Chapter 6

A Lethal Diet

They are the "super-gainers," and becoming one is many smokers' worst nightmare. When most smokers quit, they gain an average of 8 to 10 lb—which is usually unpleasant, but manageable. When super-gainers quit, they pack on 20 or 30 lb. Some of them carry it around the rest of their lives. Even beyond the health risk this extra weight might pose to them, it stands as a warning to other smokers: Beware all who enter here. You, too, could "balloon out."

Exactly why the super-gainers add so much weight when they quit is still being explored. What is known is that this excessive weight gain appears to be genetically driven and appears to be the exception rather than the rule. Different studies' estimates of an average smoker's postcessation weight gain vary by a few pounds, but the reports all fall within the range of approximately 6 to 12 lb.

David Williamson of the U.S. Centers for Disease Control and colleagues from the CDC and the National Center for Health Statistics first described the super-gainer phenomenon in 1991 in The New England Journal of Medicine. Analyzing longitudinal data from more than 2,600 men and women, the investigators found that the average weight gain that could be attributed to smoking cessation was about 6.2 lb in men and about 8.4 lb in women. However, a weight gain of more than 28.6 lb occurred in nearly 10 percent of the men and 13 percent of the women who quit smoking.

Those at the highest risk for excessive weight gain were African-Americans under the age of 55 and smokers who consumed 15 or more cigarettes per day. Among women, being underweight and having a sedentary lifestyle were related to major weight gain. The finding that African-Americans who quit smoking tended to gain more weight than persons of other races was unexpected, since race otherwise has little effect on weight gain among nonsmokers, former smokers, or continuing smokers.

Williamson and his colleagues also found that the average body weight of smokers who quit did not exceed that of individuals who had never smoked. Rather, quitting smoking merely brought that group up to the average weight of nonsmokers. In other words, because of nicotine's effects, a smoker weighs several pounds less than if he or she did not smoke. When the smoker quits using nicotine, body weight returns to what would be normal for that person.

The researchers emphasized that the average person who quits smoking will gain less than 10 pounds, and that about half of the quitters will gain even less. "The beneficial effects of quitting smoking are not likely to be negated by the weight gain that may follow," they wrote.

In a companion commentary to the Williamson article, researcher Neil E. Grunberg of the Uniformed Services University of the Health Sciences raised a possibility that he and others were exploring: "Some investigators believe that the reason nicotine and other drugs of abuse are addictive is that these drugs affect mechanisms controlling body weight and appetite and thereby come to be interpreted as food." Grunberg added that postcessation weight gain "does not mean that potential weight gain is a sound excuse to continue smoking, because the health benefits of smoking cessation exceed the risks of weight gain associated with quitting."

Exactly what causes this relatively uncommon response to quitting smoking remains unknown. Researchers Gary Swan and Dorit Carmelli of SRI International determined that the phenomenon may be influenced by underlying genetic factors. By studying smoking and weight changes in more than 2,100 men in the National Academy of Sciences-National Research Council Twin Registry, in 1995 they confirmed the existence of the super-gainers previously identified by the Williamson group.

They examined the likelihood of being a super-gainer in twin pairs and found that identical (monozygotic) twins were more likely than nonidentical twins to both be super-gainers. However, they also noted several additional relevant factors, including marital status and alcohol consumption. Super-gainers were more likely to be unmarried, and thus may have experienced less outside influence in managing caloric intake and being weight conscious. Also, super-gainers' liquor consumption increased more than twice as much as did that of the average ex-smoker.

"The large increase in consumption in super-gainers suggests the possibility of an especially strong compensatory mechanism," Swan and Carmelli wrote, "with alcohol playing a central role in replacing the effects of nicotine at either the neural-genetic or the behavioral level."

The researchers also commented on the existence of a subgroup of former smokers who reported losing—yes, losing—weight after quitting smoking. These smokers were older when they quit, they had a higher body mass index (proportion of body fat) than other smokers, and they were more likely to be retired and to have cardiovascular disease. Swan and Carmelli proposed that this group's weight loss could have resulted from aging, from disease, or from efforts to reduce the progression of disease by losing weight.

