Financial Rewards for Quitting Smoking Work Well

by on May 18, 2015 at 11:23 am in Economics, Medicine | Permalink

A large study in the New England Journal of Medicine verifies that financial rewards for quitting smoking are effective. Participants were randomly offered one of a variety of incentive schemes that paid participants who successfully quit smoking (verified with saliva and urine tests). Participants were free to decline the offer.

The most interesting variation of the study was to compare a carrot model which paid up to $800 for success with a carrot-stick model in which participants lost $150 if they failed to stop smoking but gained $800 if they succeeded (i.e. $650 of reward plus refund of $150). In theory, the carrot-stick model should work better because it harnesses loss-aversion. And statistical analysis suggested that for those who would accept either the carrot or the carrot-stick model, the carrot-stick model did work better. The problem is that far fewer people who were offered the carrot-stick model chose it compared to those offered the carrot model. Overall, therefore, the carrot model was far more successful.

Smokers are costly so even a pure carrot model of $800 paid by employers would more than pay for itself:

…Finally, the finding that individual rewards of $800, as compared with usual care, nearly tripled the rate of smoking cessation among CVS Caremark employees and their friends and family confirms and extends the generalizability of our finding from a previous trial involving General Electric employees. In addition to the public health effects of such smoking reductions, these findings are important for employers. Because employing a smoker is estimated to cost $5,816 more each year than employing a nonsmoker, even an $800 payment borne entirely by employers and paid only to those who quit would be highly cost-saving.

1 AndrewL May 18, 2015 at 11:31 am

Alright 800! I could by 40 packs of cigarettes with that kind of money!

2 cheesetrader May 18, 2015 at 1:13 pm

$20/pack? Where do you live? “Only” $7 here in flyover country

3 AndrewL May 18, 2015 at 1:21 pm

Okay I was exaggerating, in NYC, cigarettes go for about 15$ a pack. Still, It seems like a lot of money to me (I am a non-smoker).

4 AndrewL May 18, 2015 at 1:24 pm

But, the point is, if you live in NYC, you could reap the reward in 2 months by simply not buying cigs. (yes, I understand, not so simple)

5 prior_approval May 18, 2015 at 11:43 am

Imagine just how well the incentives to not start smoking at all would have worked – it is really easy to never start, after all. Well, apart from being a nightmare of the government getting involved in the personal choices made by those surrounded by overt and ominpresent tobacco industry marketing – anyone else remember the days that it only took ripping out a full color glossy insert from many mass market paperback to get a free carton?

6 Axa May 18, 2015 at 12:08 pm

Nowadays you exchange your contact data for a pack of cigarettes. Marketing people from some brand made me a 3 survey question in a bar: do you smoke?, which type do you smoke? would you like to try the new flavor? 2 weeks later I got 2 free packages by mail at home. It was funny, along the cigarettes I also got a bag of seeds to plant flowers…….tobacco companies are now green 😉

7 MOFO. May 18, 2015 at 7:56 pm

Here here! Overt and everpresent marketing is what got me to buy my Edsel and started me in on 5 drink a day Zima and Crystal Pepsi habit.

8 Alex Godofsky May 18, 2015 at 1:11 pm

Smokers are costly so even a pure carrot model of $800 paid by employers would more than pay for itself:

You mean until people claim that they, too are smokers just so they can get into the free money?

9 Dan Weber May 18, 2015 at 2:33 pm

While it’s good to worry about cheaters, 1) they can do a chemical test, and 2) $800 is not very much money to get someone to smoke.

You’d probably have to smoke for 3 months or so to build up to the right level.

10 Blaise May 18, 2015 at 2:37 pm

Step 1: Start smoking.
Step 2: Enroll in the program that promises $800 if you stop smoking.
Step 3: Stop smoking and pocket the money.

11 Dan Weber May 18, 2015 at 2:59 pm

Now that you’ve got me thinking, nicotine patches could work to get my blood levels up, and it’s probably easier to quit wearing patches than to quit smoking. And probably avoids most (all?) of the health side effects.

12 cheesetrader May 18, 2015 at 1:16 pm

Seems to me that CVS is ripping off the smokers – they should be offering $5,815…

But then as Alex points out, you’d have a bunch of people claiming to be smokers just to get the money

Curious – how much does obesity cost and how much of the cost is simply insurance companies jacking up the rates rather than actual cost?

