High school student use of cigarettes and e-cigarettes fell after the San Francisco flavor ban took effect

San Francisco’s experience continues to be a central issue as localities and states continue to debate legislation ending the sale of flavored tobacco products, including the tobacco companies’ effort to overturn California’s flavor ban by forcing a referendum on it.

Results from the 2019-20 California Student Tobacco Survey show that that, compared to the survey conducted two years earlier (2017-2018), “San Francisco” high school student cigarette smoking prevalence fell from 3.1% to 1.6% (graph above). This fall continued a downward trend compared to 2015-2016, when 3.9% of “San Francisco” high school students smoked cigarettes.

The changes were comparable to changes observed statewide (graph above).

I put San Francisco in quotes (and put an asterisk on the graph) because the California Student Tobacco Survey reported San Francisco results pooled with other local counties in 2017-2018 (with San Mateo County) and 2015-16 (with Bay Area counties San Mateo, Martin, Contra Costa and Solano). This means that the SF numbers are not 100% comparable, but the latest 2019-2020 numbers are just from San Francisco (which is both a city and a county) and clearly show low high school smoking.

Most important, the latest San Francisco cigarette smoking is well below the 6.2% estimated for 2019 in Abigail Friedman’s analysis of the CDC’s Youth Risk Behavioral Surveillance System (YRBSS) data. This paper, which was not based on any data collected in 2019 after the SF law started being enforced, has been widely quoted by tobacco and e-cigarette companies in opposing flavor restrictions. As illustrated in the graph below which superimposes the California Student Tobacco Survey data on her data for comparison shows that the most recent estimates show low cigarette smoking among San Francisco high school students.

The California data, while not 100% comparable with the YRBSS, undermines the confidence policymakers can have in statements that youth smoking went up substantially after the San Francisco ordinance. As I wrote when discussing her paper before, Friedman’s conclusion is based on a single point collected while the SF ordinance was still being implemented. In contrast, the California Student Tobacco Survey was conducted from September 2019 to March 2020, well after implementation of the ordinance.

The fact that sales of flavored tobacco products dropped precipitously in San Francisco, but not other California cities, also supports the conclusion that the ordinance worked.

Having said, that, there are some technical differences between Friedman’s YRBSS analysis and my presentation of the California Student Tobacco Survey that people should keep in mind:

• Friedman limited her analysis to high school students 18 and younger; the CSTS has no age restriction. This is unlikely to make much difference because only about 1% of high school students are 19 or older.
• The California statewide numbers in the graph above include San Francisco. To do a formal statistical analysis the SF students should be removed. This is unlikely to make much difference because the 2371 San Francisco students were only 1.6% of the total of 150,608 students who participated in the SCTS.
• As noted above, what counted as “San Francisco” changed over the three years that the CSTS was reported. Fortunately, the 2019-2020 result was for San Francisco alone.
• The sample size for San Francisco was relatively small, so there is a lot of uncertainty in the precise smoking rate, which is indicated by the broad 95% confidence interval for the estimates. Even with this uncertainty, however, the level of smoking was still dropping.

The CSTS also reported levels of e-cigarette use for the three survey years:

Consistent with national trends, e-cigarette use was much higher than cigarette use. (Note the different scales on this graph and the one at the top of this blog post.) Most important, e-cigarette use dropped a lot more than cigarette use. This is an important finding because reducing youth e-cigarette use was one of the major aims of the San Francisco flavor ban ordinance. But both cigarette and e-cigarette use dropped following the ordinance.

Thus, the most recent data adds to the case that the San Francisco flavor ban ordinance was a success.

Note: The San Francisco flavor ban covered all flavors, including menthol cigarettes and flavored cigars and other flavored tobacco products. These are important, too, especially since the tobacco companies use these products to target African Americans and other groups.