Youth Tobacco Use before and after Flavored Tobacco Sales Restrictions in Oakland, California and San Francisco, California

Abstract

The tobacco industry has used recent findings from the Youth Risk Behavior Surveillance System Survey (YRBSS) to claim that a sales restriction on flavored tobacco products might increase youth combustible cigarette use. In this special communication, we examined YRBSS data and reached the opposite conclusion. We observed the patterns in youth cigarette smoking in Oakland, California following its 2017 convenience store flavored tobacco sales restriction. We also found that 2019 YRBSS data from San Francisco, California cannot be used to evaluate the effect of the sales restriction on all flavored tobacco products in San Francisco as the YRBSS data for this city was collected prior to enforcement of the sales restriction. For future studies, we suggest triangulating with corroborating sales, behavioral, and qualitative data over time to assess the effects of tobacco control policies on youth tobacco use. We recommend that policy enactment and enforcement dates, as well as the exact data collection periods for population health surveys, be published to facilitate more rigorous policy evaluation.

Introduction

Flavored tobacco product sales restrictions have shown promising results in reducing youth access to and use of vaping products.^1,2 Vaping products are “electronic vapor products” that include e-cigarettes, e-cigars, e-pipes, vape pipes, vaping pens, e-hookahs, and hookah pens, as defined by the Youth Risk Behavior Surveillance System Survey (YRBSS).³ However, other research studies have argued that a sales restriction on flavored tobacco products might increase youth combustible cigarette use, such as a recent analysis using 2019 YRBSS data regarding the flavored tobacco sales restriction in San Francisco, California that was enforced in January 2019.⁴ Following this publication, non-academic community members and an organization associated with tobacco industry have used this YRBSS data to justify their opposition to flavored tobacco restrictions.^5–7

In this special communication, we examined the patterns in youth cigarette smoking in Oakland, California, a city adjoining San Francisco, following the 2017 Oakland convenience store flavored tobacco sales restriction as a comparison with the findings from San Francisco, and reached the opposite conclusion. In September 2017, Oakland passed a policy to restrict the sale of all flavored tobacco products, but “adult-only” tobacco stores were exempt from this policy.⁸ If youth smoking rates increased similarly in Oakland following that city’s sales restriction, this would lend credence to the call for caution against flavored tobacco sales restrictions. However, if the patterns differ, we should identify alternate explanations for the rise in San Francisco’s youth smoking prevalence. The objective of this Special Communication was not to replicate the recent analysis,⁴ but rather provide additional descriptive evidence as to why that analysis was not valid.

Methods

We examined patterns of current (past-30-day) youth vaping and combustible cigarette use prevalences from Oakland’s 2015-2019 YRBSS data.⁹ We requested information from the Centers for Disease Control and Prevention (CDC) regarding the specific timing of when the YRBSS 2019 survey was administered to students in San Francisco relative to the January 2019 enforcement of the flavored tobacco sales restriction there. We also considered the findings from a recently published paper on tobacco sales in San Francisco following the city’s flavored tobacco sales restriction.¹⁰

Results

Oakland implemented a convenience store flavored tobacco sales restriction, effective July 2018, and data for the 2019 Oakland YRBSS was collected after this restriction went into effect. This timing of the YRBSS makes Oakland a better model than San Francisco for understanding what happens to YRBSS-measured tobacco use outcomes following the implementation of a local flavored tobacco sales restriction. Following the sales restriction, high school youth vaping declined from 11.2% to 8.0% (p=0.04) and smoking declined from 4.4% to 2.4% (p=0.02) between 2017 and 2019 in Oakland.⁹

Through an inquiry using the YRBSS request form (https://www.cdc.gov/healthyyouth/data/yrbs/contact.htm), we found that the timing of the survey period of the 2019 YRBSS survey in specific school districts varied from one another. San Francisco’s assessment period was Fall 2018 (defined as September to December). Thus, the San Francisco survey preceded the enforcement of its flavored tobacco sales restriction (January 2019), making the 2019 YRBSS an innappropriate data source for evaluating the effects of the city’s flavored tobacco sales restriction. Research on compliance with San Francisco’s flavored tobacco sales restriction found that compliance was fairly low at only 17% in December 2018 and increased through 2019 to up to 80%.¹¹ Thus, even at the end of the San Francisco YRBSS survey fielding in December 2018, few retailers were proactively enforcing the flavored tobacco sales restriction in the city. This lack of retailer enforcement suggested that the impact on actual access to flavored products did not really start to materialize until well after the YRBSS completed data collection in San Francisco.

