) $C11 or more 0.84 (0.48 to 1.48) 0.34 (0.20 to 0.56)*** Frequency of sharing cigarettes Never 1.00 1.00 Sometimes 1.63 (0.80 to 3.30) 1.33 (0.67 to 2.67) Usually 1.82 (0.88 to 3.75) 1.70 (0.84 to 3.45) Always 1.68 (0.75 to 3.76) 0.74 (0.31 to 1.77) Average cigarettes per day Few puffs to 1 1.00 1.00

2–3 2.11 (1.01 to 4.40)* 0.91 (0.46 to 1.80) Protected by copyright. 4–10 4.02 (2.00 to 8.10)*** 1.49 (0.79 to 2.82) 11 or more 2.87 (1.34 to 6.12)* 2.15 (1.06 to 4.33)* *p,0.05 **p,0.01 ***p,0.001. {Includes both occasional and daily smokers. {Odds ratios adjusted for all other variables in the table and controlling for region and grade. Model 1: 1 = discount cigarettes (n = 569), 0 = premium cigarettes (n = 1240). http://tobaccocontrol.bmj.com/ Model 2: 1 = native cigarettes (n = 298), 0 = premium cigarettes (n = 1240). a usual brand of cigarettes who (a) smoke discount cigarettes image of the brand (3.2%). Daily smokers were more likely to versus premium cigarettes, and (b) smoke native cigarettes report that they smoke a particular brand because they cost less versus premium cigarettes. Survey weights were used to adjust (x2 = 55.7, df = 1, p,0.001), whereas occasional smokers were for non-response between provinces and groups, thereby more likely to report that they smoke a particular brand because minimising any bias in the analyses caused by differential their friends smoke the same brand (x2 = 5.9, df = 1, p,0.05). response rates across regions or groups (table 1). The statistical Additional analysis revealed that among youths who report that package SAS 8.02 was used for all analyses.15 their reason for smoking a particular brand is because they cost less, 24.8% smoke discount cigarettes and 35.9% smoke native

RESULTS cigarettes, whereas 27.7% smoke ‘‘other’’ brands and 11.6% on 20 August 2019 at University of Cape Town Libraries. In 2006, 5.8% (188 200) of Canadian youths in grades 5–12 were smoke premium cigarettes. current smokers. The prevalence of current smoking was higher for males (6.5%) compared to females (4.9%) (x2 = 97.14, df = 1, Cigarette brand preference p,0.001). Among current smokers, the prevalence of occasional In 2006, premium cigarettes were the most prevalent brand of and daily smoking was roughly the same (50.5% occasional vs cigarette youths report usually smoking (49.4%), however, a 49.5% daily), although students in higher grades were more substantial number of youths report usually smoking either likely to be daily smokers than students in lower grades discount (12.9%, 20,700) or native (9.3%, 15 000) cigarette 2 (x = 23.3, df = 4, p,.001). The prevalence of daily smoking was brands. Interestingly, occasional smokers were more likely to highest in Atlantic Canada (61.8%) and lowest in British report usually smoking premium cigarettes whereas daily Columbia (46.6%). Descriptive statistics among Canadian smokers were more likely to report smoking either discount or youths who are current smokers in grades 5–12 by smoking native cigarettes (x2 = 170.1, df = 4, p,0.001). As shown in status are presented in table 2. figure 1, when examining the distribution of cigarette brand preference by weekly spending money, smokers with $C10 of Reasons for smoking a particular brand spending money or less per week appear to be more apt to The most prevalent reasons for smoking a particular brand of report smoking native cigarettes, whereas smokers with more cigarette reported by current smoking youths in 2006 were that than $C10 per week in spending money are more apt to report they like the taste (55.6%), they cost less (17.0%) and their usually smoking premium cigarettes. As shown in table 3, when friends smoke the same brand (15.3%); the least prevalent examining the distribution of cigarette brand preference by reasons reported were that they have less tar (2.8%), they are average daily cigarette consumption, heavier smokers appear to the only ones they can get (3.1%) and because they like the be more apt to report smoking native or discount cigarettes