Oh wait – is it politically correct to charge fat people extra? I like the idea of health care per pound

13 Dan Weber May 18, 2015 at 2:33 pm

It’s hard to find something that makes a third-party quit being overweight.

14 Mike Hunter May 18, 2015 at 4:51 pm

I thought they already charge people who are over weight more for healthcare. I’m 20 lbs overweight according to height/weight tables and I’m casually trying to lose it. I’d certainly be a lot more motivated if my employer, or even a third party offered an incentive system.

A 3rd party could even set up a self funding carrot/stick incentive system if people’s weight could be independently verified. Kind of like gambling on the outcome of a sporting event I suppose.

15 Moelicious May 18, 2015 at 2:04 pm

Nicotine is a powerful anti-depressant. Will the insurance companies save that much money if the smokers switch to anti-psychotics (which can cost up to $450 per bottle) and therapy? I know a guy who smokes e-cigs because it’s much cheaper than a shrink plus happy pills.

16 Doug May 18, 2015 at 4:08 pm

ADHD meds too. Nicotine is a highly effective focus aid.

E-cigs are the ultimate solution. Nicotine itself is no more harmful than caffeine. All the health impact of smoking comes from the fact that inhaling smoldering organic compounds.

17 Mike Hunter May 18, 2015 at 4:46 pm

Is that true though? I thought that they were still studying e-cigs, and no definitive evidence has yet been found.

18 carlospln May 18, 2015 at 5:00 pm

‘Nicotine is no more harmful than caffeine’

Read much?

http://theoncologist.alphamedpress.org/content/9/3/353.full

19 Cooper May 18, 2015 at 8:02 pm

http://www.cancer.org/cancer/news/why-lung-cancer-strikes-nonsmokers

According to the American cancer society, a male smoker has 23X the risk of developing lung cancer than a male non-smoker. For women the risk is 13X greater.

Also, many of the non-smokers who develop lung cancer do as a result of second hand smoke. These are people who live in houses with smokers, ride in cars with smokers, etc.

If we lived in a world where nobody smoked cigarettes and every single person smoked e-cigarettes, I’m inclined to believe that lung cancer rates would still fall off a cliff.

20 carlospln May 19, 2015 at 2:46 am

“If we lived in a world where nobody smoked cigarettes and every single person smoked e-cigarettes, I’m inclined to believe that lung cancer rates would still fall off a cliff”

That’s great-we’d all be better off in your ‘best of all possible’ worlds.

Unfortunately, you have no data & there is no counterfactual: pure mood affiliation.

The End.

ps don’t shove your tobacco habit down my throat

21 Jim Jones May 19, 2015 at 3:54 am
22 andy May 19, 2015 at 7:00 am

“Also, many of the non-smokers who develop lung cancer do as a result of second hand smoke. These are people who live in houses with smokers, ride in cars with smokers, etc.”

And some numbers would be welcome, unless you want this claim to be misinterprented. In order to develop a lung-cancer with second-hand smoke you would have to spend with the smoker literally most of the day and even then the risk is very low. I would say it is quite unlikely that these days there are many such people as most work is non-smoking.

23 carlospln May 19, 2015 at 7:39 am

@ jim jones @ 3:54 am

jim, the what do you think the reason is for civil aviation to have universally banned tobacco smoking?

??

24 gwern May 19, 2015 at 11:51 am

Carlos: read your own link and its citations! Besides being extremely shallow and short, it confounds nicotine and tobacco smoke and offers no evidence that former-smokers have similar risks as current smokers, much less showing that never-smokers who use nicotine (*not* tobacco products) have elevated risks.

25 carlospln May 19, 2015 at 4:34 pm

G

The post cited showed in exacting detail the mechanism by which nicotine can damage DNA, the precursor to mutations, Ca…..& worse.

It did NOT show the other mechanisms by which PAH [poly aromatic hydrocarbons] can intercalate between base pairs to do the same.

If you’re a smoker, knock yourself out.

But shitcan your tendentious nit picking.

The only person you’re fooling is yourself.

26 ChacoKevy May 18, 2015 at 3:31 pm

Another fishy thing about this study is that the study population comes from CVS employees. Remember, just last year in October, CVS stopped selling cigarettes in their stores. 6 months plus a little bit of time later to write a paper and they have a study saying their financial incentive program was successful but no mention that there was a huge restriction to cigarette access. I’m not sure this one is going to make waves.