Further, recently published data showed that total tobacco sales in 2019 did not increase in San Francisco after the enforcement of the flavored tobacco product sales restriction and actually declined by 25%; all flavored tobacco product sales declined to near zero by end of January 2019.¹⁰ Since there was no ban on non-menthol cigarette sales, we would have expected to see an increase in sales of cigarettes if youth had been switching products. And even if an increase in sales of total cigarettes was observed in San Francisco, it still would not have been attributable to youth switching products, especially given that the prevalence of menthol and non-menthol use among youth was, prior to the policy in San Francisco, already trending in the direction of lower menthol and higher non-menthol use.¹² The study actually found an overall trend of a reduction in both total tobacco sales and cigarette sales in San Francisco following the flavored tobacco product sales restriction, further suggesting that flavored products were not being substituted by other unflavored tobacco products or cigarettes.¹⁰ Although this study did not look at youth use specifically, sales data still provides insight on overall access at the time to products. Youth tend to access tobacco products from social sources, but a proportion of youth still do access tobacco products from retailers.^13,14 Additionally in Oakland, seven months after the partial flavored tobacco sales restriction went into effect, a study found that availability of menthol cigarettes was lower in census tracts that were further away from exempt tobacco stores, showing how such policies do reduce flavored tobacco product sales.¹⁵

Discussion

Data from the 2019 YRBSS survey in Oakland showed how a city with a convenience store flavored tobacco sales restriction also saw a reduction in youth vaping that was not accompanied by a rise in combustible cigarette smoking. Across the cities surveyed in this analysis, higher vaping increases from 2017 to 2019 tended to accompany lower combustible cigarette decreases over the same time period, also bolstering the argument against vaping substituting for cigarettes. It is important to evaluate the impacts of public health policies through surveillance of population behaviors to ensure they do not lead to unintended effects that harm vulnerable groups.

However, in order to imply causality, there cannot be ambiguous temporal precedence. The 2019 YRBSS survey in San Francisco predated the enforcement of the city’s flavored tobacco sales restriction, and therefore should not be used to evaluate the effect of the policy.⁴ Important data gaps exist for rigorous evaluation of tobacco control policies. For example, public and proprietary sources of tobacco control policy data, such as for local flavored tobacco sales restrictions,^16–19 do not include the policy enactment and enforcement dates that are required to avoid erroneous conclusions like those in the recent analysis of the San Francisco flavored sales restriction. We recommend that more granular policy dates, as well as the exact data collection periods for YRBSS, National Youth Tobacco Survey, and other population health surveys be published to facilitate more accurate policy evaluation. When using single school district surveys, we suggest triangulating with corroborating data, such as sales data, behavioral data, compliance checks, and qualitative interviews to assess what effects tobacco control policies have on youth tobacco use.

Additionally, we want to emphasize the need for quality over quantity of data – it is important that such studies utilize more robust study designs and methods, such as studies that include multiple time points and not just a pre-post assessment, to make meaningful conclusions. Moreover, self-reported longitudinal and qualitative data coming directly from youth would be particularly compelling in documenting the actual impact of policy on the target population.^20,21 The non-statistically significant increase in cigarette smoking in youth before the flavored tobacco sales restriction in San Francisco may be due to unstable measurement of a relatively small number of smokers, or transient effects of the legalization of marijuana in San Francisco in 2018.^22,23

Other critiques of the analysis have been published and highlight similar concerns, such as the use of only a single post-policy data point, lack of an analysis of vaping products, potential presence of other confounding factors, and lack of recognition of dual use.^24–26 The author addressed these critiques and called attention to the need to more explicitly identify what substance(s) youth are vaping when conducting surveys.²⁷ The purpose of our special communication is to highlight additional concerns and continue to provide supporting evidence for the importance of enacting flavored tobacco product sales restrictions in reducing youth use of tobacco and vaping products.

Tobacco use begins as a social behavior, and access to flavors leads to re-normalization of tobacco use among youth and a higher risk of initiating cigarette smoking.^28,29 Vaping is not harmless – inhaled ultrafine particles cause lung inflammation and asthma exacerbations, and nicotine has serious mental health impacts on the developing brain.³⁰ It is important that public health practitioners, healthcare professionals, and researchers continue to support and advocate for tobacco control policies that limit youth access to vaping products.