Keeping smoking affordable in higher tax environments via smoking thinner roll-your-own cigarettes: Findings from the International Tobacco Control Four Country Survey 2006-15

Abstract

Background: Roll-Your-Own tobacco (RYO) use is increasingly popular in many countries: it is generally cheaper than factory-made cigarettes (FM), and smokers can further reduce costs by adjusting the amount of tobacco in each cigarette. However, the level of risk of RYO compared with FM cigarettes is similar and does not meaningfully change with cigarette weight. We assessed the weight of tobacco in RYO cigarettes across jurisdictions with differing tobacco taxes/prices and over time.

Method: Six waves of the International Tobacco Control 4 Country longitudinal study of smokers and recent ex-smokers, providing 3176 observations from exclusive RYO users covering 2006-15, are used to calculate the weight of tobacco used in RYO cigarettes in the US, Canada, Australia, and the UK. Multilevel regression analyses were used to compare weights across countries, socio-demographic factors, and over time.

Results: Smokers in the UK and Australia, where tobacco is relatively expensive, show higher levels of exclusive RYO use (25.8% and 13.8% respectively) and lower mean weights of tobacco per RYO cigarette (0.51 g(sd 0.32 g) and 0.53 g(0.28 g)), compared with both Canada and especially the US (6.0% and 3.5%, and 0.76 g(0.45 g) and 1.07 g(0.51 g)). Smokers in the UK and Australia also exhibited a statistically significant year-on-year decrease in the mean weight of each RYO cigarette.

Conclusions: Taxation of RYO should increase considerably in the UK and Australia so that RYO and FM cigarettes are taxed equivalently to reduce RYO attractiveness and inequalities. Other measures to reduce the price differentials, including taxing RYO solely on weight, are also discussed.

Keywords: Hand-rolled; ITC; Roll-your-own; Taxation; Tobacco; Weight.

Figure 1.

Mean weight of tobacco (in grams) per roll-your-own cigarette over time for the United Kingdom (UK) and Australia (AU), alongside the model predicted values from a fully-adjusted multilevel linear regression, controlling for time to first cigarette, sex, age, ethnicity, level of education, and income. *Note: no data was collected from the UK in 2011 and no data collected from the UK or Australia in 2012. Sample sizes from Australia for 2009 and 2011 were low (n = 26 and n = 25, respectively, see text and Table 1 for more details).