The Weight-Control Smoker

Weight gain after quitting apparently results both from metabolic adjustments the body makes and from increased consumption of food and alcohol after quitting smoking. Even if smokers do not take in more calories, they probably still will gain some weight. Using nicotine replacement during smoking cessation delays the weight gain, although the eventual gain will be about the same as if the smoker had never used the alternative source of nicotine. Even so, this delay may provide some benefit, since it allows the smoker to deal with one problem at a time. First, he or she can focus on quitting smoking and staying quit. Then, some weeks or months later, the ex-smoker may be better able to manage the weight gain that might accompany tapering off the nicotine replacement.

Using tobacco as a weight-control mechanism is not a universal motivation, but it is a common one. Nicotine suppresses appetite and alters metabolism in such a way that someone taking up smoking is likely to drop some weight. Smoking, particularly at a high level, also is believed to delay the rate at which solid foods empty from the stomach, an effect identified by Ellen Gritz and others working with researcher Murray Jarvik at the University of California at Los Angeles. Many smokers, primarily women, use nicotine as a way to control weight, as researcher Cynthia S. Pomerleau and colleagues of the University of Michigan found when they profiled the "female weight-control smoker."

Several scientific reports have shown that although weight-control smoking is rare in male smokers, it occurs in as many as 40 percent of female smokers. Identifying smokers with this profile highlighted the cultural impact on concerns about weight, since (in Pomerleau's words), "[A]fter all, no rat would be likely to self-administer nicotine to maintain a svelte profile, nor would individuals living in a culture that valued plumpness (e.g., as a symbol of prosperity) be likely to respond to this property of nicotine."

According to Pomerleau, it appeared that the tendency of many girls and women to smoke as a means of controlling weight had several bases. First, slenderness is valued throughout much of Western society. Second, women appear to be more sensitive than men to the effects of nicotine on their food intake and their weight. Additionally, women have a higher expectation that nicotine will help them control their appetite and weight.

"A Moment on the Lips . . ."

The researchers examined whether there is a subgroup of smokers, primarily women, who would be likely to be "weight-control smokers." With items embedded in a questionnaire, they identified women who endorsed statements indicating that they smoked to keep from gaining weight or to control their appetite. Women who scored high on the weight-related smoking questions were compared with women who did not report that smoking helped control their appetite or that they smoked to control their weight. Women who were smoking to control their weight reported that when they had been abstinent from tobacco in previous quit attempts, their appetite and weight had changed.

As Pomerleau was quick to note, the ways in which the weight-control smokers and their counterparts did not differ were as interesting as the ways in which they did. The investigators found no evidence that the weight-control smokers were more depressed or anxious. The weight-control smokers were no more nicotine dependent than were those who did not smoke to control weight. However, they had a greater tendency to use nicotine to trigger changes in their internal state, and perhaps an enhanced tendency for such changes to occur.

The researchers also measured withdrawal symptoms in a subsample of the same women. The team wondered whether the weight-control smokers would experience abstinence from tobacco differently than other women smokers would. Of the nine withdrawal symptoms measured in the groups of women over two days of abstinence, only "increased eating" was greater in the weight-control smokers. The researchers inferred that the weight-control smokers might have patterns of excessive or unpredictable eating, which nicotine might help them control.

Pomerleau and her colleagues recommended several possible changes in smoking cessation strategies, based on their findings and those of other scientists. They proposed that researchers should consider studying the following approaches to cessation with these smokers:

• Targeting weight-control smokers so that more attention can be given to their needs in cessation programs.

• Prescription of nonnicotine medications that can prevent weight gain, either alone or with nicotine replacement that is slowly discontinued through tapering.

• Therapeutic behavioral techniques that help in dealing with "disinhibited" or binge eating. These could be taught in a formal stop-smoking program.

