Objective: In this paper, we characterize how potential policies restricting access to tobacco products may impact use behaviors among adult, past 30-day, smokers and e-cigarette users. Methods: We conducted an online experiment with 820 smokers, e-cigarette users, and
dual users (April 27-June 8, 2020). We randomized participants to one of 4 hypothetical access scenarios: (1) tobacco retail stores open + pharmacies open (TOPO); (2) tobacco stores open but favorite brand unavailable + pharmacies open (TOPO-NFB); (3) tobacco stores closed + pharmacies open
(TCPO); and (4) tobacco stores closed + pharmacies closed (TCPC). Outcomes (measured on 0-100 visual analog scales) included the likelihood of quitting, reducing, switching brands or products, and finding another source of tobacco products. Seemingly unrelated regressions tested for associations
between access scenarios and prospective tobacco use behaviors. Results: Participants in the TCPO and TOPO-NFB scenarios were more likely to reduce use, switch brands/products, and find another source (ps < .001) than those in the TOPO scenario. Dual and flavored product users
were more likely to switch products (ps < .01). Conclusions: When tobacco retailers are closed, tobacco users may be more likely to quit and/or reduce use compared to when retailers are open. However, access restrictions could prompt users to switch tobacco brands/products
or sources.
An alternative numerical method for estimating large-scale time-varying parameter seemingly unrelated regressions models
