The Noncommunicable Disease Divide: How Tobacco Became a Disease of Poverty

In 1965, when the US Surgeon General's report on smoking was released, smoking prevalence among American adults was 42%. The gradient was socioeconomic, but it ran in the opposite direction from today: smoking was more common among the affluent and educated than among the poor. College graduates smoked at higher rates than high school dropouts. White-collar workers smoked more than blue-collar workers. The cigarette was, in the mid-20th century, a marker of sophistication and social status. The public health campaign against smoking—the most sustained behavior-change effort in history—changed that. But it changed it unevenly. By 2023, smoking prevalence among US adults with a graduate degree was 4.5%. Among adults with a GED, it was 24%. The gradient had reversed. Smoking had become a disease of poverty.

The reversal of the socioeconomic gradient of smoking is one of the most consequential and least discussed facts in public health. The mechanisms are multiple: the affluent are more likely to be exposed to anti-smoking messaging, more likely to have access to cessation support, more likely to work in smoke-free environments, more likely to live in social networks where smoking is stigmatized, and more likely to have the resources—financial, psychological, social—to navigate the challenges of quitting. The poor face the opposite conditions in every dimension. They are more likely to live in environments where smoking is normative, less likely to have access to cessation support, more likely to work in jobs where smoking is prevalent, and more likely to experience the chronic stress—financial insecurity, housing instability, discrimination—that makes nicotine withdrawal harder to endure. The 'choice' to smoke is shaped by circumstances that make quitting dramatically harder for some people than for others. Treating smoking as an individual behavioral choice, independent of context, ignores the structural conditions that determine who smokes and who quits.

The mental health dimension of the gradient is particularly stark. Smoking prevalence among people with serious mental illness (schizophrenia, bipolar disorder) is two to three times the general population rate—estimated at 40-60% depending on the population and the diagnosis. Among people with substance use disorders, smoking prevalence exceeds 70%. The relationship is bidirectional: nicotine provides short-term relief from psychiatric symptoms (anxiety, negative affect, cognitive disorganization) that are inadequately treated by the mental health system, and the psychiatric symptoms make nicotine withdrawal more intense and more distressing. The mental health system has historically tolerated—and in some cases, facilitated—smoking among patients, providing cigarettes as behavioral incentives and accepting smoking as a lesser evil compared to other substance use or psychiatric decompensation. The result is a population of smokers who are highly nicotine-dependent, poorly served by the healthcare system, and systematically excluded from the cessation support that the general population receives.

The incarceration dimension adds another layer of structural inequality. Smoking prevalence among incarcerated populations is approximately 50-70% in countries where smoking is permitted in correctional facilities. In the United States, where many prisons and jails have implemented smoking bans, the prevalence upon entry remains high, and relapse upon release is near-universal. The criminal justice system—which disproportionately incarcerates poor people, people of color, and people with mental illness—functions as a mechanism for concentrating smokers and intensifying their addiction during incarceration (for those in facilities that permit smoking) or imposing forced abstinence without cessation support (for those in smoke-free facilities, who almost universally relapse upon release). The carceral dimension of smoking inequality is almost entirely absent from the tobacco control discourse, despite the fact that the populations cycling through the criminal justice system are among the heaviest-smoking populations in any society.

The cessation support infrastructure compounds the inequality. Clinical cessation services—NRT, prescription medications, behavioral counseling—are most available to people with health insurance, flexible work schedules, transportation, and the social capital to navigate the healthcare system. These are precisely the resources that poor smokers lack. Quitlines are available to the uninsured, but awareness and utilization are low among the populations that could benefit most. Community-based cessation programs—the most accessible model for marginalized populations—are underfunded and unevenly distributed. The result is a 'reverse targeting' of cessation resources: the smokers who are most likely to succeed with minimal support (the affluent, educated, highly motivated) have the most access to support, while the smokers who need the most support (the poor, mentally ill, socially marginalized) have the least. The public health infrastructure for smoking cessation is not designed to address inequality. It is designed in a way that amplifies it.

Addressing the inequality of smoking requires more than better cessation services. It requires addressing the structural conditions that make smoking rational—the chronic stress, the inadequate mental health care, the environments where smoking is one of the few available coping mechanisms. The conversation about smoking and inequality is inseparable from the conversation about poverty, mental health, housing, and social justice. The public health community has historically been reluctant to engage with these broader structural determinants, preferring to focus on tobacco-specific policies (taxation, advertising restrictions, smoke-free environments) that are within its institutional mandate and technical expertise. But the policies that have worked for the general population have diminishing returns for the populations where smoking is most concentrated. The remaining smokers in high-income countries are not the smokers who could be reached by better messaging or modest tax increases. They are the smokers whose smoking is embedded in conditions of structural deprivation that no tobacco-specific policy can address. Reaching them requires a public health that engages with inequality—not as a background condition, but as the primary determinant of who smokes, who quits, and who dies.

Shareable insight: Smoking is no longer a habit of the affluent. It's a symptom of inequality—concentrated among the poor, the mentally ill, the incarcerated, and the marginalized. Reducing smoking in these populations requires not just better cessation support, but addressing the structural conditions that make smoking one of the few coping mechanisms available.