Researchers Set Out to Gauge Stop-Smoking Financial Incentives; End Up Proving Smokers Can’t Quit

The New England Journal of Medicine just published a smoking cessation study that measured how financial rewards and penalties affect quit rates.  While the study generated moderate media interest (examples here and here), the big story was buried deep in a supplemental table: the quit rate was always abysmal.

Investigators from the University of Pennsylvania, led by Dr. Scott Halperin, recruited 4,017 CVS Caremark employees, relatives and friends who smoke and randomized them to one of five interventions.  The control group received cessation pamphlets from the American Cancer Society; CVS employees were also provided with free behavior therapy and pharmaceutical nicotine.  In addition to pamphlets, the other four groups had one of four interventions: they received up to $800 rewards for quitting as (1) individuals or (2) in small-group settings, or they paid deposits of $150 that were refunded and coupled with rewards upon quitting, either as (3) individuals or (4) in small groups. 

Halperin et al. analyzed quit rates at six months after enrollment, concluding: “Reward-based programs were much more commonly accepted than deposit-based programs, leading to higher rates of sustained abstinence from smoking.  Group-oriented incentive programs were no more effective than individual-oriented programs.”  More importantly, they also reported the results after what is normally considered the gold standard for follow-up -- one year.  Here is a summary of the findings:

Quit-Smoking Rates From the CVS Caremark Study at One Year
Intervention	Number of Participants Who Quit / Total	Success Rate
Usual Care	16 / 468	3.4%
Individual Reward	37 / 498	7.5%
Group Reward	45 / 519	8.7%
Individual Deposit	21 / 582	3.5%
Group Deposit	29 / 471	6.2%
All Participants	148 / 2,538	5.8%

All told, only six of every 100 smokers quit at one year.  This represents about a 50% relapse rate from the six-month results that Halperin and colleagues used for their analyses.  That rate applies only to the 2,538 smokers who did not drop out before the study started.  Considering all 4,017 eligible smokers, the quit rate was 3.7%, the same or even lower than the quit rate in the general population.

Halperin and his colleagues deserve credit for trying to use economic incentives to change smoking cessation rates.  Unfortunately, their study proves once again that most smokers are unwilling or unable to quit.

Smokers who won’t or can’t quit can still benefit from the smoke-free options outlined in For Smokers Only, available for Kindle (here) or as an audiobook (here).

FDA Rejects Reynolds Petition to Change Smokeless Warning

Congress in 1986 legislated that smokeless tobacco products carry a warning label, “This product is not a safe alternative to cigarettes.”  In 2011, Reynolds American petitioned the FDA to replace that deceptive language. As I have noted (here), Reynolds’ request was straightforward: “…the Government should, in suitably brief form, tell the whole truth, not mislead by telling only part of the truth” about the health risks of smokeless tobacco use.

On May 11 the FDA denied Reynolds’ petition.  Center for Tobacco Products Director Mitch Zeller defended the action on the grounds that (1) the current not-safe-alternative warning is factual, (2) there is no evidence that the current warning is responsible for the fact that a vast majority of Americans incorrectly believe smokeless tobacco is just as hazardous or more hazardous than smoking, and (3) there is no evidence that the proposed replacement warning would “promote greater understanding of the risks associated with the use of smokeless tobacco products.”  The "greater understanding" phrase is repeated throughout the FDA’s 12-page decision letter. 

The agency’s rejection rests on distorted logic.  First, as I have pointed out for 20 years, the not-safe-alternative warning sets up a straw man; a perfectly safe alternative to cigarettes – or to any consumer product – doesn’t exist.  Extending the FDA’s logic, every consumer product should carry a warning that it is not “safe.” 

I do agree with the FDA’s assertion that the not-safe warning is not responsible for Americans’ misperceptions about smokeless tobacco; major medical organizations (here, here and here),  federal agencies (here and here) and state governments share the blame.  However, the FDA shifts the burden of proof.  Reynolds clearly demonstrated that the warning is inaccurate and misleading, but the agency wants the company to prove that the warning causes the misperceptions.  In effect, the FDA acknowledges that the warnings accomplish the government’s purpose – to portray all tobacco products as equally dangerous.

Most importantly, the FDA says that there is no evidence that Reynolds’ proposed replacement warning – No tobacco product is safe, but this product presents substantially lower risks to health than cigarettes – would “promote greater understanding of the risks associated with the use of smokeless tobacco products.”  Mr. Zeller then repeats a list of diseases vaguely “known” and “shown” to be linked to smokeless tobacco.  But he doesn’t concede that the risks are so low that the government has never produced or cited specific estimates of deaths from smokeless tobacco use.

The Swedish Match MRTP application, discussed previously in this blog, would delete the cancer, gum disease/tooth loss and not-safe-alternative warnings and add a relative risk warning virtually identical to the one Reynolds proposed.  It remains under FDA review.

Scaremongering Tactics Against Smoke-Free Tobacco Could Also Undermine Nicotine Medicines

Tobacco prohibitionists routinely cherry-pick data to put tobacco products in the worst possible light.  For example, crusaders have used calls to poison control centers to paint smokeless tobacco (here) and, more recently, e-cigarettes (here) as toxic time-bombs.  To underscore how deceptively simple and grossly misleading such scaremongering can be, I offer this demonstration of selective data analysis applied to nicotine medicines.

Opponents of safer smoke-free tobacco promote nicotine medicines as quit-smoking aids that are “safe and effective.”  In previous posts I have shown that nicotine medicines are not effective; their success rates are pitifully low (here and here). The question remains: Are nicotine medicines safe?

Data from the FDA Adverse Event Reporting System provides the starting point for a scaremonger case against nicotine medicines’ safety.  The FAERS is a collection of volunteered reports from consumers and health professionals, plus mandatory reports from manufacturers.  The dataset is difficult to use and interpret, so I worked with Joe Fuisz and Tom O'Connell of Vapor Tobacco Manufacturing LLC and Paul Danese of FDAble.com to prepare this analysis.

We looked at adverse events linked to nicotine medicines from 2007 to 2014, a period consistent with the availability of e-cigarettes.  Adverse events for e-cigarettes were also recorded by the FDA during this timeframe, however, those numbers may be underestimated, as e-cigs are not considered medicines by most consumers and health professionals.

During this period, there were 690 adverse events related to nicotine gum/lozenges, 2,006 related to patches, 1,123 for the spray and 28 for e-cigarettes.   

Table 1 illustrates the top 5 adverse events for each product.  It is not possible to make direct comparisons or to calculate rates, as these are individual reports, but it is noteworthy that “ineffective” was a top 5 adverse event for both gum/lozenges and patches.  The spray is commonly known to produce throat symptoms and cough, and nausea is also a frequent complaint for smokers who switch to smoke-free tobacco products.

Table 1. Top Five Adverse Events (Number) for Nicotine Medicines and E-Cigarettes, 2007-2014
Gum/Lozenges (690)	Patches (2,006)	Nasal Spray (1,123)	E-Cigarettes (28)
Intentional drug misuse (243)	Itching (396)	Cough (96)	Headache (6)
Nicotine dependence (216)	Redness (395)	Product quality issue (71)	Nausea (5)
Administration error (86)	Drug ineffective (368)	Throat irritation (64)	Dizziness (4)
Drug ineffective (65)	Administration error (364)	Nausea (63)	Muscle spasms
Nausea (56)	Nausea (199)	Dizziness (47)	Chest pain (3)

While the most common adverse events were not serious, the list of outcomes in the following table is worrisome at first glance.  There is no representation in the reporting system that the products “caused” the negative outcomes, but those aiming to trash nicotine medicines would make that linkage and issue press releases with hyperbolic safety warnings to drive their point home.  It’s that simple.

Table 2. Number of Serious Outcomes* for Adverse Events Related to Nicotine Medicines and E-Cigarettes, 2007-2014
Outcome	Gum/Lozenges	Patches	Nasal Spray	E-Cigarettes
Congenital Anomaly	0	5	1	0
Disability	4	7	13	1
Hospitalization	17	101	78	7
Life Threatening	3	7	3	6
Death	6	18	8	0

*Numbers are not mutually exclusive

Scaremongering is a despicable, yet all too common practice that undercuts legitimate scientific discourse, distorts public health policymaking and ultimately costs smokers’ lives.  Recreational use of nicotine in smoke-free delivery systems, like recreational use of caffeine, is relatively safe for consumers.  

Memo to the FDA: Release Survey Data

An update on the activities of the FDA Center for Tobacco Products was provided two weeks ago by Director Mitch Zeller at a meeting of the National Association of Tobacco Outlets (NATO).  Zeller repeatedly referred to the Population Assessment of Tobacco and Health (PATH) Study, “…a longitudinal cohort study on tobacco use behavior, attitudes and beliefs, and tobacco-related health outcomes among youth…and adults in the United States.” (description here).

Zeller said that agency staff presented information from the study at a scientific meeting earlier this year, and that additional reports would be forthcoming.  In fact, the FDA and a privileged group of investigators at Roswell Park Cancer Institute, the National Institute of Drug Abuse and Westat, a private consulting firm, are producing information based on a comprehensive dataset that is not available to anyone else.   

During the NATO public Q & A session, I asked Zeller if the FDA had set a date for public release of the PATH dataset.  He answered in the negative.

Here we go again.  For years, I have documented that the CDC routinely withholds the underlying data it uses to calculate the “Big Kill” estimates of smoking related deaths (here and here) which are used to drive tobacco regulations and federal policy.  Additionally, the CDC and American Cancer Society refuse to release data (here) on the minuscule risks of smokeless tobacco use -- data that could support product applications for reduced-risk status (I have described what the Society should do to remedy this here).  These organizations have treated smokeless tobacco users with contempt by refusing to release this information.

This institutional data embargo now affects e-cigarette users.  Every year, the CDC parlays its exclusive access to the National Youth Tobacco Survey to convince Americans that e-cigarettes will enslave and kill their children (here and here).  When the CDC releases the data months or years later, a completely different picture emerges (here).

As for the PATH study, the FDA timeline shows that it plans to “finalize restricted Use Data Set [public release] for baseline data” in the Fall, with simultaneous release of a “complete first manuscript of baseline results.”  In other words, the FDA will start spinning the results before it releases the data for independent analysis.  This is unacceptable; if the dataset is good enough for the FDA to produce a publication, then it is good enough for public perusal.   

For 20 years, biased analyses of restricted federal datasets have driven tobacco litigation, legislation and regulation.  The FDA should make the PATH survey publicly available at the earliest possible date.