468 Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from

Table 2 Descriptive statistics for the sample of youths who are current smokers* in grades 5–12 by smoking status, 2006–2007, Canada Occasional smoker Daily smoker (n = 2,399) (n = 3,129) Student characteristics %{ (n){ %{ (n){ x2{ Sex Boy 56.7 (1255) 59.3 (1654) x2 = 2.82, df = 1, p = 0.093 Girl 43.3 (1135) 40.7 (1440) Grade (Elementary) 5–8 12.6 (296) 8.9 (280) x2 = 21.48, df = 4, p,0.001 9 14.5 (368) 16.2 (431) 10 19.0 (533) 21.4 (721) 11 27.2 (648) 24.6 (880) 12 26.7 (554) 28.9 (817) Region of Canada Atlantic 6.9 (1136) 11.4 (1839) x2 = 25.88, df = 4, p,0.001 Quebec 21.4 (492) 21.1 (512) Ontario 36.2 (343) 34.7 (386) Prairies 18.8 (268) 18.1 (240) British Columbia 16.5 (160) 14.7 (152) Type of cigarette usually smoked based on the brand price No usual brand 13.6 (298) 4.6 (158) x2 = 170.14, df = 4, p,0.001 Premium 51.2 (925) 47.4 (1134) Discount 9.3 (323) 16.5 (563) Native 5.6 (139) 13.1 (320) Other 20.3 (363) 18.4 (458) Reasons for smoking a particular brand of Protected by copyright. cigarette" Those who responded ‘‘yes’’ Friends smoke the same brand 16.9 (312) 13.8 (266) Parents smoke the same brand 6.2 (154) 9.9 (268) Likes the packaging 4.6 (76) 7.6 (154) Costs less 11.8 (291) 21.8 (646) http://tobaccocontrol.bmj.com/ Likes the image 1.8 (46) 4.6 (94) Likes the taste 54.8 (882) 56.3 (1308) Only ones they can get 2.8 (77) 3.3 (102) Have less tar 2.8 (50) 2.8 (60) For the nicotine buzz 9.2 (126) 12.4 (253) *Includes both occasional and daily smokers. {Based on weighted estimates. {Unweighted sample size; numbers may not add to total because of missing values. "Only among current smokers who report having a usual brand of cigarettes: occasional smokers (n = 1750), daily smokers (n = 2475). on 20 August 2019 at University of Cape Town Libraries. (x2 = 218.8, df = 12, p,0.001). As illustrated in figure 2, the who only smoke a few puffs or one cigarette per day, youths majority of youths in the Prairies, British Columbia and Ontario were more likely to smoke discount cigarettes if they reported report smoking premium cigarettes. However, the prevalence of smoking 2–3, 4–10 or 11 or more cigarettes per day. Conversely, youths smoking discount cigarettes was highest among youths Aboriginal youths were less likely to smoke discount cigarettes in the Prairies and Atlantic Canada, and the prevalence of than non-Aboriginal youths, as were youths with $C1 to $C10 a youths smoking native cigarettes was highest in Ontario and week in spending money relative to youths with no weekly Quebec; the prevalence of youths smoking native cigarettes was spending money. the lowest in the Prairies and British Columbia. Interestingly, among youths from Quebec, the most prevalent response for cigarette brand preference was ’’other’’. When examining Factors associated with smoking native cigarettes vs premium cigarette brand preference by Aboriginal status (fig 3), it was cigarettes identified that 8.7% (11 000) of non-Aboriginal youth smokers Male youths were more likely than female youths to smoke report that their usual brand of cigarette is a native brand; native cigarettes. Daily smokers were also more likely than 11.6% of Aboriginal youths report usually smoking a native occasional smokers to report smoking native cigarettes, as were brand. youths who smoke 11 or more cigarettes per day compared to youths who smoke a few puffs or one cigarette per day. Factors associated with smoking discount cigarettes vs premium Conversely, Aboriginal youths were less likely to smoke native cigarettes cigarettes than non-aboriginal youths, as were youths with Daily smokers were more likely than occasional smokers to more than $C11 a week in spending money relative to youths report smoking discount cigarettes. When compared to youths with no weekly spending money.

Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 469 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from

Figure 1 Prevalence of different cigarette brand preferences among current smokers* by weekly spending money. Canada, 2006–2007. Source: 2006–2007 Canadian Youth Smoking Survey. *Current smokers include both current occasional smokers and current smokers.

DISCUSSION important for guiding future tobacco control policy, program- The data presented here suggest that discount and native ming and surveillance activities. cigarette brands are commonly used by a substantial number of The current findings highlight an important limitation in smoking youths in Canada, although premium cigarettes common tobacco control strategies among youths populations. Protected by copyright. continue to be the brand of choice for almost half of the youth That is, the consumption of potentially contraband native smokers. As expected, discount brands appear to be appealing tobacco products may not be a negligible problem among among smoking youths with less spending money or those who Canadian youths. In 2006, almost one in 10 Canadian youths are heavier smokers, and among smoking youths in the Prairies who were current smokers reported that they usually smoked a

and Atlantic Canada where other discounted cigarette options native brand of cigarette. Our finding that heavier smokers were http://tobaccocontrol.bmj.com/ (that is, native brands) are not widely available.5 Consistent more apt to smoke cheaper native brands is consistent with with research findings for adult populations,5 we also identified existing evidence for young adult populations (aged 19–29).5 the prevalence of youth smokers reporting that they usually Among the youths who reported that they usually smoke native smoke native cigarettes was highest in Ontario and Quebec, and brand cigarettes, over a third reported that one of the reasons modest but non-negligible in Atlantic Canada. As expected, was because they cost less. Considering that youths don’t have smoking youths who had less spending money and were heavier problems accessing cigarettes,16 17 and evidence suggests that smokers were also more likely to report that they usually smoke native brands may be even easier to access than premium and native cigarette brands. Such timely and relevant data are discount cigarettes in many jurisdictions,5 10 the combination of

Figure 2 Prevalence of different cigarette brand preferences among on 20 August 2019 at University of Cape Town Libraries. current smokers* by region of Canada. Canada, 2006–2007. Source: 2006–2007 Canadian Youth Smoking Survey. *Current smokers include both current occasional smokers and current smokers.

470 Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from

Figure 3 Prevalence of different cigarette brand preferences among current smokers* by Aboriginal status. Canada, 2006–2007. Source: 2006–2007 Canadian Youth Smoking Survey. *Current smokers include both current occasional smokers and current smokers.

‘‘easy-to-obtain and cheap’’ is of particular cause for concern in of priorities for action that include supply prohibitions for relation to youth tobacco control. Youth populations are unlicensed manufacturers and sanctions for licensed manufac- sensitive to the price of tobacco,2–4 so it is clear that policies tures who act unlawfully.18 Interestingly, the emergence and

designed eliminate the availability of native tobacco products, growth of native cigarette brands has hit the tipping point where Protected by copyright. or at the very least to increase the price of native tobacco even the large tobacco manufacturers are developing strategies to products for all youths (both Aboriginal and non-Aboriginal) take action on reducing the market share of illegal and legal native should be a critical component of future tobacco control brands (for example, Imperial Tobacco Canada5). Future research programming; as long as affordable and accessible native brands is required to evaluate the impact that future programmes and of cigarettes remain available to youths, existing tax policies policies have on the illicit consumption of contraband tobacco will probably be less effective. products being produced and/or distributed by native manufac- http://tobaccocontrol.bmj.com/ In order to address the current problem of native cigarette tures within the Canadian context. brands being illegally consumed by non-Aboriginal smokers, the Overall, these findings illustrate that a substantial number of Canadian Coalition for Action on Tobacco has suggests a variety youth smokers in Canada smoke discount brands of cigarettes