27 carlospln May 18, 2015 at 5:06 pm

For Christ’s sake, how much data, how many suss ‘conclusions’ from how many schite experimental designs do we have to wade through from Tabarrok? http://nautil.us/issue/24/error/the-trouble-with-scientists

28 Jimbino May 18, 2015 at 4:40 pm

So I take it that an independent contractor or consultant can expect cash remuneration for the $5000 smoker premium, the 6.60% SS premium, the 2% unemployment insurance premium and the 3% workers compensation premium? That’s a good reason never to take a captive employee job!

29 Mike May 18, 2015 at 5:32 pm

Alex, the carrot-stick treatment doesn’t necessarily leverage loss-aversion, at least not at the point when the abstinence from smoking decisions are made, though it does introduce the potential for a changed reference point. By the time people face the choice of abstinence from smoking, they have already given up the $150. While it isn’t a sunk cost, it does alter the status quo, and that may affect the marginal utility of the $650 reward. No such status quo shift happens in the carrot only treatment. I’d wager that most of the treatment effect is due to selection bias caused by avoiding the treatment with potential loss.

30 Clyde Schechter May 18, 2015 at 5:40 pm

“Smokers are costly so even a pure carrot model of $800 paid by employers would more than pay for itself. ”

Not so fast. From a societal perspective that is true. But from the employer’s perspective, not so much. If I’m already in my 60’s and have been smoking for 40+ years, my employer will gain relatively little by getting me to quit now. A big part of the payoff is getting the younger smokers to quit, before they have smoked enough to establish a slow-to-improve risk of lung cancer and obstructive pulmonary disease. (Another part, equally effective at all ages, is the reduction in risk of coronary heart disease, which sets in after only about a year of tobacco abstinence.)

But if I’m in my 20’s and my employer pays me $800 to quit smoking, the savings will likely accrue to whatever competitor of his/hers (possibly myself, if I’m entrepreneurially inclined) I’m working for decades later.

31 derek May 18, 2015 at 6:27 pm

All these initiatives seem to hit a hard wall of around 20% or so of the population, probably the proportion who smoke to self medicate.

32 robert May 18, 2015 at 6:59 pm

You want incentives to stop smoking? Check out Stephen King’s way of dealing with the problem: http://en.wikipedia.org/wiki/Quitters,_Inc..

33 cheesetrader May 19, 2015 at 8:47 am

That was an excellent story

34 andy May 19, 2015 at 7:02 am

Because employing a smoker is estimated to cost $5,816 more each year than employing a nonsmoker, even an $800 payment borne entirely by employers and paid only to those who quit would be highly cost-saving.

Cost whom? Why don’t employers pay $5.816 less smokers a year than non-smokers? If they do, it costs the smokers. If they don’t, then the claim is very dubious, unless there is some government regulation forcing employers to equalize the pay.

35 Daniel Reeves May 19, 2015 at 3:11 pm

This is so much better than all previous studies and statistics I’ve seen on the efficacy of commitment devices! I’ve been frustrated by Beeminder competitors who tout statistics about how 80% or whatever of people who risk money succeed. For starters they usually don’t even distinguish from the hypothesis of “people will lie to keep from losing their money!” — which this study seems like it’s robust to, with the cotinine tests. But more to the point, other studies I know of haven’t accounted for the selection effect (only super serious people being willing to risk money). So the intent-to-treat methodology here is great and it seems encouraging that the effect is still significant even with 86% of people in the commitment contract group refusing to pony up.

Here are the key numbers for smoking cessation rates:

* 6% quit in the control group with standard treatment like nicotine replacement
* 16% quit in the pure $800 reward group
* 10% quit in the commitment contract group risking $150 + $650 reward
* 50%+ quit in the subset of the commitment contract group (14% of
them) who actually participated

I still need to dig deeper on the claim that “statistical analysis suggested that for those who would accept either the carrot or the carrot-stick model, the carrot-stick model did work better”. I’m highly biased to believe that but even if it’s not true, we still have the result that, for cases where you don’t have a third party funding rewards for you, you can always find a third party to collect penalties for you!

Comments on this entry are closed.

Previous post:

Next post: