How Many Deaths Will New Cigarette Warnings Prevent?

Assuming they survive an ongoing legal challenge and are as effective as advocates hope, the new graphic warnings to appear on cigarette packages beginning next year could prevent millions of deaths from smoking over the following 8 decades, researchers estimated.

Modeling by Rafael Meza, PhD, of the University of Michigan in Ann Arbor, and colleagues indicated that, depending on the degree to which the warnings discourage smoking initiation and prompt current users to quit, anywhere from 275,000 to 5.5 million deaths would be averted by the year 2100.

These warnings use unflinching photographic imagery to depict the many adverse effects from smoking: fogged eyes from cataracts, sternum scars from coronary bypass surgery, gangrenous toes, bloody urine — and, of course, diseased lungs.

A middle-of-the-road scenario, in which smoking initiation declines 10% and quit rates increase 50% because of the warnings, would lead to 539,000 fewer smoking-attributable deaths, the group reported in JAMA Health Forum. As well, under this scenario, some 7.9 million life-years would be gained.

The lowest figures came from a scenario in which quit rates increase only 25% as a result of the warnings, and rates of new smoking fall by just 5%. Even then, according to Meza and colleagues, 4.0 million life-years would be gained by 2100, along with 275,000 fewer smoking-attributable deaths.

An online-only supplement to the group’s main paper included estimates for shorter time frames. Within 10 years of the warnings’ first appearance, for example, the middle-of-the-road scenario would see 30,900 fewer deaths from smoking, though only 90,000 life-years gained. With more optimistic assumptions — 15% reductions in smoking initiation and 75% increases in cessation — 45,400 deaths would be prevented and 130,000 life-years gained by 2032.

But the caveats noted in this article’s lead sentence can’t be ignored. Obviously one could dispute the model’s assumptions about quit rates and new smoking. The investigators acknowledged “uncertainty about the true magnitude” of the warnings’ eventual effects, though their modeling relied on previous research into the issue, including a Canadian study that found substantial decreases in smoking after graphic warnings were introduced.

Meanwhile, a separate article by a senior JAMA system editor sought to address the legal aspect. The FDA had originally set the requirement to take effect in 2012, but the tobacco industry went to court to block it. A federal district judge agreed that the warnings, as then designed, were “not … purely factual and uncontroversial,” and thus the requirement to put them on cigarette packs violated manufacturers’ First Amendment rights, and this was upheld on appeal.

Since then, the FDA went back to the drawing board, literally, to devise new image-based warnings that would better meet the “purely factual” standard. These, too, are now being litigated in a suit brought by the U.S.’s leading cigarette maker, R.J. Reynolds.

In the second JAMA Health Forum article, Gregory Curfman, MD, editor for health policy and law in the JAMA Network, argued that the new versions do meet that standard. “FDA explicitly created new graphic images that are factually based and accurately depict human anatomy and disease pathology. The images are based on expert professional knowledge,” he wrote. These, he added, “are objectively based and were not created to evoke an emotional response.”

But Curfman, who has no formal legal training, also predicted that the R.J. Reynolds suit would fail because the courts offer substantially less protection to commercial speech than that of a political nature. “[T]he new graphic warning labels, by providing accurate information vetted by expert professionals that is of value to the public, contribute to democratic competence and warrant constitutional protection on this basis,” whether cigarette makers like it or not, he wrote.

A constitutional law scholar contacted by MedPage Today didn’t think the courts would buy that argument. “Commercial speech has been treated as practically the equivalent of political speech for years,” said Bruce Ledewitz, JD, of Duquesne University in Pittsburgh.

He pointed out that the Supreme Court struck down state regulations on tobacco company billboards and store displays in a 2001 case, Lorillard Tobacco v. Reilly, on First Amendment grounds. (He didn’t mention it, but in the 20 years since that decision, the Supreme Court has appeared to become friendlier to businesses in First Amendment cases, including Masterpiece Cakeshop v. Colorado Civil Rights Commission and Citizens United v. Federal Elections Commission.)

Ledewitz also said the content of the government’s warnings are irrelevant to this issue.

Curfman wrote that if the new warnings are indeed struck down in court, it could have broad ramifications for other regulations meant to protect public health, particularly those that force companies to publish factual information (such as adverse health effects of soft drinks) against their wishes.

A secondary analysis in the paper by Meza and colleagues also deserves note. The group estimated the numbers of prevented deaths and life-years gained had the FDA’s graphic warnings gone into effect in 2012 as originally planned, using the same parameters around quit rates and new smoking.

Under the middle-of-the-road scenario, with 10% declines in smoking initiation and 50% increases in cessation, prevented deaths by 2100 jumped to 718,000 — an increase of 179,000 — and the increase in life-years reached 11.2 million, or an extra 3.3 million compared with introducing the warnings in 2022.

By 2032, that earlier introduction in the same scenario would have saved nearly 132,000 lives.

“This study’s findings suggest that preventing delays to policy implementation should be a public health priority,” Meza and colleagues concluded.

Besides the lack of certainty around the graphic warning’s true effects, the investigators noted other limitations. The base model included projections of quit rates and new smoking in the absence of warnings, based on previous trends that, of course, may not predict the future. Also, potential effects from the COVID-19 pandemic were not modeled, and these could be significant.