Table 3 Descriptive statistics for the sample of youths who are current smokers* in grades 5–12 by cigarette brand price, 2006–2007, Canada No usual brand Premium Discount Native (n = 456) (n = 2,059) (n = 886) (n = 459) Other (n = 821) %{ (n){ %{ (n){ %{ (n){ %{ (n){ %{ (n){ x2{ on 20 August 2019 at University of Cape Town Libraries. Sex Boy 64.9 (239) 58.9 (1128) 54.9 (405) 67.4 (250) 51.6 (430) x2 = 31.43, df = 4, p,0.001 Girl 35.1 (213) 41.1 (922) 45.1 (478) 32.6 (203) 48.4 (385) Grade (elementary) 5–8 18.7 (99) 8.0 (131) 11.8 (75) 11.4 (61) 7.8 (80) x2 = 170.93, df = 16, p,0.001 9 21.6 (79) 11.0 (250) 14.9 (113) 27.8 (93) 18.4 (122) 10 22.7 (101) 19.5 (439) 18.0 (216) 20.7 (112) 19.3 (176) 11 25.5 (103) 28.6 (617) 24.1 (257) 19.3 (98) 28.2 (253) 12 11.5 (74) 32.9 (622) 31.2 (225) 20.8 (95) 26.3 (190) Frequency of sharing cigarettes Never 3.8 (21) 5.7 (117) 5.7 (39) 9.9 (38) 5.9 (51) x2 = 52.52, df = 12, p,0.001 Sometimes 31.8 (162) 46.3 (978) 47.8 (421) 43.2 (208) 42.0 (387) Usually 55.2 (208) 37.0 (705) 35.1 (319) 36.4 (152) 41.3 (294) Always 9.2 (57) 11.0 (237) 11.4 (97) 10.5 (53) 10.8 (81) Average cigarettes per day Few puffs to 1 24.1 (96) 9.1 (169) 3.2 (40) 5.1 (22) 11.3 (66) x2 = 218.80, df = 12, p,0.001 2–3 30.0 (108) 25.1 (427) 17.7 (153) 12.0 (55) 27.7 (163) 4–10 24.7 (138) 43.1 (890) 55.0 (425) 46.3 (179) 45.9 (381) 11 or more 21.2 (105) 22.7 (542) 24.1 (258) 36.6 (196) 15.1 (195) *Includes both occasional and daily smokers. {Based on weighted estimates. {Unweighted sample size; numbers may not add to total because of missing values.

Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 471 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from on a regular basis. Among the youths who reported that they usually smoke discount cigarettes, over a quarter of them What this paper adds reported that one of the reasons was because they cost less. This is consistent with our finding that smokers with less disposable c Considering the link between the price of cigarettes and youth income and heavier smokers were more likely to smoke discount smoking prevalence, the recent emergence of tobacco cigarettes. Considering the link between youth smoking and the manufactures providing cheaper cigarette options for smokers price of tobacco,2–4 it appears that discount brands are a viable is cause for concern. Tobacco manufactures offer a variety of and attractive alternative to more expensive premium brand ‘‘discount’’ brands that sell for approximately $C10 to $C20 cigarettes for many smoking youths. Interestingly, we identified less per carton (200 cigarettes) compared to premium brands that the prevalence of smokers using discount brands was the despite being taxed at the same rate as premium cigarette highest in the Prairies and Atlantic Canada; jurisdictions where brands. There are also substantially cheaper contraband native the cheaper native brands are not yet widely available.5 Future cigarettes available that sell for approximately $C8 to $C10 for research should monitor the popularity of discount cigarette a bag of 200 cigarettes (equivalent to a carton). Research has brands among Canadian youths and across different provinces, previously identified that price sensitive adult smokers switch especially as other cheaper cigarette options become more to these cheaper cigarette options, but their use among youth widely available. Such ongoing research is critical for future smoking populations is largely unknown. If price sensitive tobacco control prevention programming considering that youth smokers are able to access the more affordable cigarette manufacturers are likely to continue providing cigarettes, there is the possibility that youth smoking rates Canadian smokers with these discounted product options may plateau or even potentially rise in the future. owing to their success in the marketplace (for example, c The data presented here suggest that discount and native according to Rothmans’ 2006 Annual Report, discount brands cigarette brands are commonly used by a substantial number accounted for over 55% of their total sales volume in 2006).19 of smoking youths in Canada, although premium cigarettes Consistent with existing research,20 few youth smokers report continue to be the brand of choice for almost half of the youth not having a usual brand of cigarette. Similarly, we also found smokers. These cheaper cigarette options appear to be that the proportion of smokers reporting no usual brand of appealing among smoking youths with less spending money or cigarette declined among older smokers. However, we did those who are heavier smokers. Discount brands appear to be more appealing to smoking youths in the Prairies and Atlantic identify that a substantial proportion of youths reported that Protected by copyright. they usually smoke some ’’other’’ brand of cigarette which was Canada where native cigarette brands are not widely available, not listed in the 2006 YSS questionnaire. There are a wide whereas native cigarette brands are consumed by variety of potential ‘‘other’’ options, such as other Canadian substantially more smoking youths in Ontario and Quebec discount cigarette brands (for example, Canadian Classics or since native brands are available in those jurisdictions. Matine´e), roll-you-own tobacco or American cigarette brands Ongoing surveillance of the cigarette brand preferences of (for example, Marlboro or Camel) which would be valuable to youths is required for guiding future tobacco control policy and http://tobaccocontrol.bmj.com/ monitor. Moreover, additional systematic surveillance of the programming activities. different types of native cigarettes being consumed by youths (for example, baggies of loose cigarettes or specific native brands such as DKs, Sago and Putters) and where and how non-native ‘‘other’’. Data were also unavailable to determine if youths were youths are accessing these illicit products is required. Such data consuming other forms of contraband tobacco, such as counter- are important as they can serve as a baseline to compare with feit tobacco products and tobacco products from other criminal future estimates of youth cigarette brand preference which may activities. It should also be noted that the cross-sectional nature be impacted by emerging programmes and policies. of the design does not allow for causal inferences regarding the This study has several limitations common to survey association between brand preference and the correlates research. Although the response rate was high and the data