Gains and Losses

Scientists also have considered whether concern about weight gain would affect the likelihood of success in smoking cessation. Andrew W. Meyers and his associates at the University of Memphis and the University of Alabama at Birmingham found that persons who used formal smoking cessation programs as a way to quit were less concerned about weight than the average smoker. Also, those who were concerned about weight and who sought help to quit smoking were less likely to quit smoking. These results confirmed findings by Meyers' colleague and co-author Robert C. Klesges of the University of Miami, who previously had found that many smokers were concerned about weight gain, but those who were the most concerned had the least intent to quit smoking. Klesges and his co-investigators found concern about weight gain in both men and women smokers, and observed that those who anticipated the gain were more likely to relapse.

Although the connection between tobacco use and weight control has been recognized for at least a century, it is only in the last decade or so that researchers have been able to study the association in detail. Some common assumptions have not yet been demonstrated, including these: (1) Smokers keep smoking to prevent a weight gain that might result from cessation. (2) Weight gain is a cause of smoking relapse. (3) Controlling weight gain during cessation prevents relapse.

Actually, the opposite of those notions may reflect the reality of smoking cessation. For example, stop-smoking interventions also designed to help prevent weight gain have even been found to heighten the risk of relapse. The current thinking is that as long as weight gain does not cause relapse, weight gain can be managed more easily if it is handled after smoking cessation is no longer the most pressing issue.

The most recent findings indicate that smoking cessation results in weight gain primarily through increased caloric intake after quitting smoking. During the first month after a smoker quits, food intake increases by some 300 to 400 calories per day, with much of this increase due to snacks between meals. This increase accounts for the approximate pound per week that the average ex-smoker gains after quitting. These effects may be particularly noticeable for women, who tend to obtain more postcessation calories from snacks than men do, and who have been found to increase food intake more than men do after quitting. Women smokers who are high in dietary restraint (in other words, women who have chronic concerns about weight and who diet to maintain an unreasonably low body weight) may use smoking as a way to suppress eating; for them, weight control is perceived as a benefit they derive from smoking.

The phenomenon of using smoking to suppress weight emerged among both men and women smokers in a 1996 study of smokers in Austria. Éva Rásky of the Institute of Social Medicine, Karl Franzens Universität at Graz, Austria, and co-authors found in a study of more than 27,000 rural Austrians that light or moderate smoking was correlated with a lower weight (relative to height and build), while heavy smoking and quitting smoking were related to higher weight.

In U.S. adults, smoking cessation does contribute to the prevalence of overweight, although the impact is small. Katherine M. Flegal and colleagues of the U.S. CDC reported in 1995 that weight gain associated with quitting smoking accounted for about one-fourth of the weight increase noted in men and about one-sixth of the increase seen in women between 1978 and 1990 in the United States. Taking into consideration other factors such as age, demographics, physical activity, alcohol use, and childbearing, they found that men in their national survey gained an average of 9.8 lb, and women gained an average of 11 lb. Smokers who had quit during the previous decade were more likely to become overweight than were nonsmokers.

Since weight control is important to many smokers, offering weight- control help in conjunction with a cessation program seems appropriate. However, an extensive clinical trial by Sharon Hall and her colleagues at the San Francisco Veterans Affairs Medical Center and the University of California at San Francisco, reported in 1992 that the opposite may be the case. The research team offered three types of adjunct treatment in conjunction with stop-smoking treatment: (1) a behavioral weight-control program that included exercise, weight monitoring, and calorie control; (2) a nonspecific weight-control group involving group therapy, with supportive help and information on nutrition and exercise; and (3) information packets on nutrition and exercise. Fewer people remained abstinent in the two weight-control groups than in the third group at three months, one year, and longer than one year.

