Commentary on Lipperman‐Kreda <i>et al</i>. (2016): Robust methods with a weak outcome measure still lead to consistent conclusions—even so, it may be time for recommendations to move from traditional tobacco control strategies to a relevant endgame strategy

This study found that tobacco retailer density is associated with smoking onset, and recommends the traditional tobacco control solutions around restricting retailer location or having stricter enforcement of age-of-majority non-compliance among retailers. In future, prevention efforts may be better served by also considering the relevant tobacco end-game strategy. Higher tobacco retailer density in the cities and communities surrounding where youth live and learn is associated with an increased risk of smoking onset and progression. The work by Lipperman-Kreda et al. 1 presents a robust methodological study on which to strengthen the evidence base in support of this conclusion. The following is not a criticism of the quality of work or the recommendations made by the authors; rather, it is a critical reflection about a subtle but important detail overlooked in their methods and an important, but not so subtle, additional recommendation for action that I believe should not be overlooked in the future. Lipperman-Kreda et al. 1 report on a large multi-phase study that involved (a) sampling a large cohort of youth from 50 communities and following them over time, (b) a protocol testing tobacco retailer compliance within those 50 communities and (c) key informant interviews with local law-enforcement agencies in each community. The authors should be commended for the successful development, implementation and execution of such a large study, which incorporated valuable contextual information to supplement individual behavior measures across multiple jurisdictions over time. From this effort, they report that tobacco outlet density was associated positively with both life-time smoking and youth perceptions of tobacco availability, while associated negatively with perceived enforcement of age of majority laws 1. These conclusions align appropriately with previous evidence 2-7. However, despite a substantial amount of time and effort that was put into developing a strong and comprehensive study protocol, it was surprising that the authors then used what may be considered the least informative outcome measure for youth smoking—cigarette ‘ever use’. While an entire scientific paper could describe the problems associated with using ever use as an outcome measure in this research, I will instead highlight only three issues. First, it ignores the volumes of research demonstrating that youth progress through a variety of stages when becoming a smoker 8; research examining the link between tobacco retailer density and youth smoking must consider transitions beyond ever trying a cigarette to be most informative. This is supported by the results presented in their Fig. 1 [1], where it appears that over time outlet density is not important; their static measure of ever use probably masks the more meaningful relationship one would expect to see once someone has transitioned beyond ever trying. Secondly, we know that many youth report trying a cigarette at one time in their life but do not progress to more advanced smoking experimentation 9; these are not the youth we are as concerned with. Thirdly, a measure of ever use is not appropriate when looking at retailer access, given the social exchange of cigarettes among youth, especially non-established youth smokers who rarely access cigarettes from retailers 10. It is not clear if other smoking behavior measures are available within the host study 1 but, if measures of experimentation or established smoking are available, they are worth exploring. Tobacco end-game concepts are designed to switch the discussion from simply trying to control tobacco use to implementing actions that eliminate tobacco use in future generations 11. The recommendations made by Lipperman-Kreda et al. 1 align with the more traditional paradigm of trying to control tobacco use, as they conclude that future efforts should focus upon controlling the density and location of retailers and more effective enforcement of underage sales laws. While reasonable, they are really designed simply to prevent underage youth from accessing tobacco until they reach the legal age of majority; tobacco use is delayed, but not entirely prevented. In future, it may be important to supplement recommendations focused on control with additional recommendations focused on the elimination of future tobacco use. One emerging end-game proposal relevant to the work presented here is restricting tobacco sales by year born (i.e. no one born after 2000 is ever legally allowed to purchase tobacco products) 11. This would create a tobacco-free generation, it would make it substantially easier to control and enforce the sale of tobacco products from retailers (e.g. there is no need to calculate age when checking identification, which is a common reason for non-compliance 12) and it would eventually phase out the need for tobacco retailers altogether. While there are many considerations associated with this approach, it is currently being proposed as a policy option in Singapore 13 and is supported by the British Medical Association 14. Considering that we have had the same recommendations in tobacco retailer and tobacco access literature for more than a decade and little progress appears to have been made, end-game recommendations may warrant consideration in this paradigm of research. S.T.L. is a Chair in Applied Public Health Research funded by the Public Health Agency of Canada (PHAC) in partnership with Canadian Institutes of Health Research (CIHR).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.