Cash rewards with free medicines doubled six-month quit rates, to 8.8%, in smokers sent for lung screening

A lung scan is offered to long-term smokers to catch cancer early. It is also one of the few times a health system has a heavy smoker’s attention, and Medicare, the public insurer for older Americans, requires that help with quitting be offered alongside it. What kind of help works for the patients least likely to get it has been an open question.

A trial published this week in JAMA tested four answers on more than 3,200 smokers from groups that health care serves poorly. Paying people to quit, on top of free medicines, roughly doubled the number who managed it. Free medicines by themselves made no clear difference. Even with the payments, more than nine in ten did not stay off tobacco for the six months.

Why lung cancer screening is a moment to quit smoking

Screening is aimed at people with decades of smoking behind them. Annual lung cancer screening is recommended for adults aged 50 to 80 years who have at least a 20 pack-year smoking history and who still smoke or stopped within the past 15 years. A pack-year is one pack a day for one year. The scan itself is low-dose computed tomography (CT) lung cancer screening, a CT scan being a set of X-ray images assembled by computer.

Quitting matters at least as much as the scan. An analysis of the large American trial that established screening compared the two. Its conclusion was that seven years of smoking abstinence reduced lung cancer-specific mortality at a magnitude comparable with the screening itself, and that doing both was better than either.

That is the reasoning behind the Medicare rule. The agency that runs the program requires that health systems offer smoking cessation services to patients undergoing lung cancer screening. What those services should be is left open, and the trial’s starting point was that the most effective methods to achieve smoking cessation among medically underserved patients undergoing lung cancer screening are unknown.

In practice the help is often thin. A review this year of 55 studies of screening programs found that implementation most often involved delivery of low-intensity behavioral interventions, and that many lung cancer screening programs globally still have not implemented formal cessation supports.

How the smoking cessation trial worked

The trial ran at four American health systems. Participating health systems included Penn Medicine, Geisinger, Henry Ford Health, and Kaiser Permanente Southern California. It enrolled people who smoked, had a screening scan ordered, and who identified as Black, Hispanic, living in a rural area, or being of low socioeconomic status. Low socioeconomic status meant a high-school education or less, or a household income under twice the federal poverty line.

Of 9,963 patients assessed, 3,259 were randomized. Their median age was 61, and six in ten were women. A quarter were Black, 8.9% were Hispanic and 38.8% lived in rural areas. More than half, 55.5%, reported smoking more than 10 cigarettes per day at enrollment.

Enrollment was unusual, and it shapes how the results should be read. The plan registered before the trial began was that eligible patients will enroll with opt-out consent. People were invited by text message or email, and for much of the trial the default was to be included. Patients who neither opted out nor clicked ‘Complete’ to be randomized received an email and/or text message after 14 days, and were then randomized unless they objected. Many of the people in this trial were therefore not volunteers who had set out to quit. They were close to the ordinary run of smokers who turn up for a scan.

Each was assigned at random to one of four groups, each adding something to the one before.

Group	What was offered
1. Usual care	Staff ask about smoking, advise quitting and refer to services
2. Free medicines	Group 1, plus free nicotine replacement and reimbursement for prescription quitting drugs
3. Payments	Group 2, plus up to $600 for quitting
4. Payments and messages	Group 3, plus a text-message tool about future health

The usual-care approach is known as ask-advise-refer. In the registry’s description, non-study clinicians ask smokers about their desire to quit smoking, advise them to quit, and provide informational resources such as telephone quit lines and classes.

The payments were financial incentives up to $600 contingent on biochemically confirmed smoking cessation. The money came in rising steps: $100 at two weeks, $200 at three months, and $300 at six months, confirmed via blood or urine testing. Nothing was paid on a participant’s word alone.

The fourth group also received a mobile health tool motivating patients to think about their future health, delivered by text message.

The measure of success was strict. It was biochemically confirmed, sustained tobacco abstinence through 6 months, meaning a person had to stay off tobacco throughout and have it confirmed by laboratory tests.

Quit rates with cash rewards, medicines and advice

Sustained tobacco abstinence rates through 6 months were 4.3% with usual care, 5.1% with the addition of pharmacotherapy, 8.8% with the addition of financial incentives, and 7.2% with the addition of mobile health tool. Pharmacotherapy means treatment with medicines.