A second reason for weight gain following cessation is the change in metabolic rate that accompanies the use of each cigarette. Nicotine does increase the metabolic rate (or resting energy expenditure) of tobacco users. Janet Audrain and colleagues reported in 1995 that although nicotine heightened the metabolism of both normal-weight and overweight women, nicotine's effects were attenuated in those who were overweight. Thus, overweight women might not be obtaining as much weight-control "benefit" from nicotine as they believe they are receiving.

This report concurred with findings from a research team at UCSF, who noted that although smokers weigh less than average, smokers who smoke more weigh more than those who smoke less. The investigators, Lidia Arcavi and colleagues, found that nicotine increased heart rate and energy utilization in most smokers, but that these effects were most pronounced in smokers who used 10 or fewer cigarettes a day. These low-level smokers also used more energy as a result of their nicotine use than did high-level smokers consuming 15 to 30 cigarettes per day.

Curiously, the researchers also found that the phenomenon of tolerance was different among low- and high-level smokers. (As explained in chapter 3, developing tolerance for a substance involves needing increasingly larger quantities of it to achieve the same drug-related effects.) The low-level smokers developed tolerance as indicated by acceleration in heart rate and in energy expenditure. High-level smokers, however, developed only tolerance indicated by heart rate. The tolerance to nicotine seen in their cardiovascular response was not matched by changes in energy expenditure; either they had a rapid development of tolerance or they showed no effect at all. The authors concluded that these differences between groups of smokers could help explain what they termed the "unusual" relationship between nicotine and body weight.

Influences on Women

Overall, nicotine dependence may be a somewhat different experience for women than for men. The appetite-suppressant capacity of nicotine may increase its appeal to women and thus increase women's possibility for developing dependence. Additionally, significant evidence indicates that individuals prone to depression are also more likely to use nicotine. Since depression is more common in women than in men, this becomes an issue of particular concern for women smokers. Many aspects of the nicotine-weight relationship seem particularly problematic for women.

A 1994 article by John Pierce and associates at the University of California at San Diego Cancer Center explored trends in smoking initiation among children and adolescents, focusing in particular on the effects of targeted advertising. In their report, which was part of an issue of The Journal of the American Medical Association devoted primarily to tobacco research, they noted that the increase in smoking prevalence among girls younger than the legal age for buying tobacco "started the same year that the tobacco industry introduced women's brands of cigarettes." That year was 1967. The researchers' examination of the temporal (time-related) correspondence between advertising campaigns and smoking initiation among women showed that the increase was notably higher in women who had never attended college. Across the period under study, 1944 through 1988, the rates at which adolescent boys started smoking changed little. The authors concluded that the increase in smoking among young women was associated with increased advertising targeting women.

Another risk factor enhanced in women smokers is the prevalence of depression. Women are more prone than men to experience depression; they also are at heightened risk of becoming nicotine dependent. This does not mean that women are biologically predisposed to becoming more physically dependent on nicotine, or that dependency is more a basic part of their nature. Rather, it means that some of the predisposing factors are more common to women.

Despite women's increased risk for problems that might contribute to dependence, smoking rates remain higher among men than among women almost universally throughout the world. An exception is the Lahanan people of central Borneo, a group of about 300 who live in a horticulture-based economy that has given them only limited contact with the world outside their community. In this setting, the Lahanan women traditionally have been the primary cultivators and disseminators of tobacco, and they also have been the heaviest users.

Hand Me the Ashtray, Dear

This reversal of the usual sex-related tobacco-use pattern is changing, however, as the group has more contact with the industrialized world. Smoking among these women is becoming less prevalent, and men are taking up cigarette smoking rather than using the home-grown tobacco. Young persons who acquire an education tend to quit using tobacco, according to researchers who have studied this group, Jennifer and Paul Alexander of the University of Sydney in New South Wales, Australia.