examined in this manuscript. Longitudinal data are required on 20 August 2019 at University of Cape Town Libraries. were weighted to help account for non-response, the findings to determine the temporal relations among correlates. are nevertheless subject to sample bias. In addition, the findings 21 likely reflect some under-reporting for tobacco use, despite Acknowledgements: The authors would like to thank Health Canada and the efforts to ensure confidentiality and truthful reporting. The SHAPES team at the University of Waterloo for providing support for this project, and prevalence rates for youth smoking discount and native brands Mr Michael Perley of the Ontario Campaign for Action on Tobacco for reviewing an of cigarettes in this paper are also very conservative considering earlier draft of this manuscript. Dr Leatherdale is a Cancer Care Ontario Research Chair in Population Studies funded by the Ontario Ministry of Health and Long-term Care. The we were only able to examine the brand that smokers reported Canadian Cancer Society provided funding to develop SHAPES, the system used to they usually smoke rather than the brand(s) that they may also collect the YSS data. periodically smoke. This is important as research has previously Competing interests: None. identified that many smokers consume a wide variety of Provenance and peer review: Not commissioned; externally peer reviewed. tobacco products on a periodic basis.22 Owing to limitations with the measures available, we also were unable to determine if the native brands reported by youth smokers were illegal or REFERENCES legal cigarettes manufactured on a First Nations Reserve and/or 1. Health Canada. 2004–05 Youth Smoking Survey, Ottawa: Minister of Supply and other brands of cigarettes purchased tax-free on a First Nations Services Canada, 2007. Available at: http://www.hc-sc.gc.ca/hl-vs/tobac-tabac/ research-recherche/stat/survey-sondage/2004-2005/index_e.html (accessed 12 Nov Reserve. In addition, the limited number of discount brand 2008). options provided to respondents and the high prevalence of 2. Ross H, Powell LM, Tauras JA, et al. New evidence on youth smoking behavior youths reporting ‘‘other’’ for their usual brand of cigarette based on experimental price increases. Contemp Economic Policy 2005;23:195–211. suggests that prevalence rates for discount brands may be 3. Liang L, Chaloupka FJ. Differential effects of cigarette price on youth smoking intensity. Nicotine Tob Res 2002;4:109–14. underestimated. It is also possible that some premium brands of 4. Emery S, White MM, Pierce JP. Does cigarette price influence adolescent cigarettes may have also been included in the response option experimentation? J Health Econ 2001;20:261–70.