Group	Stayed off tobacco for six months
Usual care	4.3%
Plus free medicines	5.1%
Plus payments	8.8%
Plus payments and messages	7.2%

The payments made the difference. Compared with usual care, adding them raised the quit rate by 4.6 percentage points, with a plausible range of 2.1 to 7.0. Compared with free medicines alone, the gain was 4.1 points. Those are the paper’s figures after statistical adjustment, which is why they differ slightly from the gaps between the raw percentages in the table. Put another way, about 22 people had to be offered the payments for one more of them to quit for six months.

Free medicines on their own did not help in any measurable way. Adding free pharmacotherapy to ask-advise-refer was not superior to ask-advise-refer alone. The adjusted difference was half a percentage point, with a range from 1.7 points worse to 2.6 points better.

The text-message tool added nothing visible. The group that received it alongside payments had a quit rate of 7.2%, below the 8.8% of the group that received payments without it.

Scott Halpern of the University of Pennsylvania, the trial’s principal investigator, described who these patients were. “These participants had smoked for decades and likely made multiple unsuccessful quit attempts in the past. We found that adding financial incentives increased quit rates in a population that can be especially difficult to reach with traditional smoking cessation approaches,” Halpern said.

The lead author, Joanna Hart, also of the University of Pennsylvania, put the stakes this way: “Finding new approaches that help people successfully stop smoking can have an enormous impact, not only on their health but also on their quality of life.”

Why free quit-smoking medicines changed so little

The medicines in question are known to help people who use them. A pooled analysis of 85 trials in people old enough and with enough smoking history to qualify for screening found that both counseling and medicines raised quit rates at six months. Its authors noted at the time that there are limited data to guide clinicians and policy-makers in choosing cessation interventions in this setting.

The new trial tested something narrower: whether removing the cost of those medicines, for people who had been enrolled by default, would move the numbers. It did not. One reading is that price was not what stood in the way. A free prescription still has to be requested, collected and taken, by someone who may not have been planning to quit that month. A payment changes the reason to try.

That is an interpretation. The trial compared offers, and an offer of free medicine can only help the people who take it up.

How cash rewards fit with earlier smoking cessation research

Paying people to quit is one of the better-tested ideas in tobacco research. The standing Cochrane review, updated last year, concluded that there is high-certainty evidence that incentives improve smoking cessation rates at long-term follow-up in mixed population studies. It also addressed the obvious worry, that people go back to smoking once the money stops. The reviewers found that the effectiveness of incentives is sustained even when the last follow-up occurs after the withdrawal of incentives.

What the new trial adds is the setting and the people. Its investigators describe it as the largest trial to assess financial incentives for smoking cessation among patients referred for lung cancer screening, and it was run in groups that quitting programs often fail to reach.

What the trial does not show about quitting smoking

Most people did not quit. The best group reached 8.8%. Doubling a small number leaves a small number. That is still a meaningful gain for a habit this dangerous: tobacco is blamed for an estimated 450,000 deaths in the US each year.

Six months is early. The published result stops at six months, when the last payment was made. The registered plan states that relapse rates will be tracked at 12 months. Whether the quitters stayed quit after the money ended is the next thing to learn.

Payments came bundled. Nobody received payments without also being offered free medicines. The trial shows that payments added to free medicines work better than free medicines alone. It cannot say what payments would do by themselves.

Enrollment by default. Opt-out enrollment makes the result more realistic and the quit rates lower than in trials of motivated volunteers. The figures should not be compared directly with those trials.

Who would pay. A health system would have to fund the rewards and the testing. The trial was supported by a non-profit research funder, the Patient-Centered Outcomes Research Institute. Whether rewards would repay their cost in later health spending is a separate calculation.

Where it was done. Four health systems in the United States took part. Results could differ where screening and insurance are organized differently.

The kind of support that suits a person trying to quit depends on their history and health, which is assessed by their doctor.

Payments of up to $600 raised the share who stayed off tobacco for six months from 4.3% to 8.8% in a trial of 3,259 smokers from medically underserved groups who were referred for lung cancer screening, while free medicines alone made no clear difference and more than nine in ten people in every group did not stay off tobacco for the six months.