Aside from economic and cultural considerations, women respond somewhat differently to nicotine than men do. Ken Perkins and colleagues of the University of Pittsburgh found in 1994 that women and men smokers reported nicotine's "subjective" effects differently, although both men and women were able to tell the difference between a dose of nicotine and a dose of non-nicotine placebo. Women reported dose-related nicotine effects in feelings described as "dizzy," "stimulated," "jittery," and "head rush," in response to increased doses of nicotine, but men did not. Nicotine was administered in relation to each subject's weight, via an inhaler device. The researchers used substances to mask the taste and smell of nicotine and the irritation of nasally administered nicotine.

Sex-based differences in nicotine's effects also can be related to the situations in which nicotine is used. Perkins noted in a 1996 review article that women may be more sensitive than men to situational factors such as the sight and taste of cigarette smoke. Research has shown that men and women may smoke for different reasons. Men apparently smoke to maintain a steady level of nicotine in their body, but women might be smoking to obtain effects that are less related to nicotine.

Tobacco is known to contain thousands of chemical compounds and is associated with many complex factors that can elicit their own responses in addition to those elicited by nicotine. In other work, Perkins' team noted that subjects responded differently to smoked tobacco and nasally inhaled nicotine spray, which suggested that smoking generates effects in addition to the delivery of nicotine. Even though most of the responses appear to be due to the delivery of nicotine, other aspects of smoked tobacco cannot be discounted or ignored.

Such gender-related differences in what makes smoking reinforcing could influence the effectiveness of a stop-smoking program in helping women quit. It is possible that women might not find nicotine replacement (such as gum or patch) as useful as men would find it, and that cessation geared generically toward both sexes could overlook women's particular needs in smoking cessation.

Males and females also appear to differ in the way nicotine affects the body's energy balance. It is possible that smoking helps some women suppress appetite and eating by adding the reinforcing effect of an increase in energy. Nicotine increases the body's expenditure of energy, or its metabolism. When male smokers engage in light physical activity, metabolism is enhanced. Women smokers, however, experience no metabolic enhancement from light activity. This gender difference, as explained by Perkins in a 1997 chapter, existed only during physical activity and was not present when subjects were at rest.

Each time a smoker used tobacco, the nicotine prompted a brief boost in metabolism of 5 to 7 percent for about a half hour. A pack-a-day smoker was likely to boost at-rest metabolism throughout the day, even though the effects were brief, because he or she used nicotine so continuously. A smoker using caffeine boosted metabolism even more, since the effects of the two substances (caffeine and nicotine) were additive. Men who engaged in light activity while using caffeine and nicotine together experienced an enhanced metabolic rate. Women did not.

Who is most likely to have a metabolic boost from smoking? The answer: men in good physical condition who typically engage in high levels of activity. Women, who are most likely to use nicotine to lose or maintain weight, are less likely to experience a metabolic benefit from using tobacco.

The use of nicotine replacement in cessation can forestall weight gain. Also, exercise and dietary changes can influence success in quitting smoking and can help a smoker avoid gaining an undue amount of weight. Even so, quitting smoking and dieting at the same time might not be the best approach. Several studies have found that combining smoking cessation with weight-control methods can worsen the abstinence rate, canceling out any presumed advantage of tackling both issues at once. Dieting can result in some of the same noxious effects as smoking abstinence, such as mood disruption, diminished arousal, and fatigue.

Another male-female difference that can affect smoking rates and success in quitting is the effect of nicotine in reducing feelings of stress. Women report a greater tendency than men to smoke as a way to reduce negative emotions. This could also make women more prone to relapse during stressful times. When a woman is attempting to quit smoking, she may find that smoking relieves her feelings of being stressed. It is possible, however, that the stressful feelings are due primarily to abstinence symptoms (or withdrawal), and thus the stress relief that tobacco provides during cessation could be due to nothing more than relief from those symptoms.

All of these factors could be part of the overall statistical picture reflecting these facts:

• Smoking has not declined as much in women as in men.

• Young women are taking up smoking at a greater rate than are young men.

• Quitting may be more difficult for women than for men.

• Women may be more likely to relapse.