472 Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 Research paper Tob Control: first published as 10.1136/tc.2009.029736 on 1 October 2009. Downloaded from

5. Imperial Tobacco Canada. Social Report 2006–2007 Imperial Tobacco Canada: tobac-tabac/research-recherche/stat/survey-sondage_term-eng.php (accessed 30 let’s talk. Imperial Tobacco Canada, 2007. Available at: http://www. Oct 2008). imperialtobaccocanada.com (accessed 12 Nov 2008). 14. National Youth Smoking Survey Research Consortium. 2006–2007 Youth Smoking 6. Non-Smokers Rights Association. Discount cigarettes and other cheap tobacco Survey Summary. Waterloo, ON; University of Waterloo, 2008. Available at: https://www. products, 2004. Available at: http://www.nsra-adnf.ca/cms/page1342 (accessed 20 yss.uwaterloo.ca/ysssite_app/controller/index.cfm (accessed 30 Oct 2008). Nov 2008). 15. SAS Institute Inc. The SAS system for windows. Cary, NC: SAS Institute Inc, 2001. 7. Non-Smokers Rights Association. NSRA’s Backgrounder on the Canadian 16. Leatherdale ST, Strath JM. Tobacco retailer density surrounding schools and cigarette Tobacco Industry and its market, 2006. Available at: http://www.nsra-adnf.ca/cms/ access behaviours among smoking students. Ann Behav Med 2007;33:105–11. file/pdf/backgrounder/2005_06.pdf (accessed 20 Nov 2008). 17. Leatherdale ST. Predictors of different cigarette access behaviours among 8. Cummings KM, Hyland A, Lewit E, et al. Use of discount cigarettes by smokers in experimental and regular smoking youth. Can J Public Health 2005;96:348–52. 20 communities in the United States, 1988–1993. Tob Control 1997;6:S25–30. 18. Canadian Coalition for Action on Tobacco. Health groups launch campaign for 9. Royal Canadian Mounted Police. Contraband Tobacco Enforcement Strategy, action on tobacco contraband. Press release, 24 April 2007. Available at: http:// 2008. Customs & Excise Branch, Border Integrity, Federal & Internal Operations, www.ocat.org/pdf/SmugglingReleaseApril2007.pdf (accessed 3 Jan 2009). Canada. Available at: http://www.rcmp-grc.gc.ca/ce-da/tobacco-tabac-strat-2008- 19. Rothmans Inc. Annual Report 2006. Rothmans’ Inc, June 27, 2006. Retrieved from eng.pdf (accessed 1 May 2009). Sedar.com. 10. McLaughlin P. Contraband cigarettes becoming a national norm. CMAJ 20. Johnston LD, O’Malley PM, Bachman JG, et al. Cigarette brand preference among 2007;176:167–8. adolescents. Monitoring the future occasional paper 45. Institute for Social Research, 11. Hyland A, Bauer JE, Li Q, et al. Higher cigarette prices influence cigarette purchase University of Michigan, 1999. patterns. Tob Control 2005;14:86–92. 21. Fendrich M, Mackesy-Amiti ME, Johnson TP, et al. Tobacco-reporting validity in an 12. Callaghan RC, Veldhuizen S, Leatherdale ST, et al. Contraband cigarette use among epidemiological drug-use survey. Addict Behav 2005;30:175–81. adolescent daily smokers in Canada. CMAJ 2009;181:1–3. 22. Klein SM, Giovino GA, Barker DC, et al. Use of flavoured cigarettes among older 13. Health Canada. 2006–07 Youth Smoking Survey, Ottawa: Minister of Supply adolescent and adult smokers: United States, 2004–2005. Nicotine Tob Res and Services Canada, 2008. Available at: http://www.hc-sc.gc.ca/hl-vs/ 2008;10:1209–14.

The lighter side Protected by copyright. http://tobaccocontrol.bmj.com/ on 20 August 2019 at University of Cape Town Libraries.

ß RJ Matson, The St Louis Post Dispatch.

Tobacco Control 2009;18:466–473. doi:10.1136/tc.2009.029736 473