Want to increase the odds that someone will quit smoking? Make them pay money up front - money they could lose if they relapse - and give them a hefty cash reward if they succeed.

That's the lesson of a new study in the New England Journal of Medicine that compared five smoking cessation techniques in 2,538 employees of the CVS drug store chain, along with their friends and relatives, using a payment system in which participants could earn roughly $800.

The study "is important because we have to figure out the most effective and efficient ways to deliver smoking cessation treatment," said Dr. Norman Edelman, senior scientific advisor for the American Lung Association, who was not associated with the research.

Compared to programs that simply offered cash for continued abstinence, programs that required the up-front cash deposit were twice as effective among people who agreed to make the deposit in the first place.

"It leveraged people’s natural aversion to losing money," said lead author Dr. Scott Halpern of the Perelman School of Medicine at the University of Pennsylvania.

The only problem: not a lot of people were willing to buy into that type of program. Most preferred to sign up for programs that simply offer the cash reward if they are successful. But their failure rate was much higher.

"The trick now is to refine the deposit programs so they'll be more popular without losing much, if any, of their effectiveness," Halpern told Reuters Health.

About 18 percent of U.S. adults still smoke. But smoking cessation programs are notoriously ineffective, which is why the American Lung Association advises smokers that repeated attempts may be required to get off cigarettes.

Previous research has suggested that an employee who smokes costs an employer an extra $4,000 to $6,000 a year. Thus, many companies use a variety of incentives, both positive and negative, to get workers to quit.

However, "Most of these programs are blind to basic human psychology," Halpern said. "Many are structured in a way that employees that stop smoking are rewarded by having less money taken out of their paychecks for insurance premiums for the next year. But by bundling the rewards into paychecks, they're fairly invisible to people. And the fact that they occur in the future makes them less influential than if people were handed the same amount of money more quickly."

To gauge which incentive programs might be most effective, all the volunteers were offered access to stop-smoking guides and other smoking cessation resources. Those getting health benefits from CVS could also get nicotine-replacement therapy and a behavior modification program.

One group stayed with that regimen. Others formed four groups, in which some individuals were randomly assigned to a “clan” of five other people, and how much they earned depended on the performance of their whole clan. Some were asked to pay $150 to participate, which they would get back later with a $700 bonus if they remained abstinent, while others did not have to pay up front.

Only 12.6 percent of individuals and 15.1 percent of clan members signed up for the up-front payments. In contrast, participation rates were 85 percent to 95 percent among volunteers who were just promised money if they were successful and didn't have to make a deposit.

However, their failure rate was much higher. Only 16 percent remained smoke-free for six months while the quit rate in the up-front payment group was 52 percent. The rate was 6 percent in the usual-care group.

Adjusting for who accepted and who did not, "The deposit programs were twice as effective as the reward programs and five times more effective than providing free smoking cessation aids like nicotine replacement therapy," Halpern said.

Edelman told Reuters Health by phone that it was "discouraging but not unanticipated” that the successful quit rate was cut in half after a year.

Ultimately, "my sense is that employers and insurers could do a whole lot more to curb smoking and reduce costs to themselves by designing their programs in a way that accounts for human psychology," Halpern said. "The fact that people actively seek to minimize loss is one of several psychological insights that can help supercharge incentive programs without increasing their costs."

Based on the results, "CVS Health is rolling out a campaign called '700 Good Reasons,'" Halpern said. "Instead of requiring a $150 deposit, it will require a $50 up-front deposit. If people are abstinent at 6 and 12 months, they'll not only get their $50 back but get an additional $700. Because they'll still have some skin in the game, it should be fairly effective."