All Is Not Lost

These findings about weight and smoking, particularly as they affect women smokers, can be discouraging. Even so, smoking—present, past, or passive—is only one of many factors affecting weight. Swiss researchers Martine Bernstein and colleagues of the University Canton Hospital at Geneva, Switzerland, examined the relationship between education, smoking status, and weight in 928 Swiss women. They found that a woman's education level was an important predictor of her current weight and her weight history (i.e., previous weight and weight gain since age 20). Smoking status appeared to have little effect on weight. This finding is ironic in light of findings that women are more likely than men to report using tobacco as a way to control weight.

Bernstein and her colleagues found that the group of women (with ages ranging from 29 through 74) who had the most education weighed the least. The least educated group weighed an average of 9 lb more than the most educated group. Differences in relation to tobacco exposure status were small. Women exposed to secondhand smoke, whom the researchers termed "passive smokers," weighed the most at 140 lb, and former smokers weighed the least at 133 lb. Weight differences associated with education status were greater.

The research team concluded that although smoking "may influence short-term weight variation," it had little long-term effect on weight. They explained: "The finding that smoking is not an efficient means of weight control has major implications with respect to public health strategies aimed at reducing smoking. Women will be less reluctant to quit smoking if they know it will not promote a long-term weight gain."

Several strategies could be helpful for a smoker concerned about weight and also contemplating quitting. Perkins and his colleagues in 1997 proposed what they termed a "cognitive-behavioral" approach to cessation that would treat not the weight gain itself, but the smoker's concerns about weight gain. As the term itself implies, cognitive-behavioral refers to a treatment that helps people change both their thinking and their behavior. In this case, a goal would be for participants to change their beliefs about the primacy of not gaining weight, and clinicians would work with them to change the behaviors associated with smoking, to facilitate quitting.

Perkins explained that concern about the typical weight gain of 8 to 10 pounds "must be viewed as dysfunctional and unreasonable in light of the health risks of continuing to smoke." A cognitive-behavioral approach would assume that the attitudes and perceptions of the importance of weight need to be modified, not the tendency to gain weight. "Weight concerns may stem from unreasonable perceptions about the weight gain and, more importantly, from the heightened importance placed on a modest weight gain in controlling one's health behaviors (i.e., whether or not to quit smoking)," Perkins added.

Other options include pharmacological (that is, drug) treatments. In this regard, smokers have several options. First, smokers who quit with the help of nicotine replacement are likely to find that substituting one source of nicotine for another delays the weight gain often associated with quitting. Once they taper off the nicotine, they may find that they then gain some weight. However, by this time they may be better prepared to deal with the weight gain than if they had attempted to quit smoking and severely restricted their diet at the same time.

Studies also have examined the usefulness of a drug called phenylpropanolamine, which is sold without prescription in the United States both as a decongestant and as an appetite suppressant. To date, phenylpropanolamine has been examined only in short-term use, so its effects on long-term abstinence and long-term weight gain are undetermined. Typically, drugs that suppress appetite do not continue to prevent weight gain once their use is discontinued.

A relatively simple but effective strategy involves changes in levels of exercise. A moderate increase in physical activity level minimizes postcessation weight gain without requiring extreme restrictions on food intake. Several studies indicate that it is not necessary to undergo strenuous exercise to achieve enough metabolic enhancement to keep postcessation weight gain within manageable levels. The addition of such common activities as walking instead of driving short distances, or climbing a flight or two of stairs instead of taking an elevator, can help keep the weight gain to levels below the usual Thanksgiving/Christmas increase most celebrants experience every holiday season.

The worst strategy, it appears, is to attempt to control postcessation weight gain through undue restrictions on food intake. One of the more interesting findings leading researchers to this conclusion was the discovery by Hall and her colleagues in 1986 that smokers who gained the least weight were the most likely to relapse back to smoking.

Evidently, the relationship between smoking and body weight is not simple; what is simple, however, is the fact that the benefits of quitting smoking outweigh the disadvantages of the weight gain that might accompany quitting.