A Review of: Rick Kosterman et al., “Adult social environments and the use of combustible and electronic cigarettes: Opportunities for reducing smoking in the 30s,” Nicotine and Tobacco Research (January 2020).

According to the CDC, about 1 in 5 deaths in the United States—more than 480,000 per year—are the result of cigarette smoking. And, although fewer U.S. adults smoke than in previous years (about 14 percent in 2018 compared to almost 21 percent in 2005), smoking remains the nation’s leading cause of preventable disease and death.

Perhaps because nicotine is so addictive, much of the success in reducing smoking prevalence in the United States results from preventing youth uptake of the habit. While almost 70 percent of adult smokers say they want to quit, only about 7 percent succeed in a given year, often after dozens of attempts. Those individuals that do manage to kick the habit are most likely to do so in their 30s.

Positing that this age range is “a transitional time that includes longer term considerations of one’s own health and functioning,” Rick Kosterman and colleagues examined the role played by the social environment of a cohort of 30-something smokers in order to better inform public health efforts to promote smoking cessation. To meet these aims, the authors drew data from the Seattle Social Development Project (SSDP), a longitudinal study of health behaviors and outcomes. The SSDP began in 1985, gathering information from a cohort of 808 fifth graders at 18 public schools in “higher crime” neighborhoods throughout Seattle, Washington. The data used in the present study were collected via in-person interviews and web-based surveys conducted in 2005, 2008 and 2014, when the students were approximately 30, 33 and 39.

Of the SSDP cohort, 37 had died by age 39, while approximately 90 percent of the remaining members were retained through all three follow-ups. The resulting population for Kosterman et al.’s sample was therefore as follows: 51 percent male; 47 percent European American, 26 percent African American, 22 percent Asian American, 5 percent Native American; 5 percent Hispanic; and 52 percent had experienced childhood poverty between 10 and 12 years of age.

Using this longitudinal approach, and guided by the Social Development Model (SDM), the study adopted the following three broad aims: 1) To identify measures of the social environment of mid-30s individuals that could provide pathways to cigarette use and health promotion; 2) To model the continuity of cigarette use during the 30s and explore how the social environment might mediate this continuity and; 3) To examine whether the social environment affected electronic cigarette use.

To achieve these aims, the authors identified several measures of the social environment across several life domains that could be relevant to smoking and health behaviors. Then, at age 33, participants were asked about the following constructs:

  • perceived opportunities to interact with individuals who engage in health-promoting activities;
  • actual involvement with individuals who engage in health-promoting activities;
  • perceived rewards—for example, support and encouragement—from people who engage in health-promoting activities;
  • bonding to individuals who engage in health-promoting activities;
  • beliefs about the personal value of active leisure and health-promoting activities;
  • perceived opportunities to interact with smokers;
  • actual involvement with individuals who smoke cigarettes;
  • perceived rewards—such as being able to count on someone for emotional support—from relationships with smokers;
  • bonding to smokers;
  • belief that cigarette use is socially acceptable;
  • skills for engagement in health management, such as getting adequate sleep or controlling stress. 

As expected, the study found that, for most smokers, cigarette use stays relatively stable over time. Of the participants in the study, 32 percent reported past-month combustible cigarette smoking at age 30, 24 percent reported past-month combustible cigarette smoking at 39 and 20 percent reported past-month combustible cigarette smoking at both ages. Smoking behaviors were not strongly correlated with demographic qualities, but were inversely associated with educational attainment: individuals with higher education were less likely to smoke combustible cigarettes at both 30 and 39 years of age.

Kosterman et al. also found that, despite the confirmed continuity of smoking over time, smoking behaviors were strongly correlated with SDM-derived measures for opportunities, involvement, rewards and bonding. For example, the model showed that individuals who affiliated with smokers at age 33 were more likely to use combustible cigarettes at age 39 than their counterparts who did not interact with smokers. Similarly, individuals who reported a belief that cigarette use was socially acceptable—for example, they thought it was okay for adults to smoke cigarettes—also predicted higher rates of smoking at age 39 than those who did not indicate that smoking was socially acceptable.

Conversely, the model showed a small but significant association between exposure to a health-promoting environment—one that included interacting and bonding with active individuals—at age 33 and later combustible cigarette use. Digging a bit deeper, the researchers found that those who had been exposed to childhood poverty were more likely to live in a cigarette-using social environment at age 33.

When it came to e-cigarette use, the study only asked participants about this technology at age 39, and thus it lacked longitudinal data. Nevertheless, they found that those who had been surrounded by cigarette users at age 33 were more likely to report later e-cigarette use. However, their beliefs about the social acceptability of cigarette use was not correlated with later e-cigarette use.

Based on these findings, Kosterman and colleagues conclude that, “after accounting for the strong stability in cigarette use over time, social environmental factors in the intervening years still played a significant role for later cigarette use.” Furthermore, they argue that, while both the cigarette-using environment and associated beliefs about cigarette-use acceptability represent a potential pathway for interventions to reduce combustible cigarette smoking, only the environment appears relevant to e-cigarette use.

These findings suggest that a person’s social environment is a potential intervention point for public health strategies aimed at encouraging people to quit smoking. However, the study does have several weaknesses that might be corrected in future studies, including geographic limitations and a reliance on self-report measures. Another major limitation is the collection of data related to e-cigarette use. As mentioned previously, because e-cigarettes are a relatively new technology, the study did not gather longitudinal data. Consequently, it is unclear whether e-cigarette use represents a similarly stable behavior to combustible cigarette use. In addition, because there were no measures of social environments or beliefs that were specific to e-cigarettes, the study may underestimate the potential roles they play on behavior. Thus, while Kosterman et al. makes an important contribution to our understanding of combustible cigarette use and pathways for intervention, the researchers strongly urge others to explicitly include e-cigarettes and e-cigarette-related social environments and beliefs